title
stringlengths 1
200
⌀ | text
stringlengths 10
100k
| url
stringlengths 32
885
| authors
stringlengths 2
392
| timestamp
stringlengths 19
32
⌀ | tags
stringlengths 6
263
|
|---|---|---|---|---|---|
What Is Inheritance? | Inheritance Types
1- Single Inheritance
Single inheritance enables a derived class to inherit properties and behavior from a single parent class.
UML for Single Inheritance
Let’s see Single Inheritance Code
When we look at the UML table, it is seen that the Dog class does not have the eat() method, but the Dog class has this method because it inherit from the Animal class. When this method is called in Main function, the eat() method of the inherited Animal class is called.
When we look at the table again, the Dog class has move() and speak() methods. Although these methods are in the inherited class, they have been rewritten using ‘OVERRIDE’. For this reason, when these methods are called in main, the related class’s move() or speak() class is called.
Output of the code:
---------- For Animal ----------
This is an Animal!
Moving... - Animal
Speaking... - Animal
Eating... - Animal
Type: Animal
Name: Animal
------------ For Dog ------------
This is an Animal!
This is a Dog!
The Dog is running.
Woof! Woof! Woof!
Eating... - Animal
Type: Animal
Name: Spike The Dog is deleting...
Animal is deleting...
Animal is deleting...
!! Also pay attention the order of calling Constructors and Destructors !!
2- Multilevel Inheritance
Multilevel inheritance refers to a mechanism in OOP where a class can inherit from a derived class, thus making that derived class the base class for the new class.
UML for Multilevel Inheritance
Let’s see Multilevel Inheritance Code
Output of the code:
---------- For Animal ----------
This is an Animal!
Moving... - Animal
Speaking... - Animal
Eating... - Animal
Type: Animal
Name: Animal
------------ For Dog ------------
This is an Animal!
This is a Dog!
The Dog is running.
Woof! Woof! Woof!
Eating... - Animal
Type: Animal
Name: Spike
------------ For Baby Dog ------------
This is an Animal!
This is a Dog!
This is a Baby Dog!
The Dog is running.
Weef! Weef! Weef!
Eating... - Animal
Type: Animal
Name: Tyke The Baby Dog is deleting...
The Dog is deleting...
Animal is deleting...
The Dog is deleting...
Animal is deleting...
Animal is deleting...
!! Also pay attention the order of calling Constructors and Destructors !!
3- Hierarchical Inheritance
When several classes are derived from same base class it is called hierarchical inheritance.
UML for Hierarchical Inheritance
Let’s see Hierarchical Inheritance Code
When we look at the UML table, it is seen that the Dog and Bird classes do not have the eat() method, while the Fish class does not have eat() and speak() methods. All three classes have these methods because they inherit from the Animal class.
When these methods are called in Main, the eat() and speak() methods of the inherited Animal class are called.
When we look at the table again, the Dog and Bird have move() and speak() methods, and the Fish class has the move() method. Although these methods are in the inherited class, they have been rewritten using ‘OVERRIDE’. For this reason, when these methods are called in main, the related class’s move() or speak() class is called.
Output of the code:
---------- For Animal ----------
This is an Animal!
Moving... - Animal
Speaking... - Animal
Eating... - Animal
Type: Animal
Name: Animal
------------ For Dog ------------
This is an Animal!
This is a Dog!
The Dog is running.
Woof! Woof! Woof!
Eating... - Animal
Type: Animal
Name: Spike
----------- For Bird -----------
This is an Animal!
This is a Bird!
The Bird is flying.
Chirp! Chirp!
Eating... - Animal
Type: Animal
Name: Tweety
----------- For Fish -----------
This is an Animal!
This is a Fish!
The Fish is swimming.
Speaking... - Animal
Eating... - Animal
Type: Animal
Name: Nemo The Fish is deleting...
Animal is deleting...
The Bird is deleting...
Animal is deleting...
The Dog is deleting...
Animal is deleting...
Animal is deleting...
!! Also pay attention the order of calling Constructors and Destructors !!
4- Multiple Inheritance
When an object or a class can inherit characteristics and features from more than one parent class, it is called Multiple inheritance.
UML for Multiple Inheritance
Derived class syntax for C++:
class derivedClass
: visibilityMode BaseClass1, visibilityMode BaseClass2
{
// body of the derivedClass.
}
Let’s see Multiple Inheritance Code
When we look the main function, we created an A Class object and we call function() method that is inherited from Class C and Class B. The compiler can not know which inherited function() method have to call. For this reason, we specified that which method of class will be called.
a.C::function();
a.B::function();
Output of the code:
This is constructor for B class.
This is constructor for C class.
This is constructor for A class. Function of C class
Function of B class This is destructor for A class.
This is destructor for C class.
This is destructor for B class.
!! Also pay attention the order of calling Constructors and Destructors !!
5- Hybrid Inheritance
Hybric inheritance is a combination of multiple inheritance and multilevel inheritance. A class is derived from two classes as in multiple inheritance.
UML for Hybrid Inheritance
Let’s see Hybrid Inheritance Code
When we look the A Class’s function() method, we use variable that is inherited class B or class C. Compiler do not know that which class’s variable have to inherit. For this reason, we use virtual keyword. We say to the compiler that only inherit one member thanks to virtual keyword.
class B : virtual public D,
......
class C : virtual public D
......
Avoiding multiple inheritance as much as possible is necessary.
Output of the code: | https://medium.com/swlh/what-is-inheritance-d4787d77e9b9 | ['Muzaffer Arda Uslu'] | 2020-12-19 10:32:22.264000+00:00 | ['Oop', 'Java', 'Object Oriented', 'Cpp', 'Inheritance'] |
Why I Stopped Penny-pinching and Can Enjoy My Damn Coffee | Why I Stopped Penny-pinching and Can Enjoy My Damn Coffee
Taking control of my finances helped keep my financial anxiety in check
Designed by Freepik
We live in a strange time in human history. With all of the complex challenges we face and opportunities before us in the world, we spend an inordinate amount of time worrying about petty issues like how much people spend on coffee.
In this article, I’ll review my experience with the “latte factor,” why I used to think it was a big deal and why I moved past it. I’ll also review how I ensure I reach all of my financial goals without stressing every time I go to a coffee shop, which I assure you is a much more enjoyable way to live.
What is the Latte Factor?
The “latte factor” is a term first coined by David Bach in his book The Automatic Millionaire.
The idea behind the latte factor is that seemingly small and routine purchases like buying a latte or some other expensive coffee can add up to large sums over time.
Bach argues that by cutting out these small purchases and automating the savings towards saving and investing, you can generate much more wealth over your lifetime.
Suze Orman is another well-known personality in the personal finance space who once claimed that Millennials are “pissing $1 million down the drain” by going to coffee shops rather than brewing every cup at home.
If you are interested, I have reviewed the dubious math Orman uses to make that claim right here.
I used to subscribe wholeheartedly to the latte factor.
The central argument behind the latte factor is that small purchases have an opportunity cost. When I first heard about the latte factor, I was finishing my master's degree in economics, so the concept was extremely appealing to me for two reasons.
1. I was an economic student, so I had a positive bias towards anyone using the term “opportunity cost” in their argument.
2. I was broke and in debt.
In personal finance, we use clichés like “every penny counts” too often. During graduate school, though, every penny did count for me.
I was working 2–3 jobs during the time, but between tuition, housing, and sending money back home to help family, I barely had enough time to get by.
When I got hired for my first “career job” out of grad-school, I had a few hundred bucks in my account and would not have been able to make rent if my job search went on any longer.
During that time, the latte factor made a huge difference. By cutting all of my expenses to the bone and throwing every penny, I had against the student loans, I was able to take control of my finances within a few years of graduation.
As my circumstances changed, my views have evolved
When I met the woman, who is now my wife, I began to slowly shed my habits of extreme frugality. She was responsible with money but did not share my warped view that money was to be hoarded.
I began spending money at restaurants, traveling, and even the dreaded coffee shop for overpriced coffee. At first, I felt sharp pains of anxiety every time I spent money. The psychological scars of feeling like I would never have enough money were still fresh.
As time passed and my financial position continued to improve, those psychological scars began to heal. As my income continued to increase, I began focusing less on pinching every penny and more on increasing my income and acquiring assets.
Today, an indulgence like going to a coffee shop does not bother me. That’s not to say I am hitting Starbucks every single day, but I do enjoy spending a Saturday afternoon writing at a coffee shop.
I’m still on track to become an “Automatic Millionaire”
In his book, the Automatic Millionaire, Back recommends cutting out all expenses like lattes and automating the savings into an investment account. He refers to this process as the path to becoming an “automatic millionaire.” Meaning you can save $1 million through setting up an automated savings plan.
While I am not so concerned about the “latte factor,” I am a huge proponent of automatic your financial goals.
I figure out how much money I need to save to achieve my goals, set up an automated savings plan, and spend whatever is leftover.
For example, If I crunched the numbers and figured out I needed to save $1,000 per month to reach my financial goals; I would call my bank, set up an automated process for $1,000 to come out of my checking account on the same day every month to be used towards saving and investing and then spend whatever is leftover.
This means I can enjoy my latte without freaking out about “the opportunity cost.”
It’s a much less stressful way to go through life, and I am a happier person for it. | https://medium.com/makingofamillionaire/why-i-stopped-penny-pinching-and-can-enjoy-my-damn-coffee-2523d97e1ad3 | ['Ben Le Fort'] | 2020-12-06 17:47:28.336000+00:00 | ['Life Lessons', 'Personal Finance', 'Psychology', 'Money', 'Personal Development'] |
5 Reasons to Use Virtual Reality for Data Visualisation | 5 Reasons to Use Virtual Reality for Data Visualisation
Virtual reality, through its ability to immerse us in new environments, could be the next big step in data visualisation efforts
The world is exploding with data.
It’s now pouring out of every home, office, computer, mobile, machine and human being. Data has become so vast and unwieldy that we’ve upgraded its status to ‘big data’. The visualisation of this data therefore plays an increasingly important role to help us break down its complexity.
Virtual reality is a new medium that offers a lot of potential for data visualisation. By immersing ourselves in the data, we can take advantage of the greater space on offer, more natural interactions, and analyse multi-dimensional data in a visceral way.
By living in a world of data rather than being a spectator in it, studies have shown that the effectiveness of the visualisation is multiplied. This leads to ‘demonstrably better perception of a datascape geometry, more intuitive data understanding, and a better retention of the perceived relationships in the data.’ ¹
Many will argue that the 3D spaces and experiences that VR brings can be emulated on a 2D screen. This is true to some extent but the 5 differentiating reasons for using virtual reality for data visualisation purposes are:
Less distractions
By focusing your entire field of vision, you can hone your concentration on the objective, whether that is to help archaeologists visualise the location of objects of interest during an excavation or to take users on a guided tour of 21 years of the Nasdaq.
Through being ‘present’ in the data you can also get a true sense of scale which is difficult to achieve when viewing the data on a desktop screen.
2. More space
With a 360 degree sphere of space to use, there is a lot more real estate to display data such as in the below proof of concept by Bloomberg of its virtual reality trader terminal.
3. Multi-dimensional data analysis
We primarily use our sight to analyse and interpret data, but what if we could also use our hearing? Through data-audio relationships, we could understand the significance, subject, and location of a particular data point through the loudness, type and direction of the sound, for example.
By using multiple senses, we can enhance our ability to process data with more dimensions. While it may be a bit radical to talk about taste and smell in a data visualisation context, it is not outside the bounds of possibility to ‘feel’ data. This is technically achievable right now with haptic feedback gloves.
4. Greater bandwidth for processing data ²
Much like a computer, our optic nerve is capable of transferring information at about 1 MB/s. When we simply read words on a screen we’re only using 0.1% of this capacity. Naturally, this would have improved with visualisation techniques that have been developed over the years, but at the end of the day this is still about reading information from a 2D screen.
Virtual reality immerses you in a stimulating 3D world that engages your brain and enables you to fully utilise your optic nerve’s bandwidth.
5. More natural interaction
In the real world we interact with objects directly with our hands. This allows us to connect with the environment around us and get a better idea of the objects we’re dealing with. For a long time, we’ve used keyboards and mice as conduits for this interaction. Through virtual reality, we can return to a more natural way of interacting — by physically pushing buttons, moving windows around and manipulating data streams (such as in this VR assisted biological specimen analysis). That is in addition to being able to walk around and through these data worlds.
Through these benefits, we can improve employee efficiency, conduct a deeper analysis of data more easily, and make faster decisions.
There are still some obstacles to overcome before data visualisation in virtual reality really takes off. The resolution of the headsets needs to increase so text is comfortably legible, and eye strain and nausea are still an issue for a segment of the population. One of the key challenges is not building a VR data visualisation just for the sake of it — we need to design useful visualisations which take advantage of VR’s strengths and offer an intuitive way of interacting, analysing and manipulating the data.
With the hardware developing and our understanding of this new technology improving every day, it’s only a matter of time before we see data being usefully visualised in virtual reality.
References | https://towardsdatascience.com/5-reasons-to-use-virtual-reality-for-data-visualisation-86cd37d5c1ee | ['Jeremy Dalton'] | 2017-02-13 18:10:39.289000+00:00 | ['VR', 'Data Science', 'Virtual Reality', 'Big Data', 'Data Visualization'] |
Five Scientific Journals to Follow as a Data Scientist | Five Scientific Journals to Follow as a Data Scientist
How to Stay Up-To-Date With the Scientific Community
Photo by Susan Yin on Unsplash
The field of data science is advancing at an incredible pace. New scientific articles are published daily. As a student, I try to stay up-to-date with the scientific literature that is published. In this blog post, I created a list of scientific journals that I believe every data scientist should follow. The journals are presented in no particular order.
International Journal of Data Science and Analytics
The journal welcomes experimental and theoretical findings on data science and advanced analytics along with their applications to real-life situations.
The first scientific journal on the list is the International Journal of Data Science and Analytics (JDSA). This is no surprise since this was the first scientific journal in data science and analytics. The editor-in-chief is Longbing Cao from the University of Technology in Sydney, Australia.
The journal contains articles from many subdomains in data science mostly focussing on machine learning and big data. It brings together thought leaders, researchers, industry practitioners, and potential users of data science and analytics, to develop the field, discuss new trends and opportunities, exchange ideas and practices, and promote transdisciplinary and cross-domain collaborations [1].
Link to a recent edition: Volume 10, issue 4, October 2020 | https://towardsdatascience.com/five-scientific-journals-to-follow-as-a-data-scientist-bc50f3590bc2 | ['Ruben Kerkhofs'] | 2020-09-06 16:55:11.308000+00:00 | ['Analytics', 'Data Science', 'Artificial Intelligence', 'Education', 'Research'] |
An In-Depth Introduction to Sparse Matrix | A comprehensive comparison of various sparse matrix formats
Recently, as I dig deeper into recommender systems, I found that the sparse matrix is the best data structure to represent the commonly used user-item matrix. In this review blog post, I will introduce scipy.sparse in-depth, and share some lessons I have learned so far.
Motivation to use sparse matrix
In recommender systems, we have the data on a rating of a user on a specific item. For example, on Amazon, a buyer may have a 4.5-star rating on the book Gregory Mankiw’s Principles of Economics. Given there are millions or even billions of users, and hundreds of thousands of book titles on Amazon, it is not possible for every user to give a rating on every single book. Often times, we will have most of the users rating on several books, and most of the books will only have some tens to hundreds of users to rate it. If we imagine this as a matrix where the rows represent the user ids, whereas the columns represent the book id (i.e. item), then we have a very large matrix, but only with a small percentage of it filled in. Most of the entries in this matrix will be empty. This matrix, with most of the values missing or empty, will be called a sparse matrix. The task of a recommender system is often to predict the ratings on these missing entries in order to make recommendations to users on new items.
Example user-item matrix encountered in recommender systems
We can, in theory, fill all these missing values with zeros, and use numpy.ndarray to pandas.DataFrame to represent it. But every 0 we store in our memory or disk is probably a waste of memory or storage space. Also, if we fill these missing values with 0, then we wouldn't be able to distinguish cases where a user literally rate a book with 0 stars (although most rating systems probably do not allow it, perhaps also consider other types of data with potentially 0 values distinguished from missing values). Therefore, it probably makes more sense to store only the entries of the matrix where it really has some data. Sparse matrix data structures are designed exactly for this purpose.
Scipy’s sparse matrix class
scipy.sparse has 7 different sparse matrix structures, each has its pros and cons in its use in practice. All 7 types are inter-convertible from each other. Some are good for slicing the matrix and do arithmetic operations on, while others are better used for constructing the data matrix from other data structures. We will analyze each one of them.
Compressed Sparse Column Matrix ( csc_matrix)
csc_matrix is good for column-wise slicing and fast arithmetics including + , - , * , / , power ** and matrix multiplication @ . csc_matrix is not efficient when adding a new element to the data, nor it is the best choice for constructing a new sparse matrix.
csc_matrix stores 3 pieces of information regarding the sparse matrix, the data itself as an array ( data ), the indices of the rows ( indices ), and the indirect register pointer ( indptr ) with respect to the indices . As an example, suppose we have the following sparse csc matrix
Then its csc representation is
csc_matrix example
data is simply a column-wise flattened version of the matrix
is simply a column-wise flattened version of the matrix indices is the row indices for the corresponding elements of the data , e.g., the first element in the data is a 1, and it is located in the row index 1(second row); the second element in the data is a 2, and the row index is 3(fourth row), etc...
is the row indices for the corresponding elements of the , e.g., the first element in the data is a 1, and it is located in the row index 1(second row); the second element in the data is a 2, and the row index is 3(fourth row), etc... indptr is a pointer for data and indices , an array of length col + 1, with max of its element as the length of the data or indices . It is representing the range of column indices that each element in the data belongs to. In the above example, we have
Column 0 has data and indices between indices A.indptr[0]=0 and A.indptr[0+1]=1 , i.e. A.data[0:1]=1 and A.indices[0:1]=1 Column 1 has data and indices between indices A.indptr[1]=1 and A.indptr[1+1]=2 , i.e. A.data[1:2]=2 and A.indices[1:2]=3 Column 2 has data and indices between indices A.indptr[2]=2 and A.indptr[2+1]=4 , i.e. A.data[2:4]=[1,2] and A.indices[2:4]=[2,4] Column 3 has data and indices between indices A.indptr[3]=4 and A.indptr[3+1]=6 , i.e. A.data[4:6]=[2,1] and A.indices[4:6]=[0,4] Column 4 has data and indices between indices A.indptr[4]=6 and A.indptr[4+1]=7 , i.e. A.data[6:7]=2 and A.indices[6:7]=2
Compressed Sparse Row Matrix (csr_matrix)
csr_matrix is good for row-wise slicing and like csc_matrix , fast for arithmetic operations including + , - , * , / , power ** and matrix multiplication @ . csr_matrix is also not efficient when adding a new element to the data.
The data structure of csr matrix is similar to that of csc matrix, except that the indices indicates the indices of the columns. For the same example matrix, the csr representation is
where
data is a row-wise flattened array
is a row-wise flattened array indices are indices of the columns
are indices of the columns indptr is a pointer for data and indices , an array of length row + 1, with the max of its element as the length of the data or indices .
csr_matrix example
Block Sparse Row matrix (bsr_matrix)
bsr_matrix has all the advantages of csr_matrix . In addition, when data is dense at certain regions of the matrix, arithmetic operations can be even more efficient than csr_matrix . To efficiently store the sparse matrix as bsr format, it is required that the block size are all the same, and are divisible by the number of rows and columns, i.e. if the block size is (R, C) and matrix size is (M, N), then it is required that M % R = 0 and N % C = 0.
The bsr sparse matrix representation has the following structure:
data : a K × R × C matrix, where K is the number of blocks, and R and C are dimensions of each block
: a K × R × C matrix, where K is the number of blocks, and R and C are dimensions of each block indices : length K array, column indices of the blocks
: length K array, column indices of the blocks indptr : length K+1 array, pointers of block data and indices
For the following example matrix, the bsr representation is
bsr_matrix example
Coordinate Sparse Matrix (coo_matrix)
coo_matrix is the best and fastest format for constructing a new sparse matrix using large arrays and row/column indices. Once the sparse matrix is constructed, one can easily cast it from coo to other sparse matrix formats.
A feature of coo_matrix format when constructing a sparse matrix is that it allows repeated rows and columns. When converting to other sparse matrix formats such as csc_matrix or csr_matrix , the duplicated (row, col) entries will be summed together. This is especially helpful when constructing finite element matrices.
To construct a sparse matrix without summing the duplicated entries, one can use dok_matrix .
For the following example matrix, the coo_matrix structure is
coo_matrix example
Dictionary of Keys Sparse Matrix (dok_matrix)
dok_matrix is the best format to use when we try to update/replace the elements in a sparse matrix. It is basically a dictionary of the format {(row, col): data, ...} .
keys() : this method will return the (row, col) tuples which index the data
: this method will return the (row, col) tuples which index the data values() : this method will return the data stored
For the following matrix, the dok_matrix 's equivalent dict structure is
dok_matrix example
Row-based List of List Sparse Matrix (lil_matrix)
lil_matrix is also a good format to construct a sparse matrix, but only if the matrix does not involve a lot of data. For larger data construction, consider using coo_matrix . lil_matrix stores the following two pieces of information about the sparse matrix
rows : lists of column indices for each row.
: lists of column indices for each row. data : lists of list of row elements.
For the following example matrix, when converting to lil sparse matrix, the structures are
lil_matrix example
lil_matrix offers a very simple and flexible interface to slice data. A key advantage of using lil format to slice the matrix is that it allows row and column indices to be broadcasted.
Suppose we have an array of row indices rows and column indices cols , to slice out a subset of the sparse matrix for every rows and cols , we can do
Diagonal Storage Sparse Matrix (dia_matrix)
dia_matrix is the best format to use when constructing diagonal matrices, or off-diagonal matrices. It stores 2 pieces of information about the sparse matrix.
offsets : an array of diagonal offsets. 0 means at the main diagonal of the matrix; a negative value means below the main diagonal; a positive value means above the main diagonal. The array is usually sorted ascendingly (e.g. below main diagonal, main diagonal, above main diagonal)
: an array of diagonal offsets. 0 means at the main diagonal of the matrix; a negative value means below the main diagonal; a positive value means above the main diagonal. The array is usually sorted ascendingly (e.g. below main diagonal, main diagonal, above main diagonal) data : K × D matrix, where K is the number of diagonal terms, and D is the size fo the main diagonal. Note that if the diagonal is at the off-diagonal (i.e. not at the main diagonal), then
If this is diagonal is below the main diagonal (offset < 0), then trailing zeros will be added to the array to match to the size of the main diagonal array If this is diagonal is above the main diagonal (offset > 0), then leading zeros will be added to the array to match to the size of the main diagonal array These zeros values can actually be any other values, but it won’t be part of the matrix. In the scipy.sparse.dia_matrix document example, the dia matrix was constructed using 3 identical length vectors, placed at the diagonals [-1, 0, 1] . Although, when accessing the data via A.data , it was shown that there were no zeros filled right after the initial construction, casting first to coo_matrix then back to dia_matrix and access data with A.data again will show these leading and lagging zeros.
For the following example matrix, when converting to lil sparse matrix, the structures are
dia_matrix example
Summary
The following table summaries the ways to construct and update the 7 sparse matrices. For dok_matrix , lil_matrix , and dia_matrix , I have included some in-object methods to update the matrix, which are not found in scipy ’s documentation.
Construction and Update of 7 Sparse Matrix Formats and Inter-conversion
The following table makes a comparison of the performance of several types of operations for each sparse matrix. In general, csc_matrix , csr_matrix , and bsr_matrix are great for arithmetic operations, while others are better for matrix construction.
Comparison of all 7 sparse matrix data structures in a table
Sparse Matrix Broadcasting
There are many cases where we want to add, subtract, or normalize over columns, rows, or the entire matrix, without affecting the missing data. In this section, we are going to consider several cases of sparse matrix broadcasting.
Scalar Broadcasting
Unlike numpy , simply adding a scalar on the sparse matrix with mat_csc + 1 will give us the error
NotImplementedError: adding a nonzero scalar to a sparse matrix is not supported
Fortunately, the actual data of the matrix is stored in .data attributes as a numpy.ndarray . Therefore, we can simply modify the .data attributes,
mat_csc.data += 1
Row-wise broadcasting
Suppose we have the csr sparse matrix mat_csr of size $M \times N$ and a row vector array_r of length $M$. We want to add (or subtract or element-wise multiplication or subtraction) each row of the sparse matrix by the row vector. In numpy , we can simply do
mat + array_r[np.newaxis, :]
With Sparse matrix, we again can access the .data attribute. We can also access the column indices of each data point via the .indices attribute, then the equivalent sparse matrix row-wise broadcasting is then
mat_csr.data += np.take(array_r, mat_csr.indices)
For other types of sparse matrices, first, convert to csr_matrix with tocsr() , make the broadcasting calculation, and then convert it back to the original format.
This can be implemented as
Column-wise broadcasting
Similarly, for column-wise broadcasting, suppose we have csc matrix mat_csc of size M × N and a vector array_c of size N. If we want to broadcast the vector to all the columns of the sparse matrix, we can do
mat_csc.data += np.take(array_c, mat_csc.indices)
Again, for other types of sparse matrices, first, convert to csc_matrix with tocsc() , make the broadcasting calculation, and then convert it back to the original format.
This can be implemented as
Sparse multiplication
In matrix factorization problems, we often need to compute a loss function after the prediction by multiplying the two factorized dense matrices. But since the loss function should only be calculated where the data is available on the large sparse matrix, it is only necessary to make the multiplication at a small subset of combinations of rows, instead of the full matrix multiplication.
Suppose we have two dense matrices U (N × K) and W (M × K) and suppose N and M can be very large, but K is very small. We have a user-item rating sparse matrix R, and we are trying to use the product
to predict R. Since R is sparse, it is better that predicted R is also sparse to avoid unnecessary computations.
A trick is to use coo_matrix to determine which combinations of rows and columns of R needs to be predicted. Then slice out only the necessary rows from the two dense matrices (for necessary rows of U, they would be the rows of R; for necessary rows of W, they would be the columns of R). After selecting only the necessary rows, we can do dot products on these sliced set of row vectors. The following snippets implement this.
Sparse Matrix in Other Python Libraries
In addition to scipy 's sparse matrix implementations, other Python libraries also have sparse data structure implemented, including but not limited to:
pandas : implemented as types SparseArray and SparseDataFrame , similar to coo, and interfaces with scipy.sparse.coo_matrix very well
conversion from pandas to scipy.sparse.coo_matrix with df.sparse.to_coo() conversion to pandas from scipy.sparse.coo_matrix with pd.Series.sparse.from_coo() | https://medium.com/swlh/an-in-depth-introduction-to-sparse-matrix-a5972d7e8c86 | ['Edward Cui'] | 2020-07-03 20:21:30.323000+00:00 | ['Scipy', 'Sparse Matrix', 'Data Structures', 'Python', 'Recommender Systems'] |
It’s Time for Tech Workers to Get Political | Jamal Khashoggi’s horrific murder should be a wake-up call. It may be the nth such canary in the global coal mine—the Committee to Protect Journalists reports that 28 journalists have been murdered in 2018—but Khashoggi’s brutal slaying in the Turkish consulate seems to have caught the nation’s attention. Because of Saudi Arabia’s deep ties to the Silicon Valley investment architecture, even the typically aloof captains of technology industry have been forced to respond.
Saudi Arabia, a perennial oil-economy powerhouse, has been reading the writing on the wall. Alternative, renewable, and less environmentally damaging energy sources become cheaper and more efficient with each passing year. Simultaneously, consumer aversion to oil-based products continues to rise. The kingdom’s plan to diversify their assets involves huge investments in the technology world, and Bloomberg reported that the Saudis hope to invest $2 trillion into the tech sector—much of that in Silicon Valley startups.
The Saudi government has invested $3.5 billion into Uber, and they’ve invested more than $45 billion into a fund owned by SoftBank called the Vision Fund. Between the princes, the government-run Public Investment Fund, and their investments via SoftBank, the Saudis own stakes in lots of big names, including Nvidia, Slack, DoorDash, Twitter, WeWork, ARM, Snapchat, Uber, Lyft, Tesla… the list goes on.
Many business leaders, including Uber’s CEO and the CEO of SoftBank, are publicly snubbing the Saudis by withdrawing from the Saudi investment conference. It’s better than nothing, but the technology world needs to do more than play Mean Girls with the Saudis. Tech companies that want to stand up for free speech should do everything they can to forcibly buy back shares tied to the Saudi government. Startups seeking investment should make it known they will refuse to take Saudi money. Big investment groups like Kleiner Perkins should similarly pledge they will not invest in a company that agrees to accept Saudi investment.
It’s unlikely that the CEOs and executives of these big tech firms will do much more than chatter without a bigger incentive.
The willingness of tech companies so far to get on the Saudi dole speaks volumes about their false commitment to idealistic “core values” that drive business decisions. It’s not as if Saudi Arabia became an international villain overnight; the repressive kingdom has a long and brutal history that stands in almost perfect opposition to the cosmopolitan values that Bay Area companies ostensibly align themselves with.
I’d love to see the industry titans make meaningful commitments to important global issues such as free speech. However, it’s unlikely that the CEOs and executives of these big tech firms will do much more than chatter without a bigger incentive. United workers could provide that incentive.
Take the Power Back—From Organizing to an Organization
Workers at Google, Amazon, and Microsoft have provided a template for shifting the decision-making process of their corporate governance. Google employees have spoken out against Project Maven, Amazon employees against Rekognition (more than once), and now Microsoft employees are opposing a U.S. military project called JEDI.
By writing and signing open letters, threatening to quit, and making demands, the highly valued technology workforce can flex its muscles and push big organizations to change. Google did not renew Project Maven and published a set of AI principles in line with the complaints levied by employees. On the other hand, Amazon continues to pitch its Rekognition software to government agencies such as ICE.
Tech workers should now call on their companies to sever ties to Saudi Arabia. But if workers want to see holistic change and serious commitment to ethical behavior, they need to do more than single-issue ad hoc organizing. They need to form organizations that span more than a single company.
The professionals of the software world need something akin to the Hippocratic oath. Companies seeking to deploy technology to less-than-ethical ends need to know it will be much harder to get anyone to build it. The first step, though, is for technology workers to think critically about the impact their work has had on society so far—especially close to home.
Technology firms and technology workers are making decisions and designing software to exploit contractors.
As conditions in the U.S. (and worldwide) continue toward increased economic stratification, the need for an organized labor force grows. The technology community is partly to blame for this growing inequality, through automation of manual labor to the invention of the “gig economy,” which insulates tech companies like Uber from the drivers who actually deliver the service to customers. Same goes for AirBnB, TaskRabbit, and the myriad other technology firms that rely on the meat cloud of mechanical turks to enact their corporate machinations.
By employing these workers as contractors, corporations avoid paying for benefits, avoid payroll taxes and liability, and generally shirk the responsibilities associated with employing workers. Not only that, but technology firms and technology workers are making decisions and designing software to exploit those workers. Susan Fowler, who famously exposed the darker underbelly of Uber, wrote a chilling piece for Vanity Fair, which included this unsettling paragraph:
A few weeks into my tenure at Uber, where I started as a software developer just a year after graduating from college, still blindly convinced I could make the world a better place, a co-worker sat down next to my desk. “There’s something you need to know,” she said in a low voice, “and I don’t want you to forget it. When you’re writing code, you need to think of the drivers. Never forget that these are real people who have no benefits, who have to live in this city, who depend on us to write responsible code. Remember that.” I didn’t understand what she meant until several weeks later, when I overheard two other engineers in the cafeteria discussing driver bonuses—specifically, ways to manipulate bonuses so that drivers could be “tricked” into working longer hours. Laughing, they compared the drivers to animals: “You need to dangle the carrot right in front of their face.” Shortly thereafter, a wave of price cuts hit drivers in the Bay Area. When I talked to the drivers, they described how Uber kept fares in a perfectly engineered sweet spot: just high enough for them to justify driving, but just low enough that not much more than their gas and maintenance expenses were covered.
The technology workers I’ve met are generally disheartened—often devastated—by the state of things. Perhaps ironically, the relatively wealthy professionals also feel disempowered to do anything about it. The economy is too complex, the power structures too powerful, the market incentives too unchangeable. And it’s true; as individuals, technology workers are relatively powerless to force change. But as an organized group, they could be a juggernaut in the economics of not just California or the United States, but worldwide.
I am hardly the first to suggest such collective action. The Tech Workers Coalition was formed with hopes to “redefine the relationship between tech workers and Bay Area communities … [t]hrough activism, civic engagement, direct action, and education.” Silicon Valley Rising is another organization that aims to, “[take] on occupational segregation and severe income inequality with a comprehensive campaign to raise wages, create affordable housing and build a tech economy that works for everyone.”
Writers at the Atlantic and the New York Times have suggested that technology workers unionize. White-collar workers are typically reluctant to bite the hand that feeds them, but the status quo won’t last forever. Wired magazine predicts that between the rise of coding boot camps, internal training programs for coders, and an increasing number computer science graduates that coding could become a largely blue-collar job.
The workers of the technology world need to use their power before capital can devalue them.
Ben Tarnoff suggested in a piece for the Guardian that technology firms are counting on and encouraging a shift toward programming becoming a blue-collar profession. The shareholders at Uber, Google, and elsewhere don’t want to pay programming staff top professional salaries; they only have to because of the currently competitive climate. The titans of technology have invested significantly into training programs, school programs, and coding boot camps in order to create a bigger supply of programmers. Doing so decreases the unmet demand, and therefore the value, of programmers in general. Tarnoff concludes that “whether a coder or a carpenter, capital will do what it can to lower your wages, and enlist public institutions towards that end.”
Sparks like the murder of Jamal Khashoggi or the selling-out of “core values” through military contracts need to find some tinder among the middle-class and upper-middle-class technology workforce and become a flame before it’s too late. The workers of the technology world need to use their power before capital can devalue them. Get a taste for collective action through letter-writing campaigns, but don’t let it stop there.
For the good of the cities and communities where we work and where our work is deployed, the empowered among us should demand the entire staff be compensated with a living wage. Doing so could bring prosperity back to the general population, but the CEOs and executives won’t do it without pressure. If programmers wait too long to stand up, they might just join the huddled masses of the gig economy instead. | https://medium.com/s/story/its-time-for-tech-workers-to-get-political-9e5db2fd0a3f | ['Tyler Elliot Bettilyon'] | 2018-10-31 20:42:44.980000+00:00 | ['Politics', 'Startup', 'Unions', 'Gig Economy', 'Economics'] |
Why You Must Never Say ‘I Want To Be a Writer’ | The legendary comic genius Peter Cook performed several sketches in the ’60s as a character called EL Wisty. He was a raincoat-wearing, bland, monotonous, droning, know-it-all and in one famous sketch, he laments on his career, saying;
‘I always wanted to be a judge, but I didn’t have the Latin.’
That was his reason for not becoming a judge.
I wanted to be an astronaut, but when I discovered just how difficult it is to become one, I lost interest.
There is a marked difference to being and wanting to be and when the requirements are stringent and demanding, then you can understand why people give up and that is why the rewards of such careers are so high. But there are degrees of ‘being’ and I’m not about to get all existential here but whilst it probably isn’t possible for the vast majority of the population to be an astronaut, or a high court judge, it is absolutely possible to follow your dreams in other areas.
Maybe you wanted to be a footballer?
Ok, well, like most of us you probably won’t be getting trials for Premier League sides any time soon, but you can play at a lower level, a Sunday league team, a pub team, a 5-a-side team. There are endless opportunities out there.
What has this got to do with being a writer? Or not?
We’ve all fallen in love with the idea of decamping to Provence like Peter Mayle and dashing off a best-seller or sending off another script, story or article for publication.
That’s the dream anyway. And it’s that dream that captures us all, because that’s all it is , a dream. But that’s why we want to be writers. It’s the romance. It’s being able to say to someone when they ask the question;
‘What do you do for a living?’
You can reply,
‘Me? Oh I’m a writer.’
It sounds great, doesn’t it. And that’s why we say things like ‘I want to be a writer’. We are caught up the romantic illusion. The image portrayed by films like ‘Love Actually’ where Colin Firth heads off to France to hole himself up and write his novel.
But think about it. It makes no sense.
What actually is stopping you from being a writer?
Presumably you can physically write? And even if you cant there is voice recognition software available to do it for you.
All you need is a pen and piece of paper and hey presto you’re a writer.
Don’t get confused by the idea that ‘being a writer’ means being PAID to be a writer and I think that’s what everyone means when they say ‘I want to be a writer’. What they actually mean is ‘I want to be paid for my writing’ .
Why?
Well for one it looks like it’s a pretty easy life. You’re your own boss, working the hours you want. Get up at noon, write for a few hours, have a leisurely dinner, a few drinks and go off to bed.
But stop a second. Take a while to look at all the top writers you can think of and ask yourself ‘Is that what they do?’ Have a look at this article from James Clear if you want some insight into just how demanding it can be; The Daily Routines of 12 Famous Writers (jamesclear.com)
Maybe there are a few literary geniuses who can get away with it, but for the majority, it’s a hard and disciplined slog. It means turning up for work every day without fail and putting the hours in. It means not sitting around waiting for ‘The muse’ to grab you, whatever that means.
The opportunities for writing are so enormous these days that there is no-one, but no-one, who shouldn’t be able to produce something and put it out there for all to see.
Maybe you won’t be able to make money from your writing straight away, but you can still practice it.
These days it has never been easier to BE a writer. Don’t ever say you WANT to be a writer. Don’t hold back, don’t hide your light under a bushel. If you want to write, then write and for goodness sake tell everyone you’re a writer. | https://medium.com/the-innovation/why-you-must-never-say-i-want-to-be-a-writer-144d703f4aa4 | ['Glyn Bawden'] | 2020-12-06 18:42:56.044000+00:00 | ['Writing Tips', 'Routine', 'Writing', 'Habits', 'Writing Life'] |
How to Win and Influence Customers — Lessons from Dale Carnegie | How to Win and Influence Customers — Lessons from Dale Carnegie
by Aida Khayatian
Back in 2018, I joined Opex (a division of Llamasoft) as a consultant. I knew that interacting with clients would be a huge part of my job, but I had limited experience with this kind of work. In time, I learned what I needed to know to get along by trial and error, observing my colleagues, and trying to learn from others indirectly by reading about working with clients.
One particular source in that latter category stood out: How to Win Friends and Influence People by Dale Carnegie. This book had a hugely positive impact on my customer relations abilities, and I still fall back on its principles on a regular basis. As a result, I’ve written this article to share with you some tips to help you more effectively deal with a client, whether that’s to sell your products or services or to just create and maintain a positive relationship with potential customers.
The Dale Carnegie Sales Approach
According to Harry A. Overstreet, a psychologist cited in Carnegie’s book, “the best piece of advice which can be given to would-be persuaders, [is to] first arouse in the other person an eager want.” For a customer-facing employee, this is the process of convincing your potential customer that what you’re talking to them about is valuable.
This persuasion can take many forms. An important distinction between effective and ineffective (and also ethical and unethical) sales efforts is whether the salesperson is attempting to genuinely influence the customer to give them consideration, or simply trying to manipulate them into a purchase. Sheer manipulation means twisting logic or facts to sway the potential customer’s opinion — selling the product or service without considering the facts or the customer’s real needs. Even if the salesperson is successful, when clients realize they’ve been swindled, any trust in the salesperson will evaporate.
By contrast, a better long-term approach to arousing Carnegie’s “eager want” is through honest influence. The only way to influence people effectively and ethically is to understand them and their needs, and then show them how to get what they want.
Truly understanding your clients’ needs helps you understand how your product or service benefits them and communicating this effectively is the essence of sales. Asking yourself the following questions can help you understand your potential customers’ needs:
Why should they want to buy what I’m selling?
How would the product or service help them?
What might they find undesirable about the product, and why? How can I overcome these hurdles?
Only after salespeople understand a potential client’s needs can they sell effectively. Below, I’ll discuss how Carnegie suggests that we understand others and show them that we can help solve their problems.
Understanding Your Customer’s Point of View
Early on in my time here, I worked with a client to develop a custom product. The tool was designed to help the client to make better, faster decisions for their supply chain. During product testing, the client brought up some questions about its features and assumptions, I responded with mathematical reasoning. I could tell immediately that the client did not connect with my answer. Fresh out of grad school, I’d failed to contextualize the response, answering with math despite the fact that my client had fifteen years of experience and likely no longer viewed his work in purely mathematical terms. Luckily, my manager quickly stepped in to explain my answer using supply chain concepts instead of mathematical logic, and the client was able to relate to his answer easily. That was when I realized the importance of understanding your customers’ points of view to effective and cooperative conversation.
Reading this book taught me some effective techniques for understanding customers’ points of view. Some of the major methods are as follows:
Become genuinely interested in your customer. Learn about their business to better understand how you can serve them. If what you’re selling isn’t a fit, so be it; understanding this sooner rather than later saves time and effort. If you can help them, focus on learning how you can best do so.
Rather than trying desperately to make the customer interested in your product, encourage them to talk about themselves. One way to do this is by sincerely and honestly admiring their achievements. For example, before meeting with a client, do a quick online search to discover key facts about and important milestones for them and their organization. After doing so, you’ll be able to remark on what you sincerely admire about their business should a pertinent opportunity arise. This will encourage them to talk — we all like to talk about our own successes! You can then ask about the challenges they overcame to get this point, which can further elucidate their needs or overall strategy.
When someone is talking, listen actively. Active listening includes:
Listening with exclusive attention to the person
Letting them know you’re listening with appropriate eye contact
Listening sympathetically, patiently, and with an open mind
Giving responses that are truly relevant to what was said
Confirming that you have heard them correctly
Doing these things sounds easy (and it is), but these tips are often ignored by people in all kinds of conversational contexts. Active listening not only helps you retain information more easily, but also usually results in increased trust.
Showing Customers How You Can Help
When talking with clients, find common ground and focus on points of agreement. Try not to spend much time on subjects on which you and the customer disagree. Respect their opinion, even if you don’t see eye to eye. Never outright tell customers that they’re wrong; if they’re off-base on a matter of pure fact, find a way to correct them diplomatically if necessary, but don’t kill a sale by focusing on a disagreement. Maintain the conversation in a friendly way by emphasizing that both parties want the customer to be better off, and that the only differences are about means, not ends.
Let me share an example from the book. A company that sells electric motors encountered a customer complaint about their motors being too hot. The customer was considering terminating the business relationship, but the company believed to be the complaint to be ill-founded: they had done extensive testing and felt strongly that their motors met temperature standards. Instead of arguing with the customer, the sales rep started conversation with series of questions emphasizing shared beliefs and goals: “You should not buy motors if they are running any hotter than standards, isn’t that so?” The client agreed. The sales rep then asked about the temperature of the room in which the motor is located, and then continued: “If the room is 75 degrees and the motor is 72, would not the motor get hotter because of the room temperature?” The customer agreed — hard to argue with basic thermodynamics. By gently and positively asking about the situation in an honestly inquisitive, perspective-agnostic manner, the company was able to reassure the customer that their motors worked as expected. By avoiding argument and looking at the problem from the customer’s point of view, the company was able to keep the account.
Make the customer feel like they’re actively buying the product. While nobody likes being sold to, people like to buy. Ask them to give you their ideal design. Describe your product in detail, and tell them that while it may not be a perfect fit for their organization, you’re eager to improve it. Then ask them to give feedback on how the product can be more useful to their business. Listening intently to this feedback will further build trust between you and the potential client, and it might even give you some fantastic ideas. If their feedback results in changes to the product/service, it’ll sell itself because it’s based on the client’s very own ideas and needs.
You may not have the exact same experiences using similar products or services to solve the problem they face, but try to share your own experience with the same (or related) challenges. Finding commonality with your customer will help them appreciate that you have a practical understanding of their problems and know what it takes to find a solution. This will help build trust between the two of you, making all future discussion easier and friendlier.
Takeaway
These principles from Carnegie’s book can improve the way you relate to your current and future clients. The best way to positively influence your (potential) clients is to understand their perspective and find ways to show them you can help. Keep their needs at the forefront and you’ll see success in no time!
_________________________________________________________________
If you liked this blog post, check out more of our work, follow us on social media (Twitter, LinkedIn, and Facebook), or join us for our free monthly Academy webinars. | https://medium.com/opex-analytics/how-to-win-and-influence-customers-lessons-from-dale-carnegie-aa877a0b9f8e | ['Opex Analytics'] | 2020-09-25 14:07:29.476000+00:00 | ['AI', 'Customer Service', 'Customer', 'Analytics', 'Data'] |
How to teach sustainable habits to kids | In a world of Greta Thunbergs, the ever-looming threat of climate change, and its already disastrous toll, what can you do to ensure your kids understand the state of the world, without overwhelming them, and help them become part of the solution?
The good news is, many young people are already alert to the climate rumblings around them. “They’re more aware today of these broader, complex socio-scientific issues than I think even our generation was when we were younger,” says Carol O’Donnell, the director of the Smithsonian Science Education Center in Washington, D.C.
She thinks that’s because we now have more evidence to show that humans are negatively impacting the environment, pointing to the consensus among 97 percent or more of climate scientists on this conclusion.
“We have evidence that we are depleting a lot of the resources that we know are necessary for us to live in the way we do,” says O’Donnell.
She wants the Science Education Center’s audience of pre-school through high school students to use that knowledge to develop more sustainable habits, whether it’s via conserving electricity or buying less plastic.
“Having a sustainability mindset means that you are focusing on issues in which the perspective of others is crucial.”
“What practices do we put in place right now to protect ourselves in the future? Folks who make decisions about mask wearing, for example, or social distancing or physical distancing, they’re not often making that decision for themselves. They’re making it to protect others,” says O’Donnell. “Having a sustainability mindset means that you are focusing on issues in which the perspective of others is crucial.”
In an unpredictable world, you’ll want to make sure sustainability is high on the list of skills you impress upon your child. Of course, there are limits to individual actions and the overemphasis on personal carbon footprints is mostly a corporate scam. To really address climate change, companies and governments need to step up. After all, 20 companies are currently responsible for a third of all greenhouse gas emissions.
Even so, it’s important for kids to learn about our role in climate change so they can understand what actions they can personally take to preserve Earth’s resources and how to pressure corporate and government actors. After all, one individual’s action may not make much of a difference, but when we band together, we can create collective change.
Ota Lutz, who leads the STEM K-12 education group at NASA’s Jet Propulsion Laboratory in Pasadena, Calif., points to the now widespread recognition that plastic straws aren’t good for the environment and the resulting plastic straw bans.
“Everyone had plastic straws lying around; well, they don’t anymore,” says Lutz. “More and more, they’re just not around as much.” We’re coming to terms with the fact that our lifestyle is largely not sustainable and, as Lutz points out, individuals coming together can make at least some difference in preserving natural resources.
We spoke with Lutz, O’Donnell, and other experts to learn how parents and caregivers can juggle the balance between not scaring kids about the future of our planet, while also teaching them sustainable habits that will serve us all well in both the short and long term.
1. Motivate your child to explore
“We should be encouraging exploratory, action-oriented learning,” says O’Donnell, “whether that learning takes place in the home or in the community in an informal environment, like at a museum, or whether it takes place in school.”
The action part kicks in when you use the information you gleaned about the world to make a shift in how you live. For example, if your child learns about clean water resources, she might use that new information to take shorter showers or turn the sink off while she brushes her teeth. By encouraging your children to explore nature (if you’re in a city, you also have a valuable learning tool with public parks), they can understand how its systems work and the daily impact they have on the Earth’s resources. Making this connection can motivate them to change their behavior.
“Do something with that information to make a difference,” says O’Donnell.
2. Ask questions, but let children observe what’s around them
If you’re on a walk with your child, try asking, “What do you notice?” suggests O’Donnell. A child might say, “It looks like there used to be water there,” or “the trees were cut down.” Or if you’re in the car or riding public transportation together, you might ask, “What do you think powers our vehicle and where do you think that energy comes from?”
To help inspire children’s own observations about the world, you might then ask, “Why do you think that is?” or “Do you have any questions about why that might be?”
“Then you start to dig into the science,” explains O’Donnell. Help your child understand, following the first example, where the water came from, why it’s no longer there, what humans may have done to impact the water’s disappearance, and what we can do differently in the future.
“Parents will learn from those conversations just as much as children do,” says O’Donnell.
3. Talk about sustainability from a local perspective to connect it to global issues
Most children are more likely to understand large, global issues when they can first learn about them locally, says O’Donnell.
To return to the water example, a parent might inform her child that the reason the water dried up is because their town constructed a dam upstream, blocking the flow of water, to provide hydroelectricity and the jobs that go with it. This can lead a child to consider the economic side of an issue as well as the environmental one. If talking about fossil fuels, you could discuss the cost of oil in your town and what might affect it. This could open up a conversation to introduce your child to geopolitics and how that connects to how much you pay at the pump.
Children, especially the younger they are, are thinking about these problems from their own perspective, says O’Donnell.
“If you’re going to develop students’ sustainable habits, then they have to be driven by local problems, local questions, local observations, and local actions,” says O’Donnell. “And then take those local actions, local decisions, local investigations, and apply them on a broader scale.”
As ever, factor your child’s age and their developmental stage into what topics you introduce, and how you present them.
4. Model the behavior you want
What you do is way more effective than what you say to encourage the behavior you want. If you don’t want your child to use throw away water bottles, you shouldn’t either, says Dr. Chuck Kopczak, a curator at the California Science Center in Los Angeles, Calif., who has a Ph.D. in marine biology and biological oceanography.
Kopczak says the science center is doing that in its own way by trying to eliminate single-use plastic water bottles, partly by replacing sales with boxed water. While not a perfect solution, Kopczak says the center also sells “high quality reusable bottles for sale to guests to use the Science Center’s newly installed water fountain refill stations.”
To use another example, if you’d like your child to understand that fossil fuels are bad for the environment, don’t idle your car when you pick him up after school, suggests Kopczak. These kinds of actions, while seemingly small, can make a big impression on kids and don’t require extra time in your day.
If your child repeats the desired behavior, such as turning off the water while he washes dishes, reward him in some way, but don’t punish him if he forgets, says Kopczak. You can prod your kid in the direction you want but, remember, building a sustainable lifestyle involves habits that take time to set in place.
5. Don’t scare them
Talk honestly to your kids using scientific facts, but be careful not to send them into an anxiety-induced spiral.
“Try not to create the sense that this is the end of the world, but it is a problem that we need to face. That will spread, and they’ll share that with friends,” says Kopczak.
You can use positive examples showing how humanity has worked together to push for progress on a problem, such as closing up the hole in Earth’s ozone layer, or the eradication of smallpox, to inspire confidence.
“The more people we can get involved thinking about things we can do, the more likely we are going to find solutions, and they can be a part of that.”
“We’ve taken on big problems in the past, and we’ve found ways to solve them. This may be the very biggest one we’ve ever faced, because it involves the entire planet, obviously,” says Kopczak. “Convey to them, the more people we can get involved thinking about things we can do, the more likely we are going to find solutions, and they can be a part of that. You don’t have to be a scientist.”
6. Remember: You don’t have to have all the answers
As a parent, you might think you have to know everything — or look like you do — but that’s simply not true.
If your child asks you a question and you don’t have the answer, you can go on a journey together to figure it out, says Lutz.
“Teach kids how to find data,” says Lutz. “[Find] trustworthy sources as opposed to reading some opinion on any given website or newspaper.” You can help guide this practice by helping children learn to vet sources. Ask them to look online for reputable websites, list a few, and ask why they think these particular sites are reliable.
This kind of education helps children figure out how to answer their own questions and not just believe what they’re told, says Lutz. That’s “a sustainable skill for any part of life,” she adds.
7. Keep kids excited to learn and act with these tools
There are also plenty of books, websites, TV shows, and movies kids can dive into to learn and inspire them to live a more sustainable lifestyle, in addition to relying on you as their teacher. We’ve listed links to resources recommended by our experts below, categorized by type.
There’s the animated series The Octonauts, aimed at preschoolers, “that encourages kids’ curiosity about the world by introducing them to some of the sea’s diverse species,” according to a review by Common Sense Media, a nonprofit that rates movies, TV shows, and books for parents. It’s currently available to watch on Netflix.
Or if your child wants to take a break from a screen, there’s the picture book The Watcher, which introduces young kids to the primatologist Jane Goodall, beginning with her childhood, all the way up to her life-long mission to preserve chimpanzees’ habitats.
If you’re looking for something a little more active, you can check out UNESCO’s webpage on the sustainable development goals’ resources for educators, which includes activities and games to help children to learn about topics like water pollution.
You know your child best, so you can of course use your judgement regarding whether something is appropriate for your kid’s learning style and developmental stage.
Books:
“100 of the best books for children on sustainability”
“Top 10 children’s books about sustainability”
“The best books about green living for children of every age”
TV shows:
“5 kid-friendly shows about the environment”
Movies:
“Top 10 kids’ movies with a green theme”
Websites:
Explore the sustainable development goals through kid-appropriate activities, games, and videos
”Raising earth-friendly kids”
”10 ideas for teaching kids about sustainable living”
Correction: An earlier version of this story misidentified the location of the California Science Center. The story has been updated to reflect the correct location. | https://medium.com/mashable/how-to-teach-sustainable-habits-to-kids-bd4db4f12cf7 | [] | 2020-07-28 17:26:00.925000+00:00 | ['Sustainability', 'Parenting'] |
The Earnings Lifecycle of a Medium Story | Learnings
I wrote a script in Python to extract my daily earnings stats and then analyze them further. The reason I did this is that I had many questions regarding my earnings. I then realized that I could share these statistics to help you better understand Medium earnings. Here are the things I learned.
1. Prioritize evergreen stories
The best type of story that you can create on Medium is one that is going to be read by many people for a long period of time. To create evergreen stories, focus on creating content that’s going to be relevant for a long time. You can also create stories with short-lived success but I believe evergreen stories are the best possible investment.
2. Take advantage of SEO to gain longevity in your stories
If your story is able to rank highly on Google, which is fairly easy to do thanks to Medium’s high domain authority, you are able to have long-lived stories that keep on making you money even after the algorithm stops distributing them.
3. Don’t lose faith if a story doesn’t earn a lot in the first week
If your story hasn’t made a crazy amount of money, don’t lose faith. I have stories that made over $30, having made less than $1 during the first month. Having one of your articles picked up by the algorithm at a much later point in time can happen, so don’t rush to think your articles as unsuccessful.
4. Create a backlog of content
The best way to gain significant and steady earnings on Medium is to create a large backlog of content that will keep on producing revenue for you. Don’t try to create the next viral story, but instead focus on offering value and produce stories that people will want to read for a long time.
By doing that, you will have a ton of articles that people keep on reading and will keep on making you money. Ideally, you want your older articles to make more money than your more recent articles. That is the only way you can grow your earnings substantially, without needing to produce new stories every single day. | https://medium.com/illumination/the-earnings-lifecycle-of-a-medium-story-27a154e3bba2 | ['Aris Pattakos'] | 2020-12-23 09:40:35.884000+00:00 | ['Money', 'Writing', 'Personal Development', 'Learning', 'Freelancing'] |
The Aroids of Our Time | Aroids, which make up some of our most recognizable houseplants, belong to a subset of modern plant aristocracy, the Araceae family. Despite their origins in the New World tropics, aroids find themselves resident in rented apartments all over the globe. Priced out of home ownership and resigned to indefinite tenancy, our rootless generation gravitates naturally towards aroids. To own one, or several, is to enjoy a sense of daily achievement, the thrill of keeping something alive. Aroid conservatorship offers us a degree of small, low-stakes responsibility without the gravity of caring for a pet or child. Let’s face it — you can kill a plant by accident and not face any major moral or legal consequences.
Photo by vadim kaipov on Unsplash
The millennial narrative is defined by a series of unfortunate economic events: emerging from university into a global recession, earning salaries insufficient to cover mortgages, delayed marriage and parenthood. And now, with a global pandemic threatening our already tenuous economic future, we tend to consume in smaller, more frivolous ways. The proverbial Starbucks, the iPhones, the travelling paradox for which our generation is often berated are minor expenses compared to the financial clout and stability required to secure and maintain a thirty-year mortgage. Among our favorite comfort buys— aroids. Monstera deliciosa, perhaps the best-loved of these, became the poster-plant of the modern indoor gardening rush some time in the 2010s. Adored by millennials and their younger counterparts, Monstera and the rest of its lush, monocotyledonous family provide us with the illusion of having our own piece of ground from which sprouts green, alive things.
Humans of the early 21st century were not the first to start a houseplant craze. In the 19th century, Victorian Britain enjoyed a fifty-year obsession with ferns. The craze was so notable that, in 1855, the writer Charles Kingsley coined the term pteridomania, wryly referring to fern-collecting young girls whose fashionable new hobby was, in Kingsley’s opinion, more agreeable than duller, less intellectual pursuits like crochet, gossiping and Berlin wool embroidery. There was an obvious element of throwaway consumerism in the fern craze. Victorian fern enthusiasts collected books, magazines and digests on the subject, much like how a modern person, having recently taken up film photography in the digital age, might buy every decently-rated book on the subject in Amazon’s cache to convey an impression of commitment to their new hobby. Since pteridology — fern studies — was still a relatively nascent field, the publications’ authors often coined different names for the same plants and agreed to disagree.
Fernery at the Geelong Botanical Gardens, Australia in the late 19th century
Fern fever went beyond mere cultivation and identification. Architecture, literature, decorative arts and expeditions paid homage to middle-class Britain’s botanical muse. Clearly, Victorians occupied their time in strange, luxurious ways. They went on fern hunts, expeditions for sourcing the most exotic, unusual plants. Localized fern hunting served as a popular theme for ladies’ tea parties. For snagging desirable specimens growing in treacherous rocky outcrops or quick-flowing brooks, a guild of professional fern hunters answered the demand. Some die-hards traveled further afield to the jungles of New Zealand and Papua New Guinea for the finest exotic fronds. For some unlucky fern-atics, the pursuit was lethal. In 1867, Jane Myers plummeted to her death while gathering ferns on a cliff edge in Perthshire, Scotland. Perhaps Charles Kingsley might’ve advised Miss Myers to stick with safer, less daring hobbies after all.
The Fern Gatherer by Charles Sillem Lidderdale (1877)
Back in England, the well-heeled built dedicated spaces for housing and displaying their fern collections. In the first half of the 1800s, George Loddiges, author of the serial Botanical Cabinet and botanical celebrity of the day, built one of the world’s largest greenhouses, the Grand Palm House, in London. To attract visitors to his new ornament, Loddiges hyped up the fern rhetoric, punting claims that ferns, and fern-collecting, boosted virility, improved mental health, and were a sign of intelligence. He may have been right about the mental health bit. | https://medium.com/an-idea/the-aroids-of-our-time-7ebd06b6126 | ['Aimee Dyamond'] | 2020-10-03 00:47:13.315000+00:00 | ['Capitalism', 'Plants', 'Mental Health', 'Houseplants', 'Millennials'] |
Why Our Future Depends on Libraries | I found a beautiful book. In his book amazing book, Neil Gaiman suggests that reading fiction, that reading for pleasure, is one of the most important things one can do. I want to share a passage about the “library” part because everybody should read this part for our future. (You can buy the book from here)
Neil Gaiman made a plea for people to understand what libraries and librarians are, and preserve both of these things.
It is obviously in his interest for people to read, for them to read fiction, for libraries and librarians to exist, and help foster a love of reading and places in which reading can occur. So he is biased as a writer. But he is much, much more biased as a reader.
Everything changes when we read.
People who cannot understand each other cannot exchange ideas, cannot communicate. The simplest way to make sure that we raise literate children is to teach them to read and to show them that reading is a pleasurable activity.
I don’t think there is such a thing as a bad book for children. It’s tosh. It’s snobbery, and it’s foolishness. We need our children to get onto the reading ladder: Anything that they enjoy reading will move them up, rung by rung, into literacy. You are finding out something as you read that will be vitally important for making your way in the world. And it’s this:
The world doesn’t have to be like this. Things can be different.
Fiction builds empathy. Fiction is something you build up from twenty-six letters and a handful of punctuation marks. And you, and you alone, using your imagination, create a world, and people it and look out through other eyes. You are someone else, and when you return to your world, you are going to be slightly changed.
Neil Gaiman says:
“I was lucky. I had an excellent local library growing up, and met the kind of librarians who did not mind a small, unaccompanied boy heading back into the children’s library every morning and working his way through the card catalog, looking for books with ghosts or magic or rockets in them, looking for vampires or detectives or witches or wonders. They were good librarians. They liked books, and they liked the books being read. They had no snobbery about anything I read. They just seemed to like that there was this wide-eyed little boy who loved to read, and they would talk to me about the books I was reading. They would find me other books they would help. They treated me with respect. I was not used to being treated with respect. I was not used to being treated with respect as an eight-year-old.”
Libraries are about freedom. They are places for freedom of reading, ideas, and communication. They are about education, entertainment, making safe spaces, and access to information. All books will or should migrate onto screens. Over twenty years before digital editions showed up, a physical book is like a shark. Sharks are old: There were sharks in the ocean before the dinosaurs. And the reason there are still sharks around is that sharks are better at being sharks than anything else is.
Physical books are sturdy, hard to destroy, bath resistant, solar operated, feel good in your hand: They are good at being books, and there will always be a place for them.
A library is a place of safety, a haven from the world. It’s a place with librarians in it.
We need to teach our children to read and enjoy reading. We need libraries. We need books. We need literate citizens. Books are the way that dead communicate with us. The way that we learn lessons from those who are no longer with us, the way that humanity has built on itself, progressed, made knowledge incremental rather than something that has to be relearned over and over.
We must read for pleasure. If others see us reading, we show that reading is a good thing. We must support libraries and protest the closure of libraries.
If you do not value libraries, you are silencing the voices of the past, and you are damaging the future.
Fiction is the lie that tells the truth. We all have an obligation to daydream and to imagine. It is easy to pretend that nobody can change anything, that society is vast, and the individual is less than nothing. But the truth is individuals make the future, and they do it by imagining that things can be different.
Albert Einstein was once asked how we could make our children intelligent. “If you want your children to be intelligent.” he said, “read them fairy tales. If you want them to be more creative, read them more fairy tales.
I hope we can give our children a world in which they will read, and be read to, and imagine and understand.
*A passage from Neil Gaiman’s “ART MATTERS.” | https://mathladyhazel.medium.com/why-our-future-depends-on-libraries-969851bed55a | ['Hazel Clementine'] | 2020-01-20 08:31:57.280000+00:00 | ['Personal Development', 'Life', 'Libraries', 'Books', 'Education'] |
Type Inspiration 01 — Gilroy. I decided to start a new typographic… | I decided to start a new typographic experiment series. First up for july is Gilroy, the perfect choice for any product, website or app.
Gilroy is a great example of an extremely modern and versatile font. I decided to pair this experiment with a limited colour palette and use the medium and light versions of the font. | https://medium.com/the-innovation/type-inspiration-01-gilroy-b7f22e931426 | ['Debbie Turner'] | 2020-08-03 19:18:59.874000+00:00 | ['Typography', 'Design', 'Inpiration', 'Typography Inspiration', 'Type Experiment'] |
How Big Data & Analytics is solving problems in the Banking and Finance Industry | Modern companies want to know everything about their customers. They want to know where they are going, who they are meeting, what they are eating and of course what they are buying. They are one step ahead of the potential customer, knowing what she’ll do before she does herself. This, of course, only a distant dream just a few years back, is a reality today. The reason behind this is Big Data & Analytics which is changing the way businesses function.
Read More | https://medium.com/data-analytics-and-ai/how-big-data-analytics-is-solving-problems-in-the-banking-and-finance-industry-1687d0ece1ec | ['Ella William'] | 2019-06-15 06:31:53.209000+00:00 | ['Insurance', 'Banking', 'Machine Learning', 'Data Science', 'Big Data'] |
Working out what we want from AI | Thomas Edison reportedly tried 10,000 times before he managed to create a filament for electric bulbs, but he dismissed any talk of failure. “I have not failed,” he corrected his detractors. “I’ve just found 10,000 ways that won’t work.”
We seem to enjoy hearing about mistakes more than celebrating success. Although AI technologies have transformed our lives, at home and at work, many recent media reports focus on the failings of smart devices, from disappointing gadgets on show at the CES to malfunctioning hotel bots. Some of the stories are very funny, but all they tell us is that the technology is still in development and that some products are better designed than others.
The Wall Street Journal writes about a guest in a robot-staffed hotel in Japan who was woken every few hours by the in-room assistant asking him to repeat his command. The hotel manager finally realized that heavy snoring by the guest had triggered the robot’s voice recognition system.
For every clanger, though, there are success stories. Even when a machine achieves something as banal as winning board games it can be the harbinger of transformative benefits. For example, a chess-playing programme called AlphaZero, developed by the Alphabet-owned (Google’s parent) AI research company DeepMind, has been making significant advances.
AlphaZero has developed a new style of playing chess which is much closer to human improvisation than traditional computer chess. That is because AlphaZero learns from its past successes and mistakes, rather than calculating millions of possible permutations as it plays. According to Wikipedia, AlphaZero searches 80,000 positions per second in chess, compared to 70 million for the Stockfish chess engine.
AlphaZero uses (deep) neural network technology — sometimes called deep learning — which has resulted over the past decade from notable improvements in machine learning. As computing power has increased, deep neural networks have produced machines capable of performing tasks in a way that would not have been possible using traditional programming techniques.
This has transformed technologies such as computer vision and natural language processing (NLP), which are nowadays being deployed on a massive scale in many different products and services. Manufacturing, healthcare and finance are just some of the sectors that use deep learning to uncover new patterns, make predictions and guide decision making.
“In the area of smart manufacturing, AI can help to streamline efficiency,” says Wael Diab, who is leading international standardization work in this field. “It can help to provide insights in terms of where improvements can happen and more importantly it can provide insights into where a particular organization may want to go in terms of its production planning.”
Sales of industrial robots have doubled in the past five years, according to the International Federation of Robotics. The IFR predicts that in 2021 the annual number of robots supplied to factories around the world will reach about 630,000 units. Industrial robots are satisfying a real need.
In contrast, much of the focus in consumer electronics is still on the novelty value of gadgets. To a large extent this is because we have not quite worked out yet how we intend to use AI-enabled devices in our everyday lives or what we expect of them.
The Korea Joongang Daily reported in October that Koreans not only use their smart speakers for changing the TV channel, but also to discuss their feelings. In people’s homes, a staggering 15% of the things said to smart assistants appeared to be attempts at conversation, including “I’m bored” and “I’m sad”. The newspaper noted a similar pattern in hotel rooms, where more than 18% of the commands were attempts at conversation. The Joongang Daily acquired the data from KT Corporation, the country’s largest telephone company.
In 2017, IEC and ISO became the first international standards development organizations (SDOs) to set up an expert group to carry out standardization activities for artificial intelligence. Subcommittee (SC) 42 is part of the joint technical committee ISO/IEC JTC 1.
SC 42 is working with other JTC 1 subcommittees, such as those addressing the internet of things, IT security, and IT governance, as well as the IEC Systems Committee (SyC) for Smart Cities. SC 42 has set up a working group on foundational standards to provide a framework and a common vocabulary. Several study groups have been set up to examine the computational approaches to and characteristics of AI systems, trustworthiness, use cases and applications and big data.
IEC Standards are playing a key role in the transition to the Fourth Industrial Revolution. IEC TC 65, for instance, carries out important work related to industrial-process measurement, control and automation.
“We’re looking at the different components that go into AI, from the computational side to the ethical side. Having standards allows for a common language and way for the different stakeholders to interact,” explains Diab.
“What that leads to is the ability to innovate on top of widely adopted standards in the market place.” | https://medium.com/e-tech/working-out-what-we-want-from-ai-23afe9aceeca | ['Mike Mullane'] | 2019-01-26 16:03:44.394000+00:00 | ['Artificial Intelligence', 'Robotics', 'Deep Learning', 'NLP', 'Smart Manufacturing'] |
Estrangement From Myself | Estrangement From Myself
It’s not being “born in the wrong body.” It’s so much more than that.
Photo by The Creative Exchange on Unsplash
I’ve been trying to place these feelings for a while now. I’ve never identified with the classic and easily digestible “wrong body” narrative myself, but I didn’t understand how I could possibly be trans if that wasn’t what was happening to me. If this is your narrative, that’s completely valid, too, but I don’t believe it expresses the essence of transness for everyone. As I was scribbling who knows what in my journal, the perfect word popped up in my head: estrangement. For me, and what might be the case for others, is the feeling of estrangement from existence.
With the advice of my friends and therapist, instead of tackling all my problems at once, I started breaking down everything into feasible chunks. I’d deal with my career separately from when I would deal with my social life, and so on and so on.
I started going to parties more, doing my absolute best as a socially anxious introvert (so quirky and different, I know) to talk to more people, even at the risk of being rejected for my identity. I started looking into careers and thinking about a future I never thought I had. Everything seemed like it should’ve been clicking into place, but no matter how bright my future shone for other people, I couldn’t even see a glimmer of that light. Even now, I can’t really imagine my future. It’s like I’m building a bridge into nothingness.
Tackling all my problems separately was a good idea, but something still wasn’t right.
I knew all my problems were connected, but I wasn’t sure exactly how. No matter what plan of action I took, it was like I was playing the Sims. I saw a paper with a resume and was challenged to make a life for them. Why would I work on all this stuff if it wasn’t even for me? What’s the point of going to parties if I don’t feel like I’m actually there? Why should I get a career when I don’t want anything? What is at the root of all this apathy and depression?
The answer is estrangement. I feel like I have a remote control behind a pane of glass, and I’m controlling who I see in the mirror to live life. I try and try to break through the glass, but nothing works. I’ve been even atoms away from shattering it and jumping into the real world, but I’ve never succeeded. I always end up in the same place, disconnected from everything.
It never made sense to me the way people could just look in the mirror and see themselves. Their bodies, despite their perceived imperfections, were theirs. People have dreams for themselves because they have their own best interests in mind. I can’t even think a day ahead in my life because it doesn’t feel like it’s mine.
They say, if you have nothing else, at least you have your life and your health. I would say the same, but the disconnect between my brain and my life is so strong that I even feel like an outsider to my own self.
When I brush my teeth before bed, I see me, but not me. A thief stole my reflection from me. When I speak, I’m able to produce a voice, but it’s like it’s coming from a far corner of the room. Someone keeps taking my words before I’m able to say them. This burglar has taken everything from me. I keep trying to revendicate what’s rightfully mine, but to no avail. This metaphorical thief has created an insurmountable estrangement in my life.
“Estrangement is like a horde of locusts. It eats away at your will, your desire, and eventually the very essence of you.”
Everyone else I know has gotten out and done what they need to do for themselves, whether that’s going to school for something, working, getting into a relationship, or whatever. I haven’t gotten the chance.
There’s a difference between estrangement and fear, both of which stop me in my tracks. It would be different if I were just scared to do something, because I’ve had moments of courage that let me pull through. Estrangement is like a horde of locusts. It eats away at your will, your desire, and eventually the very essence of you. When you’re afraid, you can at least try to get something done. You might be afraid to study abroad, for example, but it’s based on a desire that makes you even want to try. When you’re completely estranged from yourself, it’s a different story.
While this has gone on a tangent similar to depression symptoms, I feel that this is a special kind of estrangement. Trans people need to compartmentalize themselves. We put the most true and intimate parts of ourselves into a box in the back of our mind that we’ve been told never to open. The fact that it’s even there is something to be ashamed of. Over time, the temptation gets too strong, and we look to see what kinds of things are stored inside. Some people let it all out. Others never even find the box, wondering what’s wrong with them. For me, I feel that my box has an infinite number of compartments. When I feel like I’ve reached the bottom, there’s an IOU.
“Sorry, you don’t really know even the basics about yourself. If you work hard for it, though, I’ll let you have it!”
I keep frantically tearing through boxes, and now I’m just tired of it. I’m tired of feeling separate from what other people consider is a given. I’m tired of having a labyrinth for what for others is a hallway. The estrangement I feel is more than “being born in the wrong body.” I don’t hate my body. I think it’s alright, but it just doesn’t feel like mine. That’s the issue. I feel like an alien on my home planet, and I’m always trying to make a home out of a place that will never be a home.
I try to go to work and it’s like everyone is whispering a secret around me that they don’t want me to hear. The only thing is, that secret is just… life. They have families, they feel in the moment, and I’ll never have that. My body moves on its own and I hang onto my shoulders like a ghost, trying to make sense of everything. I don’t know if I’ll ever break the glass and really feel everything. For now, I guess I just need to get by facing the shadows on the cave wall. | https://medium.com/gender-from-the-trenches/estrangement-from-myself-da23cf309ad4 | ['Stella Luna', 'They She'] | 2020-12-28 23:55:35.725000+00:00 | ['Transgender', 'Identity', 'Mental Health', 'LGBTQ', 'Self'] |
Bill Maher Gave Me the Best Advice by Way of Josef Stalin…Sort Of | No Person, No Problem
I was crying. I usually called Bill when I was crying. Twenty-two years my senior, he was old enough to be my father. Though there was nothing paternal about him, he was never shy about dispensing compassionate, sometimes fatherly advice that would turn my tears into a laugh and my laughter into a good day.
On this particular afternoon, someone had broken my heart. Bill and I ended our relationship the year prior, and I was still reeling from the finality of it all. I still loved him but was doing my best to move forward. In the interest of full transparency, I have since dated and married nothing my losers, abusers, and straight-up crazy people since Bill and I split, so he has often been called upon to ease my tattered spirit –– and he has always answered that call.
I sobbed as I recounted the events of my personal life and the crushing weight of all that had gone wrong. Bill listened before he spoke, and when he did, what came out changed my life.
“No person, no problem,” he said. “They say Stalin said it, but that doesn’t make it any less true,” he continued as he chuckled in his trademark way.
And it hit me like a ton of bricks.
A brief search of this famous quote. which has been largely attributed to Josef Stalin, the former Premier of Russia, uncovered that, “No person. No problem,” isn’t the exact quote and that it hadn’t derived from Stalin at all!
Russian author Anatoly Rybakov wrote the original quote in his novel Children of Arbat. In this fictional account, Stalin is speaking about an army officer’s execution in Tsaritsyn and says, “Death solves all the problems. No man, no problem”.
Yikes.
Taken out of context, “No man, no problem,” or rather, Bill’s version, “No person, no problem,” has solved alot of conundrums for me throughout the years. Naturally, I wouldn’t wish death on a man just because I’m sick and tired of looking at him, nor do I think “death solves all problems.” However, what Bill helped me realize that day is all my heartaches were attached to other people, and once I got rid of those people, the hurt would go away.
After all, I’m not out here breaking my own heart.
So, to this day, when I find myself in a situation the makes me feel unhappy, I remind myself that even though I cannot control others' actions, I can control mine and other people's proximity to me. I can walk away. I can put space between the source of my discontent and me. I can get rid of the person and with them goes the problem. I don’t have to try to solve or resolve it. There is often no working it out, and that’s okay!
Once I’ve removed someone from my life, I always feel lighter. I feel refreshed and renewed. The arguments stop, and so do those long-drawn-out conversations where nothing gets resolved. Gone are those fake understandings we agree to just so the other person can shut the fuck up. It’s nice not feeling drained from having to talk to them or from complaining about them when they’re not around. It’s lovely not to commiserate. | https://lilibetovesen.medium.com/bill-maher-gave-me-the-best-advice-by-way-of-josef-stalin-sort-of-e1ec01bf435d | ['Elisabeth Ovesen'] | 2020-10-12 21:30:45.068000+00:00 | ['Self Improvement', 'Life Lessons', 'Inspiration', 'Mental Health', 'Advice'] |
Choices | Choices
21st — Writing Prompt
The decision weighed heavy on him.
“Take the blue pill and forget everything. Every bad memory gone…
Take the red pill and remember everything. Retain every bit of information, but… that includes bad memories.”
A woman’s voice, “Seriously Frank? Quit dinking around and take the damn Advil.”
“Tylenol it is!” | https://medium.com/the-friday-fix/choices-884cf19fd12c | ['Josh Petersen'] | 2020-08-21 11:01:01.112000+00:00 | ['Microfiction', 'Writing', 'Short Story', 'Fiction', 'Short Fiction'] |
Stop Wasting Your Life in the Pursuit of More and Different | Stop Wasting Your Life in the Pursuit of More and Different
The grass isn’t always greener on the other side, so let’s stop lying to ourselves
Photo by Pablo Varela on Unsplash
If I could break my whole sad life down to one word that would allow me to show you the folly of not being grateful for what you have in life, that word would be longing.
No matter where I was currently in my life, one thing was always the same. No matter what I did or did not have, or how much better off I was than my neighbor, I not only wanted more, but I also wanted different.
20 years old. Child on the way. Married for three years. With not even a high school diploma, I worked my way into a shift manager position in a major hamburger chain. I was getting the bills paid, sometimes with a little help from my dad. I had a car. I made more money and lived better than the majority of the people I knew. Was I happy? Nope. I spent every waking hour thinking about what it would be like to be with a different woman, win the lottery, drive a fancy car, travel, on and on ad nauseam. I was never happy and would never be.
35 years old. Three children. Big mortgage. Big second mortgage. Making over 100K per year. I hadn’t sat around the past 15 years dreaming about a different life. I went to college. I taught myself coding and web design. I landed increasingly higher paying jobs until I settled in Boston, and I finally had money, but I wasn’t happy. I spent all my time wishing for a different life, wife, job, house, face, body, and mind. I was a mentally ill burnout who spent all his waking hours stoned and still wishing for more and different.
52 years old. After a divorce and moving to the other side of the world, I finally have a few things I am grateful for. I have a wife who appears to love me unconditionally and a total of five kids — two being under 8-years-old. I’ve fought a bloody battle with mental illness and have come to somewhat of a truce at times. I can write consistently when I am not having psychotic episodes, and I am starting to build my reputation.
You would think I would finally be happy, right?
I still want more and different. Even though I have little ones at home, I want to travel. Even though I am working as hard as I can to be financially successful, it’s not happening fast enough. I daydream about what it will be like when I am making thousands instead of hundreds. I still dream of winning the lottery. I don’t think about more and different women because I am delighted with my wife, but money is almost like an obsession — a mistress that would fulfill my every need.
I have a do-over. I screwed up my first marriage and destroyed my relationship with my first three kids because I was always too sick and too focused on getting the things I longed for. I have a chance now to impact my children’s lives by being present and grateful, but all I can think about is working harder. I have a partner who craves my attention, and I’m too focused on giving her all the things she ever wanted.
I work myself until I am so tired that I can’t fight the depression and the voices that try to control my brain. When I’m not sick, I am feverish in my quest to do more. I feel like If I can do more, I could finally get all those things I have been dreaming about my entire life and be able to give everyone I love the things they want.
I’m always working.
And when I’m not working, I’m thinking of work. Even when I have the chance to spend quality time with my kids, I’m often distracted or staring at social media on my phone, trying to promote and engage with my audience.
If I can break through this ceiling and start making money, I can travel. I can have that new laptop I want. My wife can have the latest car she keeps talking about. I can start saving for my kid’s college. I can help my mom and dad and my three boys in the States.
We could finally move my whole family to the U.S. and allow my wife and kids to see another way of life. They only know life here in the Philippines, and I would love to take them for a $5 ‘Pizza! Pizza!’ in the land of the free and home of the brave.
So I work. And I work harder. And maybe I miss out on some things with my kids, and I don’t have the best relationship with my wife because I’m distant and distracted all the time.
If I could finally fill the longing for more and different, aren’t some sacrifices worth it? | https://medium.com/the-personal-essayist/stop-wasting-your-life-in-the-pursuit-of-more-and-different-ff99e8e5696a | ['Jason Weiland'] | 2020-11-15 18:33:49.328000+00:00 | ['Family', 'Mindfulness', 'Life', 'Mental Health', 'Life Lessons'] |
Istinomer fact-checked COVID-19 and the Serbian national election at the same time. Here’s how. | Istinomer fact-checked COVID-19 and the Serbian national election at the same time. Here’s how.
Fighting misinformation about the virus during the run-up to June’s boycotted election was an arduous task for this small Belgrade-based outlet. Now it’s thinking about how to use its readers as an early warning system.
Image: Istinomer
This case study is part of Resilience Reports, a series from the European Journalism Centre about how news organisations across Europe are adjusting their daily operations and business strategies as a result of the COVID-19 crisis. | https://medium.com/we-are-the-european-journalism-centre/holding-serbias-government-accountable-via-a-simple-browser-plugin-e7ffb3ed2e38 | ['Tara Kelly'] | 2020-09-07 09:49:24.972000+00:00 | ['Misinformation', 'Journalism', 'Media', 'Fact Checking', 'Case Study'] |
7 Historical Facts That Are Too Weird and Crazy to Believe! | There are many things in history that make sense. However, many historical incidents and stories seem insignificant and unknown when compared to major events such as the world wars. Those things are crucial parts of history, but they’ve got nothing on these wacky facts we dug up.
This is part of a series about historical facts, you can also read the following: | https://medium.com/history-of-yesterday/7-historical-facts-that-are-too-weird-and-crazy-to-believe-96d0b5df26d7 | ['Hossein Raspberry'] | 2020-11-20 15:58:17.876000+00:00 | ['History', 'Science', 'Short Story', 'Culture', 'Feminism'] |
Difference Between Data Science & Business Analytics | Data Science and Business Analytics, often used interchangeably, are very different domains. A layman would probably be least bothered with this interchangeability, but professionals need to use these terms correctly as the impact on the business is large and direct. In this article, we will elaborate on the difference between the two.
Overview
Data Science and Business Analytics are unique fields, with the biggest difference being the scope of the problems addressed. Simply put, The science of data that uses algorithms, statistics, and technology is known as Data Science. It provides actionable insights on a range of structured and unstructured data solving a broader perspective such as customer behavior.
On the other hand, the statistical study of mostly structured business data is known as Business Analytics. It provides solutions to specific business problems and roadblocks.
These two terms are interchangeably used in either of the above scenarios, i.e., a business analytics problem could be wrongly addressed to be solved with the help of Data Science. The implications of carelessly using the term ‘Data Science’ in this context could be adverse because the tools and techniques used in Business Analytics are different than Data Science and using wrong tools to assess a data set will yield imperfect and undesirable results.
Data Science is an umbrella term for all things dedicated to mining large data sets. An intersection of programming, statistics, and data analytics, Data Science is not limited to only statistical or algorithmic aspects. Business Analytics is the end-product of data science. It includes two broad categories, that are Statistical Analysis and Business Intelligence.
Business Intelligence
Another term often confused with Data Science is Business Intelligence. It is also an umbrella term that portrays ideas and strategies to improve decision making by utilizing fact-based support systems. Modern Business Intelligence is much beyond just business reporting. It is a mature framework that encompasses intuitive dashboards, mobile analytics, what-if planning, etc. It additionally incorporates enormous back-end machinery for maintaining control around reporting.
Although it sounds similar to Data Science, it is not. The principal difference lies in the type of problems that they address. Business Intelligence deduces the new unknown values of previously known elements using a formula that is already available. On the other hand, Data Science works with unknown scenarios without any formula or algorithm in hand, to solve data queries that nobody has ever answered in the past. Data Science problems are solved by exploring data, finding the best method, building a model around it, and finally operationalizing the model.
Conclusion
Business Intelligence is well established with deep roots in a typical corporate landscape. Corporate professionals are familiar, comfortable, and confident with the BI concepts and framework. As BI projects work on known unknowns, the projects can be planned well in advance and timelines could be efficiently followed. Also, there is minimal trial and error with several successful BI projects in a company’s kitty, who would have developed good project expertise over the years.
There is a massive career scope in the fields of Business Intelligence and Business Analytics. Professionals who are genuinely thinking of making a shift in the BA and BI roles can consider upskilling with the right course. Great Learning’s PG program in Business Analytics and Business Intelligence helps working professionals make a smooth and successful transition. The course offers the choice of online or classroom-based learning with Dual Certificate from the University of Texas at Austin, McCombs School of Business (world rank #2 in Analytics), and Great Lakes (India rank #1 in Analytics). It helps you with hands-on practical learning with case studies and projects, without the need of quitting your job. The course is also tailor-made keeping in mind the professionals from the non-IT background. With our career guidance and support, you can easily land your dream job in Business Intelligence and Business Analytics.
Originally published at https://www.greatlearning.in on August 28, 2019. | https://medium.com/my-great-learning/difference-between-data-science-and-business-analytics-8d7dc1f4ff15 | ['Great Learning'] | 2019-09-17 10:50:16.792000+00:00 | ['Data Science', 'Technology', 'Data Analysis', 'Science', 'Business'] |
The Physics of Superfluidity | Superfluidity is a property of some fluids to flow apparently without any viscosity (with constant kinetic energy). Examples of superfluids include helium-3 (or ³He) and helium-4 (or ⁴He). For temperatures below 2.17 K, helium-4 becomes a superfluid. Helium-3 becomes a superfluid only below 0.0025 K. Also, when superfluids are stirred, they form vortices that “rotate indefinitely” (see Fig. 1). The appearance of the superfluid phase in ⁴He, which is a boson, is related to Bose condensation, where a macroscopic fraction of the atoms occupies the lowest-energy state. The mechanism that generates the superfluid behavior in ³He is different since ³He is a fermion and the Pauli principle forbids more than one fermion to occupy the same state. Superfluidity occurs in ³He via the formation of pairs of fermions, which behave like bosons (Cooper pairs).
Figure 1: The dark spots are the cores of the vortices (source).
Superfluids also exhibit the so-called fountain effect: recipients containing superfluids empty themselves spontaneously. Fig. 2 shows a superfluid creeping up a cup’s wall as a thin film and coming down on the outside, forming a drop. After the drop falls into the liquid below, a new one will form. The process continues until the cup is empty.
Superfluidity was first discovered by the leading Soviet physicist and Nobel laureate Pyotr Kapitsa and the Canadian-born physicist John F. Allen. The mathematical theory of superfluidity was developed by renowned Soviet physicist and Nobelist Lev Landau.
Figure 2: On the left, we see the fountain effect: liquid helium in the superfluid phase slowly empties the cup spontaneously. It creeps up the bowl’s wall as a thin film then forms a drop outside which falls into the liquid below. This process continues until the cup is empty (source). On the right is a picture of Lev Landau, who developed the mathematical theory of superfluidity (source).
This article is based mostly on Zee and Lancaster and Blundell. I will follow both and use quantum field theory in the nonrelativistic limit (the analysis can be also done using Bogoliubov transformations) to derive the results.
Warm-Up: The Klein Gordon Field
In the Lagrangian formulation of classical mechanics, a particle is idealized as a point of mass m. Now, suppose its position at some time t is given by x(t) and that the particle moves in a region with potential energy is V(x). Newton’s second law of motion then reads:
Equation 1: Newton’s second law of motion.
The corresponding Lagrangian function L is:
Equation 2: The Lagrangian of a particle moving in a region where the potential is V(x).
where K and V are, respectively, the kinetic and potential energies. We can envision the passage from a point particle at x(t) to a field Φ(x, y, z, t) as the “replacement” of x by Φ, and of t by (x, y, z). The Lagrangian density has, in this case, the following form:
Equation 3: The Lagrangian density of the real scalar field theory Φ.
Note that, Φ is a real scalar field, not a wave function. Eq. 3 is Lorentz invariant (its mathematical form does not change for observers moving with respect to one another within an inertial frame). The scalar field Φ obeys the so-called Klein-Gordon equation:
Equation 4: The Klein-Gordon equation for the free field Φ.
Figure 3: The field Φ moving in spacetime (source).
Complex Scalar Fields
To describe the properties of superfluids we will need to add a layer of complexity to Eq. 3. The Lagrangian density of a complex scalar field is:
Equation 5: The Lagrangian density of the complex scalar field theory Φ.
This Lagrangian describes interacting bosons. We are specifically interested in the behavior of the slowly moving bosons. For simplicity, let us first find the dynamics of such bosons starting with the equation of motion Eq. 4 for a free scalar field Φ. The solutions of Eq. 4 are modes with the following time dependence:
Equation 6: Solutions of Eq. 4 are modes of the form
Since we are interested in the behavior of slowly moving bosons we will write the energy in Eq. 6 as follows:
Equation 7: The energy in Eq. 6, written as a sum with m >> ε.
We can separate the mode given by Eq. 6 into a product of two factors:
Equation 8: The mode given by Eq. 6 where φ oscillates much more slowly than the exponential factor (since m>>ε).
Using the approximation:
Eq. 4 becomes the Schrödinger equation:
Equation 9: Schrödinger’s equation.
The nonrelativistic version of the Lagrangian density is obtained by plugging
into Eq. 5 (the prefactor is just for later convenience). After some simple algebra we get:
Equation 10: Lagrangian density of nonrelativistic bosons interacting with a short-ranged repulsion.
From the minus sign in the potential term g²ρ², we see that the interaction is repulsive. To avoid the possibility of having zero density in Eq. 10, one usually introduces in L a finite density of nonrelativistic bosons:
Equation 11: Eq. 10 with the addition of a chemical potential term.
The term:
Equation 12: Mexican hat or champagne-bottle potential shown in Fig. 4
is the Mexican hat or champagne-bottle potential shown in Fig. 4.
Figure 4: The Mexican hat or champagne-bottle potential (source).
A consequence of the form of Eq. 11 is that φ approaches the expectation value of the field φ:
Equation 13: The Mexican hat potential will force Eq. 12 to be small.
To study spontaneous symmetry breaking we first write φ in polar coordinates:
Equation 14: The field ϕ written in polar coordinates.
then add a small perturbation to the expectation value:
Equation 15: A small perturbation is added to the expectation value of φ.
Substituting Eq. 15 in Eq. 11 we obtain:
Equation 16: The Lagrangian density written in terms of polar coordinates.
where a total divergence term was dropped. Now, to eliminate h we plug the Lagrangian density into a path integral and integrate it out:
Equation 17: Integrating out h via a path integral (which is omitted).
The Lagrangian density becomes:
Equation 18: The Lagrangian density of a fluid of bosons with a gapless mode.
where the spatial derivative in the denominator of the second-hand side of Eq. 17 was dropped. Eq. 15 tells us that there exists a fluid of bosons with gapless mode. The linear dispersion relation (see Fig. 5 below) is given by:
Equation 19: The linear dispersion relation of the bosons in a superfluid.
The form of the Lagrangian obtained after spontaneous symmetry breaking gives us the momentum superflow. The massless bosonic field θ is called a Goldstone boson. The presence of Goldstone bosons is a general consequence of the spontaneous symmetry breaking (in Eq. 15).
In Fig. 5 we see the dispersion relation of superfluid liquid obtained experimentally. When the momentum is low, the dispersion relation is indeed linear and at large momentum, the energy goes as p² as we expect for free particles. | https://medium.com/cantors-paradise/the-physics-of-superfluidity-ca42ea1b72f3 | ['Marco Tavora Ph.D.'] | 2020-12-09 19:31:12.613000+00:00 | ['Science', 'Fluid Mechanics', 'Math', 'Physics'] |
What I Learned About Life, Love & Sex From My Great-Grandfather | What I Learned About Life, Love & Sex From My Great-Grandfather
Two writers, a century apart, trying to understand the workings of the human heart
The Wolfes, late 1930s (Photo from Yael Wolfe’s family collection)
My grandfather died in 1994, when I was 18 years old. I was obsessed with genealogy at the time and was thrilled when I discovered Grandpa’s boxes of old photos, family journals, and genealogical notes.
One of the most fascinating finds was his father’s — my great-grandfather’s — journal. It was filled with musings on society, economics, social observations, and notes about love…and sex. Yes, sex.
Some of it, I could not read. My great-grandfather brought his family to America around 1920, from Denmark. He, his wife, my grandfather, and my grandaunts and -uncles were fluent in many languages — Danish, Norwegian, English, French, German, Latin — and they often wrote or spoke in multiple tongues.
What Great-grandpa wrote in English, though, I read over and over. Oddly, I wasn’t able to grasp a better understanding of his personality by reading through his journal. My opinion of him vacillated quite violently. Was he a sexist, privileged, wealthy, overly-educated man who treated women like a commodity? Was he a curious philosopher whose sexism was simply a product of his time, his culture?
He died long before I was born and my grandfather never spoke of him. So who was this man who felt as compelled as I did to scribble away about sex and love in his notebooks?
Great-grandpa Anton (Photo from Yael Wolfe’s family collection)
His name was Anton. He was born in Copenhagen in 1873, one hundred three years before my own birth. He married a young woman from Norway, named Louise. Her family was in the business of making copper pots and kettles.
Together, they had four children — two girls, then two boys, just like my family.
We know they lived a good life in Denmark, but after their eldest child died at the age of 21, they made the decision to come to America with the remaining three.
I know all the facts, but I want to know more. What was their life like? Were they happy? What went on behind closed doors?
A Good Wife
I have a fascination with love stories. Not for the typical reasons. I’m not trying to look for confirmation that “true love is real” or couples can “make it.” I’m interested in how humans express love, what love looks like in different moments in history, how humans choose to define love and relationships.
So of course, I would love to know how my great-grandparents felt about one another. Were they in love? Did their well-to-do parents orchestrate the marriage for financial or social reasons? How did they fare in their long marriage?
Great-grandpa Anton appeared to want a sensible, good wife — such a cliché, I know, but to be fair, it was a very different time. In one of his journals, he wrote:
One of the funniest things about the cult of modern beauty is that while it is encouraged on the theory that it will hold husbands, its devotees lose theirs with alarming regularity. Beauty culture for hobbling straying husbands is largely a failure. A lot of women who depend on it now suffer from diminishing alimony returns as well as from plucked eyebrows. If they had coaxed along their husbands half as hard as they did their complexions, they could have had a comfortable fireside seat cinched forever. But they were so busy enhancing their charms that they had no time to practice the arts of the good wife.
The “cult of modern beauty” is still alive and well and I love to write about it just as much as Great-grandpa…but not for the same reasons. I think the cult of modern beauty is just as dangerous as the cult of the good wife. (Sorry, Great-grandpa.)
Louise & Anton, 1921 (Photo from Yael Wolfe’s family collection)
The Hunter
The older I get, the more open-minded I become about relationships. I grew up around monogamous couples, and yet rarely saw that model work in the long run. Are affairs inevitable in this structure? Or does it stem from multiple factors?
My great-grandfather also appeared to be fascinated with the topic of infidelity, though he came at it from a very different perspective. Men, he surmised, were hunters, and collecting women was their natural instinct.
In his essay called The Second Woman (in which I have made appropriate grammatical corrections), he wrote:
Into the life of every man…comes the second woman, and the minute she appears, sense and reason go flying out of the window. Why will any sane man who truly loves his wife, cherishes his family, recognizes the worth of all the things he has spent years in building, chance the destruction of all for a few hours in the arms of the second woman? Why? Because he has no choice in the matter… From Adam on, every man has been a hunter. He wins the first great hunt when he finds the woman whom he makes his wife… Then the second woman makes her appearance. It may be her sexual lure which first attracts him — it may be an intellectual attraction, pleasantly natural. No sooner does the second woman appear than visions of a new hunt creep back into his mind… And he deceives himself that the affair is entirely innocent, so he is inclined to deceive the woman — either he poses as an unmarried man or he tells her his wife does not understand him, and he needs her sympathy, her aid, in solving his problems… Nine times out of ten, the man still loves his wife more than the second woman… No man can be as happy with the second woman as with the first, with very few exceptions…
How interesting that he speaks with such authority on this subject, hmmm? Great-grandpa, did you have a secret or two?
Interestingly, he doesn’t place any blame for this deception in the hands of the perpetrator, but calls it the fault of civilization. But I try to be forgiving, since he wrote this at a time when the cultural mores around marriage were, I imagine, suffocating for all genders. I’d guess many couples accepted their fate: men being denied in the bedroom, women playing the part of the dutiful, virtuous wife. That was what people were supposed to do, what marriage was supposed to look like, having absolutely nothing to do with what our hearts, souls, and bodies actually want.
That, Great-grandpa, can be blamed upon civilization.
Sexual Frustration, Circa 1920
One of my favorite subjects is sexual frustration and the cultural factors that contribute to it. This is a sensitive subject and I’ve found that many men can be quite reactive when this comes up. I’ve heard so much bitterness around this subject, but bitterness aimed at women, and bitterness that births false information about female sexuality that in turn creates even more bitterness. And beneath it all is this strange sense of entitlement — as though some men feel they deserve to have their every sexual need met by their female partner.
I see something similar in my great-grandfather’s musing on this subject:
There is no evidence to prove that woman has any right to assume herself the possessor of any man during a prostrated period. Man’s taste changes and what for him was the sublime in his young life may not fulfill his desires in later life. Man has a hunting nature or instinct. For this reason, it is quite natural that he is always on the hunt for new game, for new diversification, or to be frank about it, for new meat. Woman wants to be conquered and is forever teasing and egging the man to begin the preliminaries for the conquest. Although in order to save the face of her chastity, she pretends to be suavely defending her so-called honor. She generally plays the game so long as she can without cooling his ardor and when he is on the verge to lay off, she submits and gives his rigid generant approved admission to the voluptuous receptacle in which the seed for a new individual is to be fertilized.
Well, shit. Great-grandpa, this is fucked up.
What’s interesting to me, though, is how often I still see attitudes like this. One hundred years later, there’s still so much enmity between the sexes — frustration, blame. The myth of the hunter and the teasing prey. The conquest, the “face” of chastity, the “so-called honor”… these are all constructs of the patriarchy, of puritanical religious beliefs, of sexism, of misogyny. We can see that today, but he would not have known that back then.
Anton’s journal (Photo from Yael Wolfe’s family collection)
Life, love, and sex
I’m fascinated by the fact that my great-grandfather used his writing to understand the world better, just as I do. Sure, he might have been a sexist dick who had a whole lot of “meat” (to use his words) on the side, but to be fair, we are all formed by our cultures. Sometimes, in his writing, I can see his inner struggle, trying to do what he’s “supposed” to do — be the good, virtuous husband who loves his good, virtuous wife. And when faced with the emotional and physical denial that I assume accompanied that role, his anger and frustration would eventually bubble over, searching desperately for a “voluptuous receptacle.”
Here we are, a century later, still dealing with this issue. Some women are still struggling so much to emerge from the sexual shame our culture used to bind us, to discard the harmful role of “good girl,” “good wife.” Some men are trying so hard to be good husbands who love their good wives, until their blue balls (or lonely hearts) send them into a momentary fit of anger or frustration and the structure comes tumbling down. And so many people (not just men), when they just cannot find a way through this, seek out someone else. We find a way to get our needs met elsewhere.
Great-grandpa Anton didn’t have any solutions to this, nor do I. But unlike him, I have the benefit of one hundred years of social progress. Admittedly, we didn’t get very far, but at least we’re talking about it now. At least there is some awareness of the problems.
Sexist or not, I’m grateful to have my great-grandfather’s words to look back on, to ponder, to explore. There is power in writing. Just like Great-grandpa Anton, I’m only scribbling away, sharing my theories, but these written words, like his, might live another century and beyond, touching the lives of people, of descendants, I will never meet.
My great-grandnieces and -nephews might look back on all my writing and shake their head, thinking, as I do about my great-grandfather, “Damn, that lady was so unenlightened…but at least she wrote, and shared, and talked…and tried.”
In the end, I think that’s all we can do — share, talk, try.
And write.
© Yael Wolfe 2019 | https://medium.com/wilder-with-yael-wolfe/what-i-learned-about-life-love-sex-from-my-great-grandfather-40287a293c8 | ['Yael Wolfe'] | 2020-08-21 04:21:32.390000+00:00 | ['Writing', 'Relationships', 'Sex', 'Love', 'Feminism'] |
Neural Networks with Memory | DEEP LEARNING
Neural Networks with Memory
Understanding RNN, LSTM under 5 minutes
We always heard that Neural Networks (NNs)are inspired by biological neural networks. This huge representation was done in a fantastic way.
Figure 1 shows the anatomy of a single neuron. The central part is called the cell body where the nucleus resides. There are various wires which pass the stimulus to the cell body and few wires which send the output to the other neurons. The thickness of the dendrites implies the weight/bias/power of the stimulus. Many neurons with various cell bodies are stacked up which forms the biological neural network.
Figure 1: Anatomy of Single Neuron ( Source, Edited by author)
This same structure is implemented in Neural Networks. The input is passed through an activation function with weighted edges. The output is generated which can be passed to another activation function. Many activation functions can be stacked up, each of these is called a layer. And in a layer, we can have multiple neurons.
Figure 2: Single neuron neural network (Image created by author)
These activation functions can be as simple as a sigmoid function
Advantages of neural networks over traditional machine learning algorithms
Various types and sizes of data can be handled
Multiple functions can be easily configured
Non-linear data can be efficiently handled
Neural Networks with memory
The main difference between the functioning of neural networks and the biological neural network is memory. While both the human brain and neural networks have the ability to read and write from the memory available, the brain can create/store the memory as well. Researchers identified that this key difference is the major roadblock for today’s AI systems to reach human-level intelligence.
Researchers at DeepMind aimed to build a differentiable computer, by putting together a neural network and linking it to external memory. The neural network would act as a CPU with a memory attached. Such differentiable computers aim to learn programs (algorithms) from input and output data.
The neural networks are used when the amount of data is huge. For example, text data has an enormous amount of dimensions or the image data which is split into a huge number of pixels.
Recurrent Neural Network
A movie consists of a sequence of scenes. When we watch a particular scene, we don’t try to understand it in isolation, but rather in connection with previous scenes. In a similar fashion, a machine learning model has to understand the text by utilizing already-learned text, just like in a human neural network.
In traditional machine learning models, we cannot store a model’s previous stages. However, Recurrent Neural Networks (commonly called RNN) can do this for us. Let’s take a closer look at RNNs below.
Figure 3: Working of a basic RNN (Image by Author)
An RNN has a repeating module that takes input from the previous stage and gives its output as input to the next stage. However, in RNNs we can only retain information from the most recent stage. That’s where LSTM comes to the picture.
Long Short Term Memory Networks
To learn long-term dependencies, our network needs memorization power. LSTMs are a special case of RNNs which can do that. They have the same chain-like structure as RNNs, but with a different repeating module structure.
Figure 4: Working of LSTM (Image by author)
LSTM has a wide range of applications in Sequence-to-Sequence modeling tasks like Speech Recognition, Text Summarization, Video Classification, and so on.
To understand how these networks can be adopted in real-life applications in a quick glance, do check the article below.
A spam detection model can be achieved by converting text data into vectors, creating an LSTM model, and fitting the model with the vectors.
Implementation
To implement these complete neural networks, Keras and TensorFlow made it simple.
We are implementing a Bidirectional LSTM with the ReLU activation function.
#Importing necessary libraries
import tensorflow as tf
from keras.layers import LSTM, Activation, Bidirectional #Addding Bi-directional LSTM
model.add(Bidirectional(tf.keras.layers.LSTM(64))) #Relu allows converging quickly and allows backpropagation
model.add(Dense(16, activation='relu')) model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
Creating each layer is just a single line of code.
Summary
In this article, we learned how neural networks are linked to biological neural networks and the working of neural networks with memory (namely, RNN, LSTM).
Thanks for the read. I am going to write more beginner-friendly posts in the future too. Follow me up on Medium to be informed about them. I welcome feedback and can be reached out on Twitter ramya_vidiyala and LinkedIn RamyaVidiyala. Happy learning! | https://towardsdatascience.com/neural-networks-with-memory-27528a242b78 | ['Ramya Vidiyala'] | 2020-09-24 13:47:06.960000+00:00 | ['Machine Learning', 'Artificial Intelligence', 'Technology', 'Education', 'Data Science'] |
Education adjusting to COVID-19 Crisis and teaching Children AI | ARTIFICIAL INTELLIGENCE
Education adjusting to COVID-19 Crisis and teaching Children AI
Scientific Discovery using AI and UK’s role in development of AI
Everybody is talking about COVID-19 and as the crisis continues, education is adjusting to the new conditions through online programs. Zoom, Slack and Microsoft Teams are enabling this transition as people take conversations online in accordance with social distancing.
There is optimism about the pandemic subsiding as countries flatten the curve according to WHO. Artificial intelligence adoption is accelerating and COVID-19 outbreak could make the technology widespread as people experience its benefits in detecting infections and predicting the next pandemic.
The United Kingdom has played a role in the development of AI since the days of Alan Turing and the country continues to experience growth of tech start-ups in AI, block chain and IOT. Artificial intelligence is growing with scientific discoveries relying on AI for research, AI powered calling services and real-time noise suppression.
Mark Cuban believes that teaching children AI will educate them about the future of technology and its implications on business. Children need lessons on AI and exposing them according to Mark Cuban will reshape our future as new technologies such as machine learning become part of our lives.
Creating Successful AI Projects
When it comes to being competitive in today’s markets artificial intelligence (AI) is clearly a must-have weapon. But even for some of the world’s top companies implementing this technology has been challenging. Creating models, generating sufficient ROI, getting result-driven data, and finding the right talent have been the common issues faced by the companies.
Many AI projects fail because of it. Only about 35% of organizations succeed in getting models into production successfully- According to IDC.
Santiago Giraldo, Senior Product Marketing Manager of Data Engineering at Cloudera said, “Unlike traditional data analytics, machine learning (ML) models that power AI is not always going to offer clear-cut answers,” It is important to understand that every experiment is not going to drive ROI. A successful AI project¹ is built on top of many failed data science experiments. The approach of taking a portfolio to ML and AI enables greater longevity in projects.
Gus Walker, Senior Director of Product Management at Veritone. said “Often times businesses take on AI projects not realizing that it might have been cheaper to continue a process manually instead of investing large amounts of time and money into building a system that doesn’t save the company time or money,”
How AI is making the world a safer place
AI is making a difference in improving public health and safety as the world adapts to a new normal.
The biggest challenge with this coronavirus has been predicting how quickly it can spread. Social distancing measures and the closure of high-risk facilities are viewed as the best way to control the spread, but many areas have been slow to enact such measures because they don’t have an accurate perception of their risk.
An AI powered survey system developed by the Weizmann Institute of Science in Israel aims to better predict outbreaks² so authorities can proactively enact measures that will mitigate the virus’s spread. The system uses a questionnaire focusing on key issues like isolation practices and health symptoms then match responses with a location-based algorithm. Moreover, AI analysis can identify potential hotspots in advance to help local authorities enact measures that will slow down the virus.
Hospitals and health organizations are getting more inquiries than ever from patients worried that they might have the coronavirus. AI tools specifically designed to address questions related to COVID19 are being introduced into healthcare apps and websites to adapt to the increased inquiries.
Scientific Discovery using AI
Steve Jobs described personal computing as a “bicycle for the mind.” He thinks that computers can be used as “intelligence amplifiers” that offer an important boost for human creativity and are now being given an immediate test in the face of the coronavirus.
The National Library of Medicine and a group of artificial intelligence research groups announced that they had organized the world’s scientific research papers about Covid19 and the documents include more than 44,000 articles, could be explored in new ways using a machine learning program designed to help scientists see patterns and find relationships to assist research.
Oren Etzioni, the chief executive of the Allen Institute for Artificial Intelligence, “This is a chance for artificial intelligence.” Allen Institute is a nonprofit research laboratory that was founded in 2014 by Paul Allen, the Microsoft co-founder.
There has long been a dream of using AI to help with scientific discovery³. The new advances in software applications raise questions about whether computer technologies such as artificial intelligence will enhance or even begin to substitute for human creativity. A rapid set of advances based on new language process techniques leading a variety of technology firms and research groups.
AI Powered Calling Service
Duplex, artificial intelligence-powered calling service automated by Google is now available in Australia, Canada, and the UK which was earlier available only in the US and New Zealand.
Google clarified to The Verge that It isn’t a full rollout of the service, and Google using Duplex mainly to reach businesses in those new countries to update business hours for Google Maps and Search.
CEO Sundar Pichai wrote in a blog post “In the coming days, we’ll make it possible for businesses to easily mark themselves as ‘temporarily closed’ using Google My Business. We’re also using our artificial intelligence (AI) technology Duplex where possible to contact businesses to confirm their updated business hours, so we can reflect them accurately when people are looking on Search and Maps.”
Duplex launched as an early beta via Google Assistant in 2018 in the US after a splashy yet controversial debut at that year’s Google I/O developer conference. There were concerns about the use of Duplex regarding proper disclosure that the automated call was being handled by a digital voice assistant⁴ and not a human being.
Google tried to address those concerns by adding disclosures at the beginning of calls and giving businesses the option to opt-out of being recording and speak with a human.
Education adjusting to COVID-19 Crisis
Learning for grades K-12 looks very different than a month ago in light of the recent events surrounding COVID19. Turning homes into classrooms may be overwhelming for parents & educators.
A team led by Media Lab Associate Professor Cynthia Breazeal has launched aieducation.mit.edu keeping that in mind to share a variety of online activities for K-12 students to learn about artificial intelligence, with a focus on how to design and use it responsibly.
This website provides learning resources to address the needs of the millions of children, parents, and educators worldwide who are staying at home due to school closures caused by COVID-19 and are looking for free educational activities that support project-based STEM learning⁵ in an exciting and innovative area.
MIT Open Learning and the Media Lab, MIT Stephen A. Schwarzman College of Computing collaborated the website, serving as a hub to highlight diverse work by faculty, staff, and students across the MIT community at the intersection of AI, learning, and education.
MIT has revolutionized the way children can learn computational thinking with successful platforms such as Scratch and AppInventor.
MIT professor of computer science & engineering Hal Abelson says “MIT has been a world leader in AI since the 1960s.”
Teaching Children AI
Mark Cuban tweeted “Give your kids an edge, have them sign up and learn the basics of Artificial Intelligence.” He also said, “The way to set your children up for success in this day and age is to ensure they learn about artificial intelligence.”
He is a star on the hit ABC show “#SharkTank” and the owner of the Dallas Mavericks NBA basketball team. He was promoting a free, one-hour virtual class his foundation is teaching an introduction to artificial intelligence in collaboration with AIForAnyone, a nonprofit organization that aims to improve the literacy of AI⁶ understanding.
He promoted the importance of learning and understanding artificial intelligence using his megaphone. He also talked about how important it is for business owners to understand AI at the South by Southwest conference in Austin, Texas, in March 2019.
Cuban told Recode’s Peter Kafka “As big as PCs were an impact, as big as the internet was, AI is just going to dwarf it. And if you don’t understand it, you’re going to fall behind. Particularly if you run a business,” He also told Kafka. “I mean, I get it on Amazon and Microsoft and Google, and I run their tutorials. If you go to my bathroom, there’s a book, ‘Machine Learning for Idiots.’ Whenever I get a break, I’m reading it,”
Real-time Noise Suppression
Microsoft announced last month that Teams, its competitor to @Slack, Facebook’s Workplace, and Google’s Hangouts Chat, had passed 44 million daily active users. The milestone overshadowed its unveiling of a few new features coming later this year.
A hand-raising feature to indicate you have something to say, offline and low-bandwidth support to read chat messages and write responses even if you have poor or no internet connection, and an option to pop chats out into a separate window among the most straightforward features. But one feature stood out, real-time noise suppression Microsoft demoed how the AI minimized distracting background noise during a call.
Thousand of times we have you asked someone to mute themselves or to relocate from a noisy area. Real-time noise suppression⁷ will filter out someone typing on their keyboard while in a meeting, the rustling of a bag of chips, and a vacuum cleaner running in the background. AI will remove the background noise in real-time to ensure you hear only speech on the call.
The coronavirus crisis forces the use of collaboration and video conferencing tools are exploding as millions to learn and work from home. Microsoft is leaning on its machine learning expertise to ensure AI features are one of its big differentiators.
Training Models and AI Learning
Amazon researchers propose an AI approach that greatly improves performance on certain meta-learning tasks (i.e., tasks that involve both accomplishing related goals and learning how to learn to perform them) in a paper scheduled to be presented at the upcoming International Conference on Learning Representations. It can be adapted to new tasks with only a handful of labeled training examples, meaning a large corporation could use it to, for example, extract charts and captions from scanned paperwork.
A model trains on a set of labeled data (a support set) and learns to correlate features with the labels in conventional machine learning. It’s then fed a separate set of test data (a query set) and evaluated based on how well it predicts that set’s labels. An AI model learns⁸ to perform tasks with their own sets of training data and test data and the model sees both by contrast, during meta-learning. AI learns how particular ways of responding to the training data affect performance on the testdata in this way.
The model is trained on tasks that are related but not identical to the tasks it saw during meta learning in a second stage called meta testing. The model once again sees both training and test data, but the labels are unknown and must be predicted for each task.
COVID-19 accelerating adoption of AI
Cloud computing kicked into high gear and started to become a pervasive, transformational technology after the financial crisis of 2008. AI applications could be central during the current COVID-19 crisis. Though the implications of AI continue to be debated on the world stage, the rapid onset of a global health crisis and concomitant recession will accelerate its impact.
AI technologies help to discover new drugs — either vaccine or treatment — have kicked into hyperdrive. Companies are forming partnerships with academia to find a cure and startups are racing to find solutions as well. Companies are also researching existing drugs to identify their potential applicability.
AI is possibly saving years of research proving a useful tool for dramatically reducing the time needed to identify potential drug candidates. AI is already screening for COVID19⁹ symptoms, automating hospital operations, and supporting the decision for CTscans. Robots have started to perform a variety of healthcare functions.
AI will be an even larger part not only for healthcare but also for other technology landscape going forward. “Americans are growing more comfortable shopping for food or electronics without the aid of another human,” according to the March 2020 Automated Retail Tracker.
Secure computation of AI models
A team of researchers from Princeton, Microsoft, the nonprofit Algorand Foundation, and Technion proposes Falcon, an end-to-end framework for secure computation of AI models¹⁰ on distributed systems in an academic paper published this week on the preprint server Arxiv.org.
It’s the first secure C++ framework to support high-capacity AI models and batch normalization, a technique for improving both the speed and stability of models as per their claim. Falcon can outperform existing solutions by up to a factor of 200 and it automatically aborts when it detects the presence of malicious attackers.
Falcon could be a step toward a pipeline tailored to domains where privacy and security are table stakes, such as healthcare. Running machine learning models in a privacy-preserving fashion without computational trade-offs remains an unsolved challenge despite the emergence of techniques like federated learning and homomorphic encryption.
Data holders who own the training data sets, and query users who query the system post-learning in a distributed AI usage scenario as per Falcon’s assumption. A machine learning model of interest is trained on data from the data holders and queried by the query users, such that the data holders share their data securely between servers.
UK’s role in development of AI
An important role played by the UK in the history and development of AI. The greatest British mathematician, Alan Turing is considered to be the father of theoretical computer science and has deep roots in AI as well. Turing envisioned the Turing test to determine a machine’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human to crafting the foundations for modern computing.
The UK was not only heavily involved in AI development from the very first years, but also helped bring about the first AI Winter¹¹ in the industry as well. The UK has been making heavy investments in AI and continues to show its strength in the field. “The United Kingdom has one of the strongest AI strategies in the world with strong government funding for AI, strong research activity in the field, strong VC funding and AI startups, and strong enterprise activity and adoption of AI”-According to a recent report by research firm Cognilytica.
The UK established an All-Party Parliamentary Group on AI to address ethical issues, industrial norms, regulatory options and social impact for AI in Parliament in 2017. We have to change with AI to ensure the future of the use of this technology.
Works Cited:
¹Building AI Projects, ²AI predicting COVID-19 Outbreak, ³Scientific Discovery, ⁴Digital Voice Assistant, ⁵STEM Learning, ⁶AI Literacy, ⁷Real-Time Noise Suppression, ⁸AI Learning, ⁹AI Screening COVID-19, ¹⁰Secure Computation of AI Models, ¹¹AI Winter
More from David Yakobovitch:
Listen to the HumAIn Podcast | Subscribe to my newsletter | https://medium.com/advantages/education-adjusting-to-covid-19-crisis-and-teaching-children-ai-908fd210b618 | ['David Yakobovitch'] | 2020-05-17 13:31:57.186000+00:00 | ['Machine Learning', 'Data Science', 'Education', 'Artificial Intelligence', 'Covid 19'] |
Smart Cities Incorporation of Technology For Waste Management | Smart Cities Incorporation of Technology For Waste Management
Introduction
By way of definition, a smart city is a city that utilizes technology and data for the creation of more efficient results, improvement of sustainability in all sectors, fostering a wide-ranging economic development and contribution to the enhancement of the living conditions of those residing and working in that particular city. The utilization of technology to achieve these purposes covers a wide range of sectors, ranging from the health sector to the education sector and even the waste management in the cities.
Technology has immensely contributed to the reduction in cost and complexity of current waste management systems thereby improving their efficiency, safety, and productivity while also reducing the impact they have on the environment. Thus, this article aims to analyze the different ways by which technology can be incorporated, and is currently being incorporated, into the waste management system of smart cities to achieve better results.
www.swachhcoin.com
Technology For Waste Management
The complexity surrounding waste management has become more pronounced than before. More tons of trash is produced by citizens around the world and this has led to a major challenge which is how to successfully dispose of waste without opening individuals to major health issues. Specifically, the waste collection which is a component of waste management is an essential service for any city and the current system that exists can be greatly improved by using smart technology to foster increased efficiency and cause an improvement in the quality of waste collection processes.
In a bid to handle these challenges and tackle the adverse effects of landfill waste, among other dangerous waste products on the environment, waste management industries globally are rethinking the different strategies. Some of these strategies include using technology for operational analytics and route optimization as these can considerably lead to achieving sustainability goals in this wise and reducing the operational costs.
In addition, they are also changing how they handle waste and its disposal so as to achieve increased efficiency and enhancement of environmental protection.
Thus, some of the major areas where technological tools are being utilized for those purposes include
1. Automated Waste Collection
With the introduction of automated sensors that function to trigger automatic alerts each time a container is filled up and needs to be serviced, technology has impacted the workings of waste management. In fact, some other tools such as advanced disk screens and optical sorters have been introduced to make the sorting process easier and faster.
In addition, the trucks used in the waste collection now run on natural gas and not diesel for operations which are quieter and more cost-effective. Further, logistics software, mobile apps and in-vehicle monitors have made the waste management process simpler while also ensuring that the driver is safe.
2. Improved Rates of Recycling
Currently, waste management companies and also recycling companies are working on improving their techniques and tools. There is a recent development available in single-stream recycling which allows people to dump all their trash in one bin. That has helped to reduce people’s burden of sorting and has also impacted the recycling rate. In addition, it has contributed to the reduction of truck count as well as the emissions.
3. Route Optimization
Route optimization is important in the protection of the environment and the reduction of hazardous emissions and accounts for why companies are going ahead to invest in advanced systems as well as software dedicated to optimization. These companies are now using automated trucks which have already installed robotic arms that function to save effort and time along normal routes. It has made pickups from one point to another quite eco-friendly. The technology is also quite viable financially and also improved energy efficiency.
4. Modernization of Landfill
Through a harness of the power of technology, the waste management industry has succeeded in modernizing garbage dumps. The systematically engineered landfills that adequately comply with relevant regulations ensure a complete protection of health and preservation of the environment. Also, the use of solar panel systems which are integrated together with geomembrane are providing essential backings for production of energy deemed sustainable and also preventing carbon from finding its way back into the environment.
5. Faster Turnaround Periods
Some waste management companies have also begun investing in technology that provides better customer experience to their customers. These companies are investing in the creation of mobile apps that are user-friendly so they can promote quicker delivery of services and even bill payments via push notifications.
6. Conversion of Waste to Energy
This is one of the core ways technology is being used to transform the operations of waste management companies in smart countries. Rather than leaving particular types of waste in the landfill, they can be converted to energy. There are new innovative machinery called digesters for this purpose. Digesters can convert the waste and the produced biogas into energy that can be effectively used onsite. It can be used on different categories of waste such as agricultural leftovers, food, animal waste, among others.
Another innovative technology used for conversion of waste is called thermal conversion. This helps to convert waste into specific products. By using cues gotten from natural geothermal processes that function using heat and pressure to convert useless materials to products that have relevant usages.
It can be used to turn waste materials into fertilizers, chemicals, oils among other important converted products. Better still, certain varieties if landfill gas can also be converted to energy. The relevant gas in this wise will be those flared or released that can be converted into energy by using certain new innovative technology. Examples of such technology are microturbine technology, bioreactors, and fuel cells.
Swachh Bins — SwBins
SwBin is an innovative device of the Swachhcoin Platform that can autonomously sustain the Ecosystem while also retaining its core fundamental governing and working principles. This tool has some of the most revolutionary features not present in a lot of traditional waste management systems. Some of these features include:
1. Dapp Support
2. Enabled IoT support
3. Automatic opening and closing lid
4. LED Advertisement Panels
5. Waste Sorting at source
6. Waste compression
7. Unique User Identification
All these innovative features of the SwBin help to foster better waste management activities and is therefore ideal for all smart cities across the world. | https://medium.com/hackernoon/smart-cities-incorporation-of-technology-for-waste-management-a8ebf22f5 | [] | 2018-09-06 18:05:25.077000+00:00 | ['Blockchain Startup', 'Waste Management', 'Circulareconomy', 'Environment', 'Cryptocurrency News'] |
Everyone Has a Dream of What They Want Their Life to be Like | When I was a kid, I wanted to be a teacher and a writer.
I don’t remember why I wanted to be a teacher, but I do remember that there were times I thought about being one.
Now as for being a writer I remember that class time journaling in school was one of my favorite times. At least in elementary school. The stories I made up in my notebook we’re long and went on for pages.
Then in middle school, I developed a love for math. I began to get straight A’s and B’s for the rest of my education. At least in math, I did. Even when I went to college in my 30’s it was for accounting.
By then, my dream was to work in an accounting office. The problem with that was that yes, I had a degree but I didn’t have 2 or more years of experience in an accounting office.
But I have decided that I am ok with that. I have since realized that although I love math and accounting, this is not my ultimate dream in life.
My ultimate dream, my one true passion in life, is my writing. Sometimes I am inspired to write a poem. And sometimes I find a cool writing prompt that inspires me to write a short story. Then there are all these nonfiction book ideas I have swimming around in my head that I want to write.
So what have I done about that? So far I wrote and published on Amazon, 3 books:
Financial Management 101
This is a book about budgeting and managing your money and getting out of debt.
35 easy healthy delicious smoothies to lose weight
This book teaches you how to make smoothies and describes the different types. It also tells how to incorporate them into your diet, and what equipment you need to make them. It also includes 35 recipes to get you started.
Be an entrepreneur
Are you tired of working to make someone else rich while you are still barely getting by? Have you ever wanted to start your own business, to be your boss? If you do but don’t know where to start or how to do it then this is the book for you. In it, I will talk about different businesses that you could choose from. You’ll find out what scams to stay away from. The pros and cons of starting your business. How to set up your own home office and a lot more.
W.M. Housouer
Then I have 2 more books that I have been working on.
One on email marketing
I have started the 2nd draft
One of writing fiction books.
I have all the research completed and am about to start the 1st draft.
Then in the middle of October, I discovered medium. I downloaded the app on my phone, became a paying member, then signed up for the paid partners program.
So far I wrote and published 61 articles and have another 14 planned out and started. This does not count this one, and one written and planned for publication on February 1st, 2020.
This is my dream, my one true passion. I want to be a full-time successful writer. I want to get to the point where I can quit my full-time job and stay home to write full time.
I want to be able to make more than enough money to pay the bills and spend the rest of my life doing what I love doing.
Eventually, I want to have the following income streams bringing
in money every month.
My books on Amazon
My articles on medium
My articles on Vocal
Email Marketing w/affiliates
Email courses that I create
A membership community on Patreon.
A travel blog on Substack.
These are my goals, these are my dreams. My family doesn’t take me seriously about these dreams. They don’t think I should waste my time. They have no faith in me when it comes to making a career in writing.
My dad retired from a kitchen cabinet factory. He has said all my life that
The only real job is a factory job.
I was never good at factory work. I always believed in the saying
Work smarter, not harder!
This is not to say I am lazy and don’t want to work. I want to be able to do what I love and make a good living with it.
I hope that by my 46th birthday I will have quit my day job and am a full-time writer. That gives me a year and a half to do this. I know that I can do it if I keep working at it.
To all you who have a similar dream, I believe in you. I know you can do it if you fight and work hard enough at it. Never give up and write every day.
So now I say: | https://medium.com/this-shall-be-our-story/everyone-has-a-dream-of-what-they-want-their-life-to-be-like-d05e6bdd6812 | ['Wendy Housouer'] | 2020-01-21 12:46:01.088000+00:00 | ['Blogger', 'Blogging', 'Dreams', 'Life Lessons', 'Writing'] |
Reading as an Intentional Practice | Do you read on purpose or for distraction?
Reading as an Intentional Practice
What happens when you read? Think about it.
How do you read? Do you sit still and sink into the material, like lowering yourself into a deliciously hot bath? Or do you skim, going way too fast, greedy for the gist, oblivious to gift of context and detail?
I find so much of my reading is incredibly selfish. There’s a kind of “do me” attitude I’m not proud of. What will this do for me? What can I take from it into a conversation or something I’m writing that will make me seem like I’ve really thought about it?
It’s rarely that sacred exchange of my attention for your words. Rarely is it me stepping onto the ship of whatever you’re telling me, and experiencing the ocean of it while the shore of my laundry-laden life recedes.
Since I’ve become aware of this attitude, my little inner guru has given me a teaching: Give ten minutes of every day to reading as a spiritual practice. Put in place the following parameters:
Be intentional.
Give some thought to what you’re going to read. Don’t just read to distract yourself or pass the time. I don’t care if you’ve chosen the National Enquirer, just make sure you’ve decided to read it on purpose and for real. Ask your mind what it wants from language right now? Cookbook recipes? Newspaper article? Novel? Poem? Tabloid? Sacred texts? Whatever you’re drawn to reading, be intentional with what you choose. Get comfortable.
This is another area where intention comes into play. Decide where and how you’re going to sit. Make sure you’re comfortable. Do you feel better at the table, sitting on a good chair? Or would you rather read on the couch with your favorite pillow supporting your back? Think about it. Give your whole body to this little ten-minute vacation into someone else’s world. Set your timer.
Set it for ten minutes. It doesn’t matter, really, if you go over — but it does matter that you reach the ten-minute mark. This is to give yourself enough time to notice how you read. If you dive right in and lose yourself in the piece, then awesome! If you find yourself fidgeting, or skipping around, maybe peeking at the end or whatever, notice that. Take a breath and invite yourself to settle down a little. (Sometimes reading out loud helps me focus.) Then READ.
Go on the ride. Hold hands with the one in you that’s learning something new. This is a precious moment. The world is telling itself to you. Are you listening? Are you on the ship, or did you wave goodbye to it from the shore, waiting for the timer to ding? Pay attention.
Tending to the quality of our reading is just one of the many gateways into this Right Now thing. This moment. The one where Alice has taken a sip and nothing is its right size. The one where Harry realizes the Patronus he thought was his father was actually himself. The one where we ride the train with Gandhi as a young man, or where we learn about the structure of the shoulder girdle, or where we’re told who really killed Laci Peterson.
Or it’s this one. The moment where you are letting my words bear you away from your laundry and into the deep waters of a language that connects us and makes us one.
Today, I bow to my eyes, my brain, to language itself. To my parents and all our people who taught me English, my nanny who taught me French, and my friends much later in Tuscany who taught me Italian. To all my teachers who taught me not only the shapes of letters and how to sound out the words but the body parts of a sentence, the anatomy of a paragraph, the nervous system of a story.
To anyone who said, “Read this,” and it was a great book.
To authors and publishers and books and bookstores, and amazon.com, and kindle, and all its brothers and sisters in the digital world.
Scripts, articles, exposés, dissertations, and declarations of independence. Footnotes. Sidebars.
Scribbles in my journal.
Let’s bring a little reverence to our reading life,
break out the anchor and set sail on the uncharted waters
of all the stories that have made their way into our hands. | https://tinalear.medium.com/reading-as-an-intentional-practice-f5261865cbe0 | [] | 2019-10-22 20:32:57.251000+00:00 | ['Storytelling', 'Mindfulness', 'Publishers', 'Reading', 'Authors'] |
Content Marketing Tip #1: Say Something | So, another year goes by and I’ve been doing this content marketing thing for what feels like a lifetime. If I ever want to depress myself I just Google myself and see how far back I go — God, is that what I published in the year 2000? I invented SEO (I didn’t really).
I do feel like I’ve sharpened my content marketing tools this year. I wrote a pretty decent paper on AI, for one. I’ve worked across two different industries. I’ve written a few articles that will go in my portfolio. Also, I got back into creative writing and that’s helped my content marketing. Here, old brain, have a sip of this imagination juice.
I’m also in the strange position of mentoring budding, young content creators. I ran a workshop earlier this year on content creation and copywriting. It’s a huge subject, worthy of a number of workshops. Great content creation, it turns out, is as much about unlearning as it is about learning.
But, my number one tip is this: say something.
I’ll break that down into several pillars, or principles.
Intention
Think fire. Think rockets. Think raging passion. Intention, always start with intention. And here’s the difficult thing to grasp: you may not always say that something, or find something worth saying, but it’s the spunky attitude, your intention, that will drive you to create content.
They talk about intention a lot in martial arts. It’s more than direction. It’s placing your sight beyond, taking aim like a bowsman. It’s where your mind goes, and the rest of you follows. When you get it right you feel superhuman.
Intention is the driver. If you start with a sort of flakey, soft attitude, sad about having to fill a blank screen with stuff that resembles writing, you’ll struggle.
The dream is, of course, to create unique content, but, when was the last time you came across truly unique content? Don’t get ahead of yourself and start talking about engagement, or great content. Everybody wants to go from learning to type to being Hemingway, overnight. Easy does it — I’m still learning after a decade-plus of doing this.
You’re the strongest angle
Find an angle, what’s the angle, God what is it… Yes, finding an angle worth writing about can be a real headache for content creators. But what if you were the angle?
I don’t mean that everything needs to sound like your hot take, but offering a glimmer of personality in even the most opaque copy never hurts. Bridge the gap between you and the reader.
You want to sound like a human being has written your copy, not a copywriting algorithm. Take risks. Flip things around.
Start noticing when you create gaps between you and the reader. You might be overusing terminology or jargon with nothing else behind it. You, the writer, the content creator, has gone missing. You’ve created elevator music. Muzak. You’re asking people to skip to the summary.
Recommendation:
Read more Felicia C. Sullivan, analyse it — what is it about Felicia’s work that stands out? There’s an energy about her work that’s hard to pin down. It’s not templated, and purposefully so.
Look at the images she uses. She’s generally not going to Unsplash or Pexels like the rest of us. The images are vibrant, punchy; they ask us to feel something, to respond.
Felicia’s copywriting is worthy of analysis in itself, but let’s just say this — you can tell she didn’t start writing yesterday. Her writing is memorable, full of metaphors, symbolism, and she constantly offers personal examples.
Her work is divisive — bristling.
So whether it’s her marketing, personal pieces, or fiction, the work of Felicia C. Sullivan always sounds like the work of Felicia C. Sullivan. Not anonymous marketer #21921921911.
Is it working? 30k followers seem to think so.
No fear
Instead of obsessively looking for an angle, challenge your own preconceptions. For example, I know that wireless earbuds are all the rage, but what if I don’t want the headache of recharging tech? What if I find the idea of a cable quite comforting? What if I have…cable fetish. Ok, cable fetish might be taking it a bit far, but you get the idea.
Be honest with yourself and with your audience. Hell, I’m such an addict to honesty I started a company called Honest Publishing (in 2010, I left in 2013, this is not a plug, I have nothing to do with them now). Don’t be afraid to admit that you know very little, or nothing.
Find something worth talking about that’s not immediately obvious, and don’t be afraid to argue your point. Or to start an argument. Or to be controversial. Are you a bit of a prick? Bring out your inner prick, then. Are you quiet and withdrawn? I’m quiet and withdrawn, but do I sound like it in my work?
Whispering can be sexy.
The amount of content that I read that is stating the obvious is unbelievable. Characterless. You see, it takes guts to write with no fear. It takes guts to put yourself and your opinions out there. You might get shot down. Someone might disagree or be better than you. You might lose everything, right?
Especially these days when social media will crucify anyone who doesn’t tow the social media party line. And towing the party lines makes for boring content. Fuck the party line.
If you’re not sweating about the content you’re about to put out there, then you’re doing it wrong.
Principles are just principles
Let me be clear, no two jobs are the same. You may be working a job that allows for minimal creativity — you’re literally pasting a bunch of buzzwords in and changing the order slightly so that Google doesn’t penalise you. I get it. I’ve been there. Your soul is taking a pounding. I know.
But that’s enough about me — what about you? How do you feel about your content marketing this year? What do you need to work on in the next year? | https://medium.com/illumination/content-marketing-tip-1-say-something-b513a688f279 | ['Bogdan Tiganov'] | 2020-12-22 12:51:31.886000+00:00 | ['Content Writing', 'Content', 'Content Creation', 'Content Marketing', 'Marketing'] |
Cloud Cost Optimization | Save $ on your cloud infrastructure | Get your Template | Cloud Costs are rising? With a combined experience of more than 22+ years between them, Sunu and Vikrant helps you on your cloud cost optimization journey, without compromising on development and user experience.
We break down tips, ticks, hacks and best practices to keep costs in check. We are talking about Network, Storage, Compute, CDN, Human Factor, Devops and pretty much the entire spectrum on cloud cost optimization.
This video is all you need to watch to know where to look to save costs and how to do it. We cut down the theory and share this with you based on our real, practical experiences.
We’ve saved Millions of $ for our clients (some of which are Fortune 500). We’ve also advised startups to set it right so they get more done out of cloud. Share this #cloudcostoptimization video with your network and start saving. | https://medium.com/cuelogic-technologies/cloud-cost-optimization-save-on-your-cloud-infrastructure-get-your-template-a29eb2f4fc36 | ['Cuelogic Technologies'] | 2019-12-02 07:58:32.242000+00:00 | ['Cloud Optimization', 'Cloud Storage', 'Cloud Services', 'Cloud Computing', 'Cloud Solutions'] |
Understanding Dice Loss for Crisp Boundary Detection | Dice Loss
Dice loss originates from Sørensen–Dice coefficient, which is a statistic developed in 1940s to gauge the similarity between two samples [Wikipedia]. It was brought to computer vision community by Milletari et al. in 2016 for 3D medical image segmentation.
Fig.3: Dice coefficient
Fig.3 shows the equation of Dice coefficient, in which pi and gi represent pairs of corresponding pixel values of prediction and ground truth, respectively. In boundary detection scenario, the values of pi and gi are either 0 or 1, representing whether the pixel is boundary (value of 1) or not (value of 0). Therefore, the denominator is the sum of total boundary pixels of both prediction and ground truth, and the numerator is the sum of correctly predicted boundary pixels because the sum increments only when pi and gi match (both of value 1).
Fig.4: Dice coefficient (set view)
Fig.4 is another view of Fig.3 from the perspective of set theory, in which the Dice coefficient (DSC) is a measure of overlap between two sets. For example, if two sets A and B overlap perfectly, DSC gets its maximum value to 1. Otherwise, DSC starts to decrease, getting to its minimum value to 0 if the two sets don ‘t overlap at all. Therefore, the range of DSC is between 0 and 1, the larger the better. Thus we can use 1-DSC as Dice loss to maximize the overlap between two sets.
In boundary detection tasks, the ground truth boundary pixels and predicted boundary pixels can be viewed as two sets. By leveraging Dice loss, the two sets are trained to overlap little by little. As shown in Fig.4, the denominator considers the total number of boundary pixels at global scale, while the numerator considers the overlap between the two sets at local scale. Therefore, Dice loss considers the loss information both locally and globally, which is critical for high accuracy. | https://medium.com/ai-salon/understanding-dice-loss-for-crisp-boundary-detection-bb30c2e5f62b | ['Shuchen Du'] | 2020-02-29 17:36:51.532000+00:00 | ['Deep Learning', 'Loss Function', 'Computer Vision'] |
Medieval UFO sightings | Medieval UFO Sightings
Fact or fiction?
UFOs have been around for thousands of years, but their explanations are always changing. Taken at face value, Unidentified Flying Object just means something that flies about which we can’t make any concrete statement on what it is.
In prehistoric times, a UFO would probably be explained as a bird. In the days of the ancient Greeks, a visit from Zeus. In medieval times, an omen from God, and in our modern age, advanced spacecraft from space.
While there is no hard evidence of extraterrestrial visitation on Earth, there are interesting records that would get even a skeptic interested. One such record of an extraordinary event is that of the 16th century celestial phenomenon over Nuremberg.
In 1561, a German broadsheet newspaper in the town of Nuremberg was released. An illustration covered the entire length of the paper, done by a local artist called Hanns Glaser. The contents? Judge for yourself:
While a little hard to read as english translated from gothic German, the news article reads:
‘In the morning of April 14, 1561, at daybreak, between 4 and 5 a.m., a dreadful apparition occurred on the sun, and then this was seen in Nuremberg in the city, before the gates and in the country — by many men and women. At first there appeared in the middle of the sun two blood-red semi-circular arcs, just like the moon in its last quarter. And in the sun, above and below and on both sides, the colour was blood, there stood a round ball of partly dull, partly black ferrous colour. Likewise there stood on both sides and as a torus about the sun such blood-red ones and other balls in large number, about three in a line and four in a square, also some alone. In between these globes there were visible a few blood-red crosses, between which there were blood-red strips, becoming thicker to the rear and in the front malleable like the rods of reed-grass, which were intermingled, among them two big rods, one on the right, the other to the left. These all started to fight among themselves……..After all this there was something like a black spear, very long and thick, sighted; the shaft pointed to the east, the point pointed west. Whatever such signs mean, God alone knows.’
It is also mentioned in later articles from that period that some of the balls of light crashed in the neighboring fields, only to evaporate into smoke.
Certainly, something interesting happened over the skies of Germany that day.
As with all extraordinary phenomena, this one also has its detractors, as it should — it does sound like a great piece of medieval science-fiction.
Some say that this was an example of a sundog — where sunlight is reflected off ice crystals in the atmosphere to create the illusion of multiple points of light in the sky.
However, that theory fails to explain the erratic movements of the orbs or the black spear-shaped object.
A more likely explanation is given by the late great psychologist Carl Jung who believed that this was a case of mass delusion or, at the least, a natural occurrence with religious interpretations. For example, the orbs fighting could have the mating dances of birds/insects. | https://medium.com/lessons-from-history/medieval-ufo-sightings-925b20e8ba85 | ['S Pats'] | 2020-12-28 08:12:30.458000+00:00 | ['Space', 'History', 'Medieval', 'UFO', 'Nonfiction'] |
Introduction to Recommender Systems: Deal with Overloading Information | This blog is written and maintained by students in the Professional Master’s Program in the School of Computing Science at Simon Fraser University as part of their course credit. To learn more about this unique program, please visit {sfu.ca/computing/pmp}.
Authors: Quan Yuan, Yuheng Liu, Wenlong Wu
Introduction
With the continuous development of Internet technology and the increasing popularity of smart devices, more and more data are generated in current days. Once we get this massive amount of data, the industry-wide personalized recommendation technology becomes easier to implement, whether it’s Amazon or YouTube, or any big tech company is undoubtedly the biggest beneficiary of this area.
Unlike ordinary personal items, smart devices can uniquely link to a specific person, and these personal smart devices are difficult to share with other people, which makes the browsing, transaction, and other behavioral data on your mobile phone play a significant analytical value on the recommender system.
From the perspective of the e-commerce platform, the essential goal of the recommender system is to recommend products that are most likely to be sold to consumers so that the data can be fully utilized. The accuracy becomes higher with the continuous enrichment of personal user data and technical approach.
Methodology
Data Collection
Recommender system uses various information of users and items to obtain a proper recommendation. That means a sufficient dataset is a basis for building an intelligent recommender system. A general dataset contains attributes of users and items and ideally also the interactions of users with items. The following figure shows basic approaches to link a certain user and items.
Several data collection methods and the problem we will meet building the system are mentioned below.
1. Attributes
Both users and items have attributes that provide critical information for the recommender system, such as users’ gender, interests, and descriptions of items. Data mining methods can be used to extract relevant key knowledge from those attributes. By calculating the similarity of the user’s attributes and item’s attributes, we could use the k-nearest neighbors algorithm to obtain the closest item for users. This is a content-based recommender system especially useful when we lack interaction information between users and items. Conversely, it is also a limit. According to the most prevailing recommender system algorithms, interactions of users and items play an essential role.
2. Interaction
To be more specific, interactions refer to the actions and behaviors of users and records of feedback on the server. User behaviors provide lots of useful information. Based on the user behavior, we could tell their recent interests, their concern about a particular item, etc. There are two types of feedback. The explicit one is that users directly provide their attitude toward items such as clicking the “like/dislike” button. However, lots of users leave the page instead of doing that. It turns out that the implicit one, the record of users’ interactions, is more important than it. Suppose a user has read a description of an item up to the end, which means he may have some interest in it. If the user views an item for a few seconds, we can infer this item is not what he wants at that moment. Besides the user’s behavior, the recommendation also depends on time, domain, and other contextual factors.
3. Context information
Context information includes time, location and mood, etc. It is also an important part of the recommender system. For example, the recent interactions are much more important than the past since they represent users’recent needs. For a music recommender system, users would choose different music based on their moods. For an upcoming festival, users would buy some specific decorations. It is like we add a real-time property to the recommender system. Due to the importance of context information, it is important to record users’interactions and update the model frequently to be able to generate new recommendations in real-time.
4. Tag system
Tag system on Instagram
Based on using attributes to provide recommendations, the tag system is a useful tool to create a relationship between users’ interests and items and is widely used on some movie-related and video-related websites. On the one hand, the tag describes the user’s interest. On the other hand, it is a brief attribute of the item. Some websites prefer to provide several related tags for users instead of letting users improvise in order to improve the quality of tags.
5. Cold start problem
Cold start is a common problem in the recommender system caused by the lack of user’s or item’s information. When a new user or a new item is added to the platform, how does the recommender system do without data? It turns out several typical solutions for obtaining original information. You definitely have ever used authorized login to create a new account on a website. As long as signing up with your other social networking account, the system can obtain some of your records from it and extract some useful information as data used in the recommender system. Some websites ask a new user to select some tags intriguing him/her at the beginning. Then, the system will recommend some popular items based on the user’s selections as the appetizer. Random strategy is also used as a solution to the “cold start” problem. Recommending random items to new users or new items to random users can give the recommender system positive feedback based on the user’s interaction with the item.
Algorithms:
1. Collaborative Filtering
Generally, Collaborative Filtering makes predictions based on the past experience of the user. The past experience reflects the user’s preference. Reviews, ratings, number of clicks, duration of browsing, choice of items, and many other factors could indicate what the user may be interested in and be willing to spend time on. There are mainly two types of Collaborative Filtering, which are user-based and item-based.
User-Based:
By collecting the user information and items that users are interested in, the system allocates users with similar preferences in the same group. Suppose user A and user B are frequent users of Netflix, and they have similar preferences in movies. If Titanic is in B’s watching history and A has not watched it, the movie Titanic will be recommended to B as they are determined to have similar tastes according to their experience.
User-Based Formula
Item-Based:
Based on a large amount of data, the system compares the items that may be preferred by similar types of users and determine whether these items are similar or not. The system lists the preferences of each user and recommends similar items to the user.
Item-Based Formula
As mentioned above, both these two types of Collaborative Filtering are based on a large amount of data. The experience of the user and the information of items play an important role in this algorithm.
By comparing the similarity among users and items, the system links potential preferred items to the users. To calculate the similarity, Euclidean Distance is the most direct approach. The smaller the result is, the higher the correlation between two items. However, a negative correlation is hard to be detected in this method. If two items are highly negatively correlated, the result of Euclidean Distance will still be large. Based on this, Pearson Correlation is a better solution as it covers both positive and negative correlations. In addition to this, cosine similarity is another effective approach, which illustrates the ratings in terms of vectors and calculates the cosine value between two vectors to find the similarity.
Collaborative Filtering involves an important machine learning algorithm — — K-Means Clustering. It classifies data points into different clusters so that data points in the same cluster have more similarities than others. First, it randomly plots several data points on the plain. Then, It implements the following steps iteratively until the centroids don’t change.
a) Assign the data points to the nearest centroid and thus construct clusters centered at those centroids
b) Find the center point in each cluster and re-define it to be the new centroid
The following graph shows how this algorithm is implemented when k=2.
2. Content-based
Unlike Collaborative Filtering, Content-based is focused on the contents of the item. It involves data preprocessing and feature extraction. For each item, the system builds a vector containing the extracted features. Also, the system builds a profile for each user based on his experience and preference. Thus, this item will be recommended to the user who is interested in the item with similar features. It might be confusing what the difference between Collaborative Filtering Item-based and Content-based is. For the Item-based algorithm, it is focused on the user’s ratings on the items rather than what contents this user is interested in. For the Content-based algorithm, it concentrates on the common features between the user profile and item profile. As this algorithm clearly lists the features of the item, it can better explain why it is recommended, and it may have a higher chance of meeting the user’s taste. However, extracting features is difficult, as it is hard to define which features are typical and important to the item.
Application
Here are some examples of the recommender system applied to real applications.
1. Amazon
Amazon is a multinational technology company that focuses on e-commerce. G. Linden, B. Smith, and J. York explained Amazon’s Item-to-Item Collaborative Filtering in their paper.
As shown above, the Item-to-Item Collaborative Filtering algorithm can provide customers with product recommendations based on the items in their shopping cart. This feature is similar to impulse goods at the checkout of the supermarket, but Amazon’s impulse goods are personalized for each customer.
In fact, Item-to-Item Collaborative Filtering is item-based Collaborative Filtering. It put all purchased and rated items to a recommendation list instead of matching the user to similar customers.
2. TikTok
TikTok, a video-sharing social networking service. With its advanced and high-accuracy recommendation algorithm, it can make users spend hours on its application a day. In research conducted by Jiamin and Leipeng, they try to reveal TikTok’s mysterious algorithms by introducing feature engineering.
The purpose of feature engineering is to select better features for generating better training data. The common methods to select features are Pearson correlation coefficient, mutual information, distance correlation. By using these similarity tests, the most useful feature can be selected and improve the outcome. For example, users click to see some travel-related videos; the system will record this feature of the user, and look for this feature from later video features (the creator, the music ID, the gender of the character in the video) to ensure there is a high probability that the following videos will also be travel-related.
Conclusion:
The recommender system is a creative solution to deal with overloading information as well as the search engine. The difference between them is that the search engine needs users to have a specific target. However, the recommender system aims to solve the situation that users have no idea about what they want at that moment. Based on the user’s interests and past behavior, the recommender system is able to provide items to the user automatically. The article briefly introduced some data collection methods and prevailing algorithms used in this domain. A few practical applications were given to interpret the deployment of the recommender system in the real world. Nevertheless, all we talked about is just a small piece of this significant topic. Generally, the different company has their recommender system framework base on their business. It’s not the simple combination of the mentioned methods. The recommender system is more like a free domain that you could come up with various innovative ideas to explore the link among users and items.
Thanks for reading! 🤗🤗🤗
References: | https://medium.com/sfu-cspmp/a-brief-introduction-to-recommender-systems-deal-with-overloading-information-58c808201a6f | ['Wenlong Wu'] | 2020-02-04 07:07:52.687000+00:00 | ['Blog Post', 'Machine Learning', 'Recommender Systems', 'Big Data', 'Data Collection'] |
How Will Quantum Computing Impact Blockchain Technology? | Cryptographers often voice the concern that blockchain technology could become obsolete upon the development of quantum computers (probably 10 to 15 years from now). They fear that once developed, quantum computers will be able to break the security mechanisms blockchains use (asymmetric cryptography and hashing). Though these concerns are somewhat justified, blockchain developers have time to proactively address the threat.
What is Quantum Computing?
To understand the security threat posed by quantum computers, it is important to understand how they differ from classical computers. A classical computer uses a base 2 numerical system: it understands, processes and communicates data exclusively using bits. Each bit can hold only one of two positions: 0 or 1. We can think of each bit as a light switch: it can either be on or off.
Quantum computers, on the other hand, use qubits which can hold two positions at once. (For example, a qubit could hold both a 0 and 1 at the same time). This phenomenon, known as superposition, is the secret weapon of quantum computing. Superposition decreases the number of operations required for the computer to solve a problem. Thus, quantum computers can handle multiple complex queries simultaneously.
A quantum computer’s ability to engage in more computations per second than a classical computer explains why many computer scientists believe that a quantum computer will be able to crack current cryptographic protocols with relative ease.
A Classical bit compared to a Quantum Bit or ‘Qubit’
Asymmetric Cryptography, Symmetric Cryptography, and Hashing
To gauge how, specifically, quantum computers will impact cryptography, it is worth understanding a few kinds of cryptography widely in use today. I have discussed cryptography and hashing previously. Here is a quick recap:
· Symmetric cryptography uses the same cryptographic key for encryption and decryption. This is similar to using the same key to lock the front door to your house in the morning, and to unlock it in the evening.
· Asymmetric cryptography, also called public-key cryptography, uses a pair of keys for each transaction: a public key and a private key. The public key encrypts a message, while the private key decrypts it. They must be used together to execute a transaction.
· Hashing algorithms scramble data into a unique string of letters and numbers that represents that specific data. Any given piece of untampered data will always produce the exact same hash.
Blockchains use two primary security tools: encryption via asymmetric cryptography, and hashing algorithms.
Asymmetric Cryptography Depends on the Difficulty of Factoring Large Numbers
Asymmetric cryptography algorithms, which secure blockchains, are based on prime numbers and factors. In every blockchain protocol, the public and private keys are both extremely large numbers that have been hashed into a series of smaller letters and numbers. Since finding the prime factors of an enormous number is virtually impossible for classical computers, blockchains are tamper-proof. | https://medium.com/blockstreethq/https-medium-com-shaanray-how-quantum-computing-will-impact-blockchain-technology-f69e4a3af25e | ['Shaan Ray'] | 2019-12-15 06:50:27.551000+00:00 | ['Quantum Computing', 'Bitcoin', 'Bhq Contributors', 'Science', 'Blockchain'] |
Avoiding jagged edges on gradients. | A really common design pattern I see come up a lot are those angled blocks of colour in the bottom of containers, you’ve probably come across it, it looks something like the image below.
Angled colour block inside a square
This happened to turn up on one of our projects and rather than implementing with an image or SVG we decided to use a linear-gradient, the benefit of this is that it’s really easy to change the angle, colour etc within the CSS. The problem we had was that the angle of the gradient in the space available caused a really horrible jagged edge.
Figure 2: jagged edged angled gradient
The real life example looked like Figure 2, as you can see the jagged edge is quite prominent.
Luckily, there is a really simply way to solve this problem! Normally when you want to create a sharp line in a linear gradient you’d create your gradient like the code below.
background:
linear-gradient(176deg, white, white 75%, black 75%, black 100%);
This basically ensures that from 0–75% we have a white colour and then from 75% on, it’s black! This creates a really nice sharp line. Unfortunately, depending on the angle, and the size of the container you can end up with the ugly zig zag.
Fixing this requires a very small change — add a decimal to the second percentage which will create a slight blur on the gradient. This will need to be adjusted depending on your use case to ensure the blur isn’t too prominent — but visually it will result in a smoother gradient!
background:
linear-gradient(176deg, white, white 75%, black 75.3%, black 100%);
Side by side example of the jagged vs smooth gradient.
Update:
Several people comment to me on twitter about using calc, this is also an option which can be achieve by doing something like the following:
background:
linear-gradient(176deg, white, white 75%, black calc(75% + 1px), black 100%);
The Codepen example:
Hopefully this helps with your issues in the future!
Enjoy!
Mandy | https://medium.com/pixel-and-ink/avoiding-jagged-edges-on-gradients-f485cc7401f5 | ['Mandy Michael'] | 2019-11-19 09:12:41.035000+00:00 | ['CSS', 'Web Development', 'Design', 'Front End Development'] |
A User Authentication Android App made in Android Studio | After some processing time, the Android Studio main window appears.
Now take a moment to review the most important files. First, be sure the Project window is open (select View > Tool Windows > Project) and the Android view is selected from the drop-down list at the top of that window. You can then see the following files:
app > java > com.example.userauthentication > MainActivity
This is the main activity. It’s the entry point for your app. When you build and run your app, the system launches an instance of this Activity and loads its layout.
app > res > layout > activity_main.xml
This XML file defines the layout for the activity's user interface (UI). It contains a TextView element with the text "Hello, World!"
app > manifests > AndroidManifest.xml
The manifest file describes the fundamental characteristics of the app and defines each of its components.
Gradle Scripts > build.gradle
There are two files with this name: one for the project, "Project: User Authentication" and one for the app module, "Module: app." Each module has its own build.gradle file, but this project currently has just one module. Use each module's build.gradle file to control how the Gradle plugin builds your app. For more information about this file, see Configure your build.
Before you run your first app, get your smart phone and enable USB debugging by following the instructions over here. After, enabling USB debugging, connect your smart phone to your PC and you are ready to go.
In Android Studio, select Run > Run ‘app’ or click the Run icon in the toolbar.
Congratulations on creating your very first app. But you are far from what we have to achieve. Next we need to create the rest of the applications.
Setting up the files:
Lets start with app > res > drawables. This is the directory where we will be storing all the graphical elements for our application. We are going to store three different categories of graphic elements.
Images. (filename: background.jpg, image.png)
Icon elements. (filenames: account.xml, email.xml, eye.xml, eye_off.xml, lock.xml)
Custom graphic file. (round_button_2.xml)
The images used here are for background and a logo in the logged-in session. You can use any image you like but be sure to use the same file name for the next part where we will be coding the back-end and creating activity layouts.
Icon elements are XML files for vector graphic icons used in our application. You can find and download these files from here.
The custom graphic file here is used for creating a rich look for our buttons in the application. It’s an XML file containing various tags and attributes specifying the properties that the button will have for its appearance. | https://harsh-k99.medium.com/a-user-authentication-android-app-made-in-android-studio-728dc92d9243 | ['Harsh Kudtarkar'] | 2020-12-10 09:37:28.775000+00:00 | ['UI', 'Android App Development', 'Rest Api', 'Java'] |
Stories of Greener Grass and Bluer Skies. | As I walk barefoot on the green lands,
Feeling the wetness and the freshness of the grass.
I looked up to the blue sky,
to those beautiful clouds sliding in mind.
And I wondered,
is this all there is to be discovered?
My mind slipped between senses of sight and touch,
flipped to those beautiful harmonics sung by birds as such.
And I wondered,
is there better music?
Nature itself can be so exclusive.
I walked and walked,
not being able to fill my gap,
when wonder overcame my mind.
Imaging things and moments
not happening now,
filled with omens and poets around.
Picturing characters
with different situations,
from all nations.
Only then I realized,
I was smiling for the first time.
I was so surprised.
My happiness was so great,
as I thought what a beautiful affair,
is to create something from thin air.
This is why I was born!
In so many categories,
I was brought to this great world to share stories.
Stories of greener grass and bluer skies.
Pictures in my mind,
of humankind.
Predesigned,
to serve a purpose,
bigger than life combined. | https://medium.com/flicker-and-flight/stories-of-greener-grass-and-bluer-skies-a3511c385e9f | ['Jacobo The First'] | 2020-10-28 03:36:40.460000+00:00 | ['Storytelling', 'Mindfulness', 'Poetry', 'Nature', 'Poem'] |
Metrics, Metrics, Teams. How to Set-Up the Ultimate Lead Generation Machine — Part 2 | A comprehensive guide to help you take actions and optimize your Customer Acquisition Cost (CAC)
[This article is part of a group of 4 dedicated to provide an insightful guide to online lead generation: “How to Set-Up the Ultimate Lead Generation Machine”. Part 1: CAC, Sales Volume & Business Optimum; Part 2: Metrics, Metrics, Teams; Part 3: Log, Build, Automate; Part 4: Arbitrage Tips & Tools]
In Part 1, we explored the two main drivers of all lead-fueled business: CAC and Sales Volume, and discussed how to set your business targets to find your Gross Margin Optimum. Even if very detailed, the first article was a helicopter view of the lead generation problem, barely touching the surface of concrete, actionable insights for marketers, sales directors and other decision makers.
In this second article, I intend to get much deeper into each step of the lead generation funnel, detail who is responsible for what within a regular organization, and introduce the main KPIs you should focus on. If there is one thing I’m 100% sure of in this industry, it is that the three factors of success are Metrics, Metrics and Metrics. I mean it, lead generation is all about key metrics!
Metrics, Metrics, Metrics. Lead generation is all about Key Metrics.
Here is what I will cover today:
The breakdown of the lead generation user experience funnel and the naming and conventions used by professionals to describe each of its steps
All lead generation key metrics you should know about (and how to compute them)
How to organize a high performing lead generation team and which KPI each team should focus on.
Are you up to it? I am, let’s get down to it.
1. The 8 Steps of the User Experience Funnel
Let’s start from the start. The lead generation process is a funnel made of different steps along the consumer online buyer journey. From the ad displayed, viewed and eventually clicked by the prospect, to the final consumption of the product or service. The art of lead generation relies on how you can actually help people with a need, a question or a project. Like my grandma looking to remodel her sunroom for instance. Guide her through the different layers of the funnel as fast, safe and effectively as it can possibly be in order to secure a good deal aligned with what she is looking for in the first place.
That is why you need to build a comprehensive understanding of your own funnel, using key metrics that are widely used among the Industry. Looking more closely into it, most user experience funnels are made of 8 steps:
I think the slide seen above is quite self-explanatory. The online buyer journey starts in a Channel (Google Search, Facebook, Linkedin, Instagram, the email inbox or an online newspaper for instance). In this channel, the user browses content and is frequently exposed to online advertising. This is how Views (Step 1) are generated. If the ad audience targeting, copy writing, and imaging are all well executed, then the user will eventually click on it. Congrats, you just generated a Click (Step 2)! Keep in mind that for many technical reasons, all clicks do not end up becoming Visitors to your website. Online marketers usually see a discrepancy of roughly 5 to 10% between the number of clicks channels claim to have generated (i.e. that you have to pay for) versus the number of visitors your website analytics provider will count for you.
When landing on your website, users will soon enough be exposed to:
Free Contents to help them self-qualify their intent and leave the page if not interested
to help them self-qualify their intent and leave the page if not interested Calls To Action (CTA) like Newsletter subscription boxes, Lead Magnets (usually a piece of digital, downloadable content, such as a free PDF, report, eBook, white-paper, video, etc.), Forms and Toll-free Telephone Numbers.
When users find what they are looking for on the website, and willingly share their contact information to start engaging with a brand, they become Conversions. However, not all conversions are equal. Some are purely useless. Others are not ready to be sent to the sales team. Finally some are good, worth working on, what the industry calls Marketing Qualified Leads (MQL).
What is the difference between Conversions (aka Captured Leads) and Marketing Qualified Leads (MQL) anyway?
Obviously fake conversions: wrong phone number, email address, zip code. Silly personal data (name = Mickey, last name = Mouse), irrelevant answers to mandatory open field questions (“blablabla”). These leads are called Scrub Leads.
wrong phone number, email address, zip code. Silly personal data (name = Mickey, last name = Mouse), irrelevant answers to mandatory open field questions (“blablabla”). These leads are called Duplicates: leads you already have in your database.
leads you already have in your database. Conversions with mandatory fields missing: theoretically this should never happen if you use automated fields validators, and if they are implemented properly. But you know, errors happen everyday in the technical world and sometimes you end up logging conversions that miss some crucial information, preventing them from being considered MQL. These are Scrub Leads too.
theoretically this should never happen if you use automated fields validators, and if they are implemented properly. But you know, errors happen everyday in the technical world and sometimes you end up logging conversions that miss some crucial information, preventing them from being considered MQL. These are Scrub Leads too. Conversions without Compliance: users browsing with an Adblock sometimes prevent the landing page from logging proper user consent (TCPA / GDPR language). This usually happens when you use a third party compliance service. From a purely legal standpoint, you don’t want to take the risk of calling back these leads.
users browsing with an Adblock sometimes prevent the landing page from logging proper user consent (TCPA / GDPR language). This usually happens when you use a third party compliance service. From a purely legal standpoint, you don’t want to take the risk of calling back these leads. Conversions that need to mature: some conversions need more time to mature through the lead nurturing journey before being put in contact with the Brand’s salesforce. For example a white-paper download is a Conversion, but the intent is usually very low: at this point the user is probably not ready to buy anything from the brand. You will have to send a couple (automated) emails to understand A) if this is an interesting lead for you and B) if the user is now ok to start the sales process. If the answer to these two questions is yes, then you can consider it a MQL.
Disclaimer: leads needing to mature are very e-commerce centric. Most telesales-based brands skip this step, and try to connect as quickly as they can when they get access to the user phone number. They consider that the intent at this point is consistent enough to give it a try.
Now, the sales team has just received a MQL to start working on. Obviously the first thing to do is to reach out and contact the consumer, usually by phone. If someone picks up the phone, then most professional outbound call softwares will consider the call a success and the MQL a Contacted Lead. If the user is not reachable after a given number of attempts, then the sales process will stop and the lead will stay a MQL. You can keep in mind that the Contact Rate depends both on the quality of the Intent (i.e. MQL quality) and the salesforce call back process (mostly the time duration between the moment a lead is flagged as MQL in your system and the moment the lead receives a call).
From Contacted to Sales Qualified Leads (SQL): once your sales representative has the prospect over the phone, he/she will try to quickly assess if there is a real business opportunity lying behind the inquiry. To do so he/she will have to answer three questions:
Does the contact information match with the person over the phone? (obvious, but critical)
Does the person have a real need / question regarding the type of product / service offered by the brand?
Is the person ready to consider buying such a product / service
Yes, Yes, Yes? Then this contacted lead becomes a SQL!
SQL are probably what you had in mind when you started reading this article. What your team is highly focused on. And yes! SQL are a cornerstones to lead generation optimization.
Last but not least, some SQL will finally be transformed into sales, and become customers. Why then split the “Sales” step into Gross Sale & Net Sale? There is nothing mandatory here, but this split is usually very useful when tied to business operations. Let me give you two examples: education brands differentiate Enroll (deposit) vs Start. Home improvement brands like home security providers or bathroom contractors differentiate Sale (deposit) vs Installed (full payment). So, as you can easily imagine, there is another drop-off between Gross and Net Sales, that you want to monitor and optimize.
Now that we have divided the buyer experience into mutually exclusive steps (meaning that a prospect cannot stand on more than one step at a time), with clear naming, we can start computing key metrics and measure drop off between steps to understand where to start the optimization work!
2. The Lead Generation Funnel Key Metrics (KPIs)
Each step of the funnel can be measured in Volume, Rate and Cost. In this section, I will present the main metric for each step, and how they interact to build the metric you need.
There are no successful companies, only successful teams.
Keep in mind that all key metrics are interlinked and that combined, they will lead you to the next stage of the journey, and eventually to your final destination: a sale. This is why I used the multiplication symbol between all of them on the slide seen above.
I believe that there are no successful companies, only successful teams. In a typical lead generation organization, each team will take care of running, monitoring, and optimizing a few parts of the funnel:
A) The Traffic Team
Dolead’s Traffic Team working from our EMEA HQ in Paris (Library Room)
The Marketing Team, more precisely the Online Traffic Team, is in charge of generating high volumes of high converting clicks on the landing page, at the lowest average cost per click possible. We already explained here why generating high clicks volume with low CPC is a very challenging target. Auction-based channels won’t let you be successful without a strong determination, extensive testing, and a good set of hacks.
The traffic team will focus first on Reach, meaning generating Hits, the number of times ads are displayed.
Views (Reach):
Volume = #Hits (number of impressions)
Rate = %Impression Share (the percentage of total possible impressions you captured)
Cost = Not to be computed as most channels are CPC-based. Still, if you buy display, video or email traffic, you may have to pay per impression (Cost Per Thousand, Cost Per Mille, Cost per View for video advertising)
Then, the team will have to answer the following questions: is the audience qualified enough? A good performing audience for a given campaign? Is the ad displayed well performing? Clear enough? Attractive enough? Visible enough?
Clicks:
Volume = #Clicks (number of clicks)
Rate = %CTR (Click Through Rate)
Cost = $CPC (Cost Per Click)
Clicks = Reach x CTR Cost (Marketing Budget) = Reach x CTR x CPC
If you want to take a closer look into today’s main CPC-based channel and its Auction Model, you can watch the video by Google Adwords’ Chief Economist Hal Varian, explaining how CPC, Bids and CTR are interlinked.
B) The Website Team (Marketing Operations / Tech Team)
There are two ways to create a well performing website. One is to ask your internal or outsourced Tech Team build and optimize it. Another one is to let the Marketing Team leverage an external landing page and A/B Testing software, like ABtasty, Unbounce or Instapage. The second option is much more agile, but less customizable.
Dolead’s Tech Team working from our EMEA HQ in Paris
The goal here is to convert clicks into actionable MQLs you can transfer to the Sales Team:
Conversions:
Volume = #Conversions (number of raw conversions logged into your system)
Rate = %CVR (Conversion Rate)
Cost = $CPA (Cost Per Acquisition, Cost Per Conversion)
Conversions = Reach x CTR x CVR CPA = CPC / CVR
Now, you need to convert conversions into Marketing Qualified Leads. Remember, some conversions will turn into scrub leads. Here is where you usually need the Tech Team to set-up rules to automatically exclude fake/bad leads.
Marketing Qualified Leads (MQL):
Volume = #MQL (number of “good” leads captured)
Rate = %MQL (MQL Rate)
Cost = $PPL (Price Per Lead = Cost Per Lead = Cost Per MQL)
MQL = Reach x CTR x CVR x MQL Rate PPL = CPA / MQL Rate
The PPL is usually the Price you pay to an external lead vendor, when you decide to outsource part of your lead generation efforts.
C) The Sales Team
Last in line, first to reach out to new potential customers. The Sales Team carries a heavy burden on its shoulders. Now that the marketing budget has been deployed and the leads have been qualified by the Marketing and Tech Teams, they have to convert all these efforts into sales, into revenue and into return on investment.
Dolead’s Sales Team working from our EMEA HQ in Paris (Canopy Meeting Room)
The Sales Team follows a three steps process: Contact, Qualify, Sell.
Contacted
Volume = #Contacted (number of contacted MQLs)
Rate = %Contacted (Contact Rate)
Cost = You don’t really need this metric. But you can compute it if need be (Price/Cost Per Contacted)
Sales Qualified Leads (SQL)
Volume = #SQL (number of Sales Qualified Leads)
Rate = %SQL (SQL Rate, Approval Rate)
Cost = $PPSQL (Price Per SQL, Per Approval)
Sale (Gross & Net)
Volume = #Sales (number of Sales)
Rate = %SR (Sales Rate)
Cost = $CAC (Customer Acquisition Cost)
3. When the Dust Settles: What Matters Most
You need 3 teams to run and optimize your lead generation funnels. Traffic, Marketing Operations and Sales. These teams have very different skillsets and know-how.
To simplify, if you had to give clear guidelines to each of your teams, you could focus on 3 main metrics: Cost Per Click (CPC), landing page Conversion Rate (CVR) and Sales Rate (SR). Teams are everything within an organization. Because all KPIs are interlinked, you need to make sure your teams are also interlinked, and learn to work together as one, even if they focus on different targets.
Organizing your teams that way will help you better serve your prospects, optimize your lead generation efforts, and monitor drop-offs between every key metrics.
At the end of the journey, this will also allow you to optimize your CAC: as you can see on the slide above, there is another way to compute CAC leveraging each team’s KPI.
CAC = CPC / (CVR x SR)
Conclusion
After reading all this, I do hope that you have a comprehensive understanding on how lead generation works, you know all the metrics you need to monitor to start optimizing your lead flow, and you have a better vision of your ideal team organization. Now, I hear you thinking “Ok this looks very powerful. But moving forward, can I build my own ultimate lead generation machine?”
The answer is Yes you can! And without a single line of code. We will cover this in my next article.
[This article is part of a group of 4 dedicated to provide an insightful guide to online lead generation: “How to Set-Up the Ultimate Lead Generation Machine”. Part 1: CAC, Sales Volume & Business Optimum; Part 2: Metrics, Metrics, Teams; Part 3: Log, Build, Automate; Part 4: Arbitrage Tips & Tools] | https://medium.com/online-marketing-and-entrepreneurship/metrics-metrics-metrics-how-to-set-up-the-ultimate-lead-generation-machine-part-2-3b92514034db | ['Arthur Saint-Père'] | 2020-05-07 14:45:12.287000+00:00 | ['Entrepreneurship', 'Lead Generation', 'Data Science', 'Sales', 'Online Marketing'] |
GitOps Observability — Visualizing the journey of a container | Trivia app
I have a very simple Trivia web application that I would like to host on Kubernetes using the GitOps pattern. It contains a simple React frontend and a Nodejs backend that queries existing APIs on the internet for fresh trivia questions.
Even in a simple application such as this one, there’s two microservices since we have a separate dockerfile for the frontend and backend. This would mean that I need to setup CI pipelines for these individually, and one common CD pipeline should be sufficient to deploy the changes to the cluster.
I’m following the 5 minutes GitOps pipeline with Bedrock guide which creates a skeleton for all my pipelines and a starter HLD. My current CI setup looks like this:
High Level Definition (HLD)
Kubernetes manifest files define the final configuration of the deployment on the cluster. Being extremely error prone, helm charts are a great tool for templating Kubernetes resource definitions. In any application, there are multiple microservices (n) and that leads to (n) helm chart configurations. Bedrock uses the concept of a High Level Definition (HLD) that allows you to define the components of your application at a higher level.
For example, in my Trivia application I need to include the frontend and backend microservices. But at an even higher level, I want to include Traefik in the cluster. I’m looking at having my high level definition at the root folder look something like below:
type: component
subcomponents:
- name: hello-world-full-stack
type: component
method: local
path: ./hello-world-full-stack
- name: traefik2
source:
method: git
path: definitions/traefik2 name: default-componenttype: componentsubcomponents:- name: hello-world-full-stacktype: componentmethod: localpath: ./hello-world-full-stack- name: traefik2source: https://github.com/microsoft/fabrikate-definitions.git method: gitpath: definitions/traefik2
And the definition at the nested level to be:
name: hello-world-full-stack
type: component
subcomponents:
- name: backend
type: component
method: local
path: ./backend
- name: frontend
type: component
method: local
path: ./frontend
Who doesn’t love recursion!
Fabrikate is the tool behind HLD that simplifies the GitOps workflow: it takes this high level description of the deployment, a target environment configuration (eg. QA or PROD) and renders the Kubernetes resource manifests for that deployment by utilizing templating tools such as Helm. As you may have already guessed, it runs as part of the CI/CD pipeline that connects the HLD repository to the final manifest configuration repository!
Deployment Rings
I want to setup two environments, let’s say dev and prod for my application. But I want to run them on the same cluster to cut cost! I also want to be able to test my new features for the Trivia app in their own separate environments before I merge into dev or prod. Several customers have expressed interest in being able to test their features within an existing cluster before they make merge it into their next release.
This scenario calls for the concept of Deployment Rings — an encapsulation of a DevOps strategy to group your users into cohorts based on the features of your application you wish to expose to them, similar to A/B or canary testing but in a more formalized manner. Rings allow multiple environments to live in a single cluster with the help of a service mesh, by setting a header on their ingress routes. Each developer in the team would be able to test their features in their own rings before merging it into the formal production branch.
In the figure above, a single cluster has multiple deployment rings, and each of these rings is running off a branch in the source repository. For example, the production ring is deployed against the master branch and the dev ring is deployed against the develop branch. When a developer is testing their features before merging into develop , they're able to deploy a ring into the cluster, let's say feature_ring_1 and get feedback from their team before merging it into develop . Once the feature gets into develop , they can go through the testing process, and make it to staging , and then finally into master for production.
Connecting all the pieces together
Using Bedrock CLI, I created a High Level Definition for my Trivia app and hooked up the CI/CD pipelines for the repositories together. | https://samiyaakhtar.medium.com/gitops-observability-visualizing-the-journey-of-a-container-5f6ef1f3c9d2 | ['Samiya Akhtar'] | 2020-11-25 02:30:52.646000+00:00 | ['Kubernetes', 'Azure', 'Continuous Delivery', 'DevOps', 'Gitops'] |
The 90's Trend Hits the World Wide Web, Again. | An ode to the 90's fad?
The 90's are back, in a big way.
Girls with false nostalgia for an era they were conceived in are rocking velvet dresses, tattoo necklaces, and yes, flannel.
But did you know that some sites are beginning to pick the 90's look back up again, too?
Right now it’s mostly limited to the stores that sell 90's clothing, or the personal webpages of Instagram darlings, but watch out — soon you might be asked to code something with scrolling or blinking text!
Here are a few examples from around the web:
Happy Monday’s store site (note the scrolling text!)
Shop Jeen’s store site, complete with music video autoplay and a custom tie-dye cursor.
O Mighty’s store page, with scroll, blink, and pig glitter.
And if that’s not meta enough for you, check out these clothes:
Windows 95 chic. From O Mighty, shown on the DollsKill website.
90's graphics galore! From DollsKill clothes.
From Happy Monday, the “Hot Butt Shirt”
Shop Jeen even named its line of risqué clothing Netgear90:
Tubular — NOT!
Nostalgia is a powerful emotion, and one that shouldn’t be ignored while building.
So to all you designers out there, dust off your HTML3 skills and figure out your Lisa Frank hexadecimals because there’s a new trend in town. | https://uxdesign.cc/the-90-s-trend-hits-the-world-wide-web-again-3d83227f60e3 | ['Kiki Schirr'] | 2017-01-12 02:38:31.442000+00:00 | ['Web Development', 'Design', 'Humor'] |
The End of Starsky Robotics | In 2015, I got obsessed with the idea of driverless trucks and started Starsky Robotics. In 2016, we became the first street-legal vehicle to be paid to do real work without a person behind the wheel. In 2018, we became the first street-legal truck to do a fully unmanned run, albeit on a closed road. In 2019, our truck became the first fully-unmanned truck to drive on a live highway.
And in 2020, we’re shutting down.
I remain incredibly proud of the product, team, and organization we were able to build; one where PhDs and truck drivers worked side by side, where generational challenges were solved by people with more smarts than pedigree, and where we discovered how the future of logistics will work.
Like Shackleton on his expedition to Antarctica, we did things no one else ever has. Similarly, though, it didn’t turn out as planned.
Much of Starsky office team circa Feb 2019. Nothing in my life has made me as proud as getting to work with this incredible team.
So what happened?
Timing, more than anything else, is what I think is to blame for our unfortunate fate. Our approach, I still believe, was the right one but the space was too overwhelmed with the unmet promise of AI to focus on a practical solution. As those breakthroughs failed to appear, the downpour of investor interest became a drizzle. It also didn’t help that last year’s tech IPOs took a lot of energy out of the tech industry, and that trucking has been in a recession for 18 or so months.
The AV Space
There are too many problems with the AV industry to detail here: the professorial pace at which most teams work, the lack of tangible deployment milestones, the open secret that there isn’t a robotaxi business model, etc. The biggest, however, is that supervised machine learning doesn’t live up to the hype. It isn’t actual artificial intelligence akin to C-3PO, it’s a sophisticated pattern-matching tool.
Back in 2015, everyone thought their kids wouldn’t need to learn how to drive. Supervised machine learning (under the auspices of being “AI”) was advancing so quickly — in just a few years it had gone from mostly recognizing cats to more-or-less driving. It seemed that AI was following a Moore’s Law Curve:
Source: TechTarget
Projecting that progress forward, all of humanity would certainly be economically uncompetitive in the near future. We would need basic income to cope, to connect with machines to stand a chance, etc.
Five years later and AV professionals are no longer promising Artificial General Intelligence after the next code commit. Instead, the consensus has become that we’re at least 10 years away from self-driving cars.
It’s widely understood that the hardest part of building AI is how it deals with situations that happen uncommonly, i.e. edge cases. In fact, the better your model, the harder it is to find robust data sets of novel edge cases. Additionally, the better your model, the more accurate the data you need to improve it. Rather than seeing exponential improvements in the quality of AI performance (a la Moore’s Law), we’re instead seeing exponential increases in the cost to improve AI systems — supervised ML seems to follow an S-Curve.
The S-Curve here is why Comma.ai, with 5–15 engineers, sees performance not wholly different than Tesla’s 100+ person autonomy team. Or why at Starsky we were able to become one of three companies to do on-public road unmanned tests (with only 30 engineers).
It isn’t incredibly unprecedented — S-curves are frequent in technological adoption (Moore’s Law is actually made up of a number of S curves as chip technologies continuously replace each other as the best candidate to continue the phenomenon’s overall curvature). The problem is when try to compare the current technology how good humans are at driving. I’d propose that there are possible options: we’ve already surpassed human equivalence (show below as L1), we’re nearly there (L2), or we’re a ways off (L3).
If L1 is the line of human equivalence, then leading AV companies merely have to prove safety to be able to deploy. I don’t think I know anyone serious who believes this, but it is a possibility. If L2 is the case, the bigger teams are somewhere from $1–25b away from solving this problem. When big AV investors say that autonomy is an industry just for big companies, this is the bet that they’re making. If, however, L3 is the line of human equivalence it’s unlikely any of the current technology will make that jump. Whenever someone says autonomy is 10 years away that’s almost certainly what their thought is. There aren’t many startups that can survive 10 years without shipping, which means that almost no current autonomous team will ever ship AI decision makers if this is the case.
There aren’t many startups that can survive 10 years without shipping
Why We Didn’t Survive
To someone unfamiliar with the dynamics of venture fundraising, all of the above might seem like a great case to invest in Starsky. We didn’t need “true AI” to be a good business (we thought it might only be worth ~$600/truck/yr) so we should have been able to raise despite the above becoming increasingly obvious. Unfortunately, when investors cool on a space, they generally cool on the entire space. We also saw that investors really didn’t like the business model of being the operator, and that our heavy investment into safety didn’t translate for investors.
Trucking Blues
If teleop solves half the challenge of autonomy, the other half is solved by being the operator. As the trucking company you can choose where you operate — allowing you to pick your battles. Your system only has to be safe on the routes and in the conditions you choose to drive in (on the easiest routes and pulling over and waiting in bad conditions).
The nature of the participants in the trucking industry also reinforces the decision to be an operator. Trucking companies aren’t great technology customers (you should see what they use), and no one knows how to buy safety-critical on-road robots. Even if Starsky perfected general autonomy and perfectly validated safety, it would take years to deploy sufficient systems to make the necessary profits.
“You can always tell how serious a company is about unmanned by how seriously they talk about teleop” a vendor once told me. Nevertheless, we found an incredible amount of industry and investor resistance to our teleop-dependent approach.
While trucking companies don’t know how to buy safety critical robots, they do know how to buy trucking capacity. Every large trucking company does so — their brokerages buy capacity from smaller fleets and owner-operators, many of whom they keep at an arm’s length because they don’t know how much to trust their self-reported safety metrics. At Starsky we found 25+ brokers and trucking companies more than willing to dispatch freight to trucks they already suspected were unmanned. While this is a lower margin business than software’s traditional 90%, we expected to be able to get to a 50% margin in time.
It took me way too long to realize that VCs would rather a $1b business with a 90% margin than a $5b business with a 50% margin, even if capital requirements and growth were the same.
And growth would be the same. The biggest limiter of autonomous deployments isn’t sales, it’s safety.
No One Really Likes Safety, they like Features
In January of 2019, our Head of Safety, our Head of PR, and I gathered in a conference room for a strategy session. The issue: how could we make safety seem exciting enough to cover. A month earlier we had publicly released our VSSA, a highly technical document that detailed how we decided to approach safety. We had pitched it to a particularly smart reporter, but instead of covering it in detail they mostly wrote about teleoperation. We left the meeting in a fluster — we couldn’t figure out how to make safety engineering sexy enough to garner its own reporting.
And we never really figured out how.
Ironically, we were planning on launching a fleet of 10 v2 trucks by January 2020. These systems were designed to be consistent enough to enable us to prove safety across the broader fleet, allowing unmanned regular service by June 2020.
The problem is that people are excited by things that happen rarely, like Starsky’s unmanned public road test. Even when it’s negative, a plane crash gets far more reporting than the 100 people who die each day in automotive accidents. By definition building safety is building the unexceptional; you’re specifically trying to make a system which works without exception.
Safety engineering is the process of highly documenting your product so that you know exactly the conditions under which it will fail and the severity of those failures, and then measuring the frequency of those conditions such that you know how likely it is that your product will hurt people versus how many people you’ve decided are acceptable to hurt.
Doing that is really, really hard. So hard, in fact, that it’s more or less the only thing we did from September of 2017 until our unmanned run in June of 2019. We documented our system, built a safety backup system, and then repeatedly tested our system to failure, fixed those failures, and repeated.
Credit: Tanya Sumang
The problem is that all of that work is invisible. Investors expect founders to lie to them — so how are they to believe that the unmanned run we did actually only had a 1 in a million chance of fatality accident? If they don’t know how hard it is to do unmanned, how do they know someone else can’t do it next week?
Our competitors, on the other hand, invested their engineering efforts in building additional AI features. Decision makers which could sometimes decide to change lanes, or could drive on surface streets (assuming they had sufficient map data). Really neat, cutting- edge stuff.
Investors were impressed. It didn’t matter that that jump from “sometimes working” to statistically reliable was 10–1000x more work.
So, what’s next?
Around November 12 of 2019, our $20m Series B fell apart. We furloughed most of the team on the 15th (probably the worst day of my life), and then started work on selling the company and making sure the team didn’t go without shelter (or visa status, or healthcare for the new and expectant parents).
We were able to land many of the vulnerable jobs by the end of January and I’m in the process of selling the assets of the company (which includes a number of patents essential to operating unmanned vehicles). Like the captain of a sinking ship, I’ve gotten most of the crew on lifeboats and am now noticing the icy waters at my ankles while I start to think about what I do next.
From my vantage point, I think the most likely line of human equivalence is L3 which means that no one should be betting a business on safe AI decision makers. The current companies who are will continue to drain momentum over the next two years, followed by a few years with nearly no investment in the space, and (hopefully) another unmanned highway test for 5 years.
I’d love to be wrong. The aging workforce which will almost certainly start to limit economic growth in the next 5–10 years; the 4000 people who die every year in truck accidents seem a needless sacrifice. If we showed anything at Starsky, it’s that this is buildable if you religiously focus on getting the person out of the vehicle in limited-use cases. But it will need to be someone else to realize that vision.
Signing off,
Stefan. | https://medium.com/starsky-robotics-blog/the-end-of-starsky-robotics-acb8a6a8a5f5 | ['Stefan Seltz-Axmacher'] | 2020-03-19 16:19:44.522000+00:00 | ['Autonomous Vehicles', 'Autonomous Cars', 'Tech', 'Robotics', 'Startup'] |
For All Mankind Season 2 Coming Soon To Apple TV+ | For All Mankind — Apple TV+
Apple has recently announced that their critically announced sci-fi drama series will be renewed for a third season ahead of the release of the third season which will premiere globally on February 19, 2021.
The series explores what would have happened if the global space race had never ended. The series presents a world where the USSR beats the US to the moon, and we follow a group of NASA astronauts, engineers and their families through their extraordinary lives in this alternate history timeline.
Apple has already released some details about the storyline for the second season. We have been told that the season will pick up a decade later in 1983. This is a time where tensions between the USSR and the USA are at their peak, as it is the height of The Cold War.
The series takes place with Ronald Reagan as President and the increasing expansion of space exploration is under threat as the US and the USSR to control different sites on the moon.
Season 1 stars Joel Kinnaman, Michael Dorman, Sarah Jones, Shantel VanSanten, Wrenn Schmidt, Jodi Balfour, Krys Marshall, and Sonya Walger will all reprise their former roles from the first season. There are also some new cast members that will be joining the series will include Cynthy Wu, Coral Peña, and Casey W. Johnson. | https://medium.com/macoclock/for-all-mankind-season-2-coming-soon-to-apple-tv-c5152b3c665a | ['Yash Patak'] | 2020-12-22 07:28:36.303000+00:00 | ['Netflix', 'Apple TV', 'TV Series', 'Apple', 'Streaming'] |
The RSI-MACD Technical Indicator — A Python Study. | The strategy does not seem to perform well. Apparently, there is not much to do when it comes to combining these two indicators. After all, not much good can come from applying a lagging indicator on another lagging indicator.
One thing to mention before we end the discussion is that the hit ratio presented may seem confusing. How can such a high hit ratio give us mostly negative results?
Well, as I have mentioned in my previous articles, the hit ratio does not mean anything without the risk-reward ratio. We can notice that for all the pairs, the risk-reward ratio was extremely low (around 0.28).
Sure, a 70–80% hit ratio on your trades is great. But what if you are risking $1 each time to earn $0.20 (20 cents)? Well, then you will lose money and get wiped out because your risk reward will be 0.2. If you make 100 trades where you always use the same position sizing and you get your 80% hit ratio which translates to 80 profitable trades with each gaining $0.20.
This gives you a profit of 80 x 0.2 = $16. Alright, not bad but let us see the remaining losing 20 trades which have lost $1 each. This gives you a loss of 20 x 1.0 = $20. Your net profit is therefore -$4.00. Hence, by getting it right 8 out of 10 times, you have managed to lose money. Be careful from high hit ratio strategies which are generally quoted first when people are selling you theirs. Always check the risk metrics.
Here is the same strategy but with a theoretical risk-reward ratio of 2.00. This means for every trade, the stop will be 1x ATR and the target 2x ATR. Financially speaking, this is how risk management should be at its most basic form. | https://medium.com/python-in-plain-english/the-rsi-macd-technical-indicator-a-python-study-71cf949707eb | ['Sofien Kaabar'] | 2020-12-28 12:38:53.817000+00:00 | ['Trading', 'Finance', 'Machine Learning', 'Data Science', 'Artificial Intelligence'] |
Nobel Prizes 2020: Roger Penrose and Black Hole Formation | Sir Roger Penrose, author of the award-winning The Emperor’s New Mind: Concerning Computers, Minds, and the Laws of Physics (part of the Oxford Landmark Science series) and his own six-part series of collected works, has been jointly awarded the 2020 Nobel Prize in Physics for his discovery that black hole formation is a robust prediction of the general theory of relativity.
Announcing the Prize, The Royal Swedish Academy of Sciences wrote of the “ingenious mathematical methods” used by Penrose in his proof that black holes are a direct consequence of Albert Einstein’s general theory of relativity, going on to state that “his groundbreaking article is still regarded as the most important contribution to the general theory of relativity since Einstein.”
Sir Roger shares the award with Reinhard Genzel and Andrea Ghez, who have been recognized for their discovery of a supermassive compact object at the centre of our galaxy
If you would like to understand more about black holes and their place within physics, take a look at our reading list:
Roger Penrose, Oxford Landmark Science series
Clifford M. Will and Nicolás Yunes
John W. Moffat
Nicholas Mee
Alvaro De Rújula
Katherine Blundell | https://medium.com/science-uncovered/roger-penrose-and-black-hole-formation-122c9357a3ec | ['Oxford Academic'] | 2020-10-14 15:02:57.656000+00:00 | ['Black Hole', 'Physics', 'Science', 'Oxford University Press', 'Nobel Prize'] |
How a 100-day doodle project helped me rekindle my childlike creativity | It all started when my older sister said that I should start drawing to pass the time during the second UK Covid lockdown in 2020. It was May and I was feeling extra tired and sick coping with my first trimester. The idea of having to stay cooped up for another 12 weeks left me feeling sad and glum.
So following my sister's advice, I decided to do a quick sketch from a photograph that I loved. But I quickly became frustrated with my attempt and I was left unhappy and at odds with myself. In drawing the faces from the photograph, in my eyes I had drawn deformed monsters!
Exasperated, I put my sketchbook down. I left believing that I wasn’t able to draw.
A few days later, my 5-year-old niece came to stay. She absolutely loves drawing. Watching her draw, she had so much passion, enthusiasm, and joy.
When she proudly showed me her final picture there was no doubt in her mind what she had drawn. It was then that I had realised what the problem was; I had simply grown up, and instead of feeling pride I was filled with self-criticism.
My 5-year-old niece with her drawing
“Every child is an artist, it is only as we get older that we believe we are not.” — Picasso
Deciding not to put my pen down
Instead of putting down my sketchpad for another few years, I decided to make a change. I recalled a blogger named Tian Tian Xu, who set herself 100-day challenges to learn something that she wanted to get better at. Inspired by her progression, I felt that this would be a great way for me to teach myself to be able to draw again.
I started with a few ground rules to help me keep this challenge going. This loose framework helped me avoid feeling defeated on the days where I wasn’t very motivated.
Doodle every day, but don’t be too precious if you miss any days. If you don’t know what to doodle then find inspiration from Pinterest. Don’t get stuck in the details, it’s more about the process rather than perfection. Keep a visual diary to help you keep track of your drawings.
It’s ok to copy
The image on the left of the French bulldog found via Pinterest, Image on the right is my version
In the beginning, I didn’t really know what to doodle. This is where rule number 2 came in handy. Knowing that copying wasn’t giving up because it was part of my process really helped motivate me.
Turns out, copying drawings is much more inspiring than you’d think! It took the pressure off how something should look in my head. I found this simpler and was much happier with the outcomes.
Stop comparing yourself to others
Every time I would draw a picture, my internal voice would tell me how bad it was. I felt self-conscious, telling myself that I would never be able to draw as well as illustrators like Malika Favre. I couldn’t help but compare my drawings to hers.
How I want to draw faces vs How I actually draw faces
To combat this self-doubt, I reframed my thinking. It wasn’t about creating something art-worthy, it was about creating a doodle that day, whether it took a minute or half an hour and sometimes looked unfinished.
By alleviating the pressure I had put on myself, I was able to draw more freely and enjoy the process rather than the outcome. In fact, looking back at my drawings, on the days when I felt I wasn’t drawing my best I actually produced some of my more interesting drawings.
Doodles on days when I really wasn’t feeling it
Creating a habit
When Jerry Seinfeld was an up-and-coming comedian, he made a commitment to write one joke a day. He had a big calendar of the whole year on a wall in his apartment and every time he wrote a joke, he put a red X on that date.
When starting a new challenge — especially a 100 day one — it can feel intimidating looking at the 100 days ahead of you. Instead, it’s better to break it down into small, manageable amounts.
For me personally, I prefer to save things as a visual picture instead of marking a red cross on a calendar. I created an Instagram account and called it @hannahdoodled. This helped me to see my progress, like a patchwork quilt I wanted to complete by creating the next square.
It wasn’t actually until about day 6o that I realised doodling had become a habit. Whether you want to draw, write, bake, exercise… Simply break it into smaller steps, then one day it will turn into a habit.
Creating my doodle grid on Instagram
You are unique and that is great
Originally my @hannahdoodled Instagram account was set as a private account. But after posting for a few weeks, my friend asked me why I was keeping it private. I usually love sharing my photography on Instagram. My main fear was that people would judge my drawings, hence why I wanted to keep them private.
Looking at other illustrators that I followed and loved, I noticed how different they all were. Yet I loved all of them for the uniqueness. Feeling better about my drawings, I decided to make my account public, allowing people to see and comment. In fact, it also made me feel accountable and encouraged me to post.
Various illustrators that I follow
It’s more than just a doodle
As I carried on doodling, I realised I was getting better at drawing things with ease. I was starting to recognise my drawings as my drawings and not expecting them to look like an illustrator that I loved. The habit of drawing every day felt satisfying: I was able to see things that I would want to draw, and quite often later I would do a little doodle or sketch in my book.
Practice can make perfect, but for me, I found that practice and consistency helped me to see the work for what it is; I was accepting my drawing style whilst building on it.
Even if it’s stick men, then become really great at getting your ideas across in the simplest stickman format. Whatever you might be interested in, definitely consider doing a 100-day challenge. | https://medium.com/ideas-by-idean/how-doing-a-100-day-doodle-project-helped-me-get-back-to-being-as-artistic-as-a-child-f90ee05e10fe | ['Hannah Bain'] | 2020-12-23 13:31:18.606000+00:00 | ['Design', 'Drawing', '100DayChallenge', 'Illustration', 'Doodle'] |
My Coming Out Story. | My Coming Out Story.
Not all Stories are Negative.
I would love to go back and tell the young Kevin to accept who you are, it will be hard for you to accept but ultimately it will be the best thing that you ever do.
Rainbow heart. Photo by @jiroe
I was talking to a friend at work about the articles I have written and he said something along the lines of ‘yeah I’ve read them, homophobia, homophobia and the straights hate the gays’. That got me thinking about my articles and he’s right, the majority of the articles I have written are directly or indirectly about homophobia. I have written that word so much that when I type the word ‘homo’ it automatically wants to finish the word with ‘phobia’. So here is the thing, I have been out for over ten years now and I haven’t suffered any homophobic abuse. I thought in this article I could write my own experience, a positive experience of what happened when I came out and what happened since.
After hiding my sexuality to myself, and to everyone else since I was around 11, at 22 years old (relatively late but I got there) I came out. It was a Sunday afternoon when I knew everyone would be at home and I asked everyone to come into the living room as I had something that I wanted to talk about. I was so nervous and the speech I had planned went out of the window but I told them I was in a relationship with a guy, probably not what they had expected. I said I was bi, I think it made it easier for me to handle. I understand how this can be bad for people who identify as bisexual, people don’t believe that you can like both sexes but I genuinely thought I did at that time.
My parents were shocked which was understandable but they said that they loved me no matter what. I think my dad found it hard to get his head around, he was from the building trade, he used to go to the pub most nights, smoked cigarettes most of his life, he was a mans man but I was his son and even though it was a world he didn’t know he never said anything bad about my sexuality. My Mum was accepting from the start, but I know it took time for her to get over the shock but she never let it on to me. My brother right from the start was awesome, it must have bothered him that he didn’t guess when he was the closest person to me but he accepted it straight away.
Friends Drinking at the Pub. Photo by @elevatebeer
I had told my immediate family but I wasn’t quite sure about the rest of my family yet so I kept my sexuality hidden from them but I did tell my friends. I was a little bit sneaky how I went about this, I changed my status on Facebook to ‘in a relationship’, I knew this would get questions, my friend Vicki was one of the first to respond asking who this mystery girl was that I was now in a relationship with. I then told her that it wasn’t a girl. I knew that telling her would mean that the next night out would force me to tell everyone. So on a traditional Friday night gathering, sat together at the pub I came out to them. I couldn’t get my words out so I think my brother told everyone, that detail is a little hazy. They were awesome right from the start, I then went about telling my friends who weren’t in that friendship circle and again they were perfect and accepting.
A couple of weeks after I came out one of my friends also did, I think that he saw the reaction I got and it encouraged him, he saw that everyone was ok with me and that he would get the same. The relationship I came out for didn’t last long (if you can call it a relationship) but I’m thankful for it, it forced me to come out and encouraged someone else to do the same. Nine months after I came out I met Stu, we had spoken online for a while but he lived near Birmingham (the one in the UK) he moved nearer to me so we finally met up. Five days later we were ‘official’, 6 weeks later we were travelling around the world together, quick but it felt right.
My extended family eventually worked out that Stu and I were a couple and he wasn’t just “my friend from school”, maybe it was the Brummie accent that gave it away. To be honest I have no idea why I didn’t tell them before, once again they accepted my sexuality and treated our relationship just like any other. I absolutely love the relationship my mum and Stu have, he even stayed with my mum the night before we got married. It really shows how far things have come from her initial shock to being friends with my other half.
Stu and I have been together now for ten years and just over two years ago we got married. The day was awesome and we shared it with our friends and family, it was the best day ever! There was only one family member who didn’t agree with our marriage, who didn’t come to the wedding but as it was someone I’m not that close to. I was and still am annoyed but it wasn’t the end of the world. The people I really care about were there and made the day special.
Friends and family at our wedding!
I was so worried about people’s reaction to my sexuality that I buried it so deep that I wouldn’t let myself accept it. I found myself looking at boys and hating myself for it, I tried to force myself to look at girls and find them attractive. It doesn’t work and it only makes it harder to accept, you know you can’t change it but you want to. It wasn’t until Stu and I had been together for a while that I accpeted I was gay and not bi.
After coming out I realised I wasted over ten years of my life worrying and battling but it was pointless. I would love to go back and tell the young Kevin to accept who you are, it will be hard for you to accept but ultimately it will be the best thing that you ever do. Having said that if I came out earlier I might not have met Stu and have the life I have now and I love my life.
I have tried to make this a positive piece, I have admitted the difficulties in coming out but I wanted to show that being gay isn’t all homophobia, homophobia and some more homophobia. I haven’t suffered any abuse at all. I know I’m a white boy from the south of England so in that respect it is relatively easy for me. There is every chance that my family and friends wouldn’t accept me and that could happen in any place and for any race. For me the grass was definitely greener on the other side and if you are thinking of coming out, it could be for you too. | https://medium.com/lgbtgaze/my-coming-out-story-41962b44fde4 | ['Kevin Laurie'] | 2018-10-04 19:19:37.345000+00:00 | ['Storytelling', 'Sexuality', 'Lifestyle', 'Coming Out', 'Love'] |
Spyder IDE and Command Line: The Basics | Learning the basics of Spyder and Command Prompt has got to be an essential step for your Data Science Journey. So, it is best to get an in-depth knowledge of at least 2 Data Science tools because it will surely step up your game! To learn more about these available analysis tools in the market, you might want to check the article below: | https://medium.com/python-in-plain-english/spyder-ide-and-command-line-the-basics-b66aa8277ff8 | ['Angel Mariano'] | 2020-10-01 12:46:55.230000+00:00 | ['Python', 'Command Line', 'Spyder', 'Data Science', 'Programming'] |
How Persuasive Design is Not a Process with Anders Toxboe | A good friend of mine, Andrew Chak, wrote Submit Now! in 2002, that amongst others deals with a persuasive design process and persuasive persona’s. That’s something I don’t hear much about anymore.
Is there such a thing as an ideal persuasive design process? (Agile Persuasion has a nice ring to it BTW ;)
I really don’t see persuasive design as a process. It is a set of ideas and insights that will help enlighten your thinking. Persuasive design will aid you in making informed best guesses as to why and how users are behaving.
As designers, we can use those insights to suggest features and create hypotheses to validate and test. There are no ultimate thruths. There is a sound possibility that what works in one context might not work in another. This is why you need to test whatever ideas persuasive design might suggest, rigorously.
At the start of a project, you try to grasp its entirety. As work gets broken down into sprints, it can be easy to see the backlog as a massive pile of “work to be done”. This is a typical trap of agile development. When you start to see your development as work to be done, you quickly get a focus on “What” and “When” to build — rather than “Why”. Too much focus on what and when to build without asking why, creates tunnel vision.
Agile is great for helping teams focus on executing incremental change, but it can also blind you. Relying on agile alone, in my experience, too often neglects the overall goals and visions. In agile, success is often measured in feature completions rather than business improvements. For agile to be effective in accommodating broader innovation, it needs to be combined with methods that question the assumption of the status quo.
I’ve personally found the Lean Startup approach particular helpful as a process to support persuasive design.
Test your prototypes or final product with your users. Use A/B tests, landing pages, feature stubs, handheld Wizard of Oz features, clickable prototypes, guerilla user tests, and user interviews to validate your assumptions.
Persuasive design is great, but it’s not a process.
In a sense using cards in brainstorming sessions could be considered a process ;) it resembles the way we worked at Booking.com. Some time ago I wrote an article on the importance to look at customer journeys in detail and plotting micro persuasion moments.
The insights of the persuasive pattern cardset can be used for new business concepts, as well as for details of website design and anything in between. I would say that are very different problems with a different level of solution to them.
What are the kind of problems you see fit best with this approach?
Depending on whether you’re innovating (new business concepts) or optimizing (details of a design), you need to choose appropriate measures of testing.
If your product doesn’t have real users yet, it’s hard to conduct meaningful quantitative A/B tests or create feature stubs to collect insights on your product hypothesis. However, you can still do early validation. Whether you’re validating the problem you’re trying to solve, the market, or the product, you can still listen to your users. Create landing-page tests, prototype tests, or just conduct interviews with potential real users. In the early stages, the important thing to remember is that you need to solve real problems and in order to find that problem, you need to listen to real people.
All of this has to do with your design process rather than on how you apply persuasive patterns. Persuasive patterns work on all levels — they can form a whole business concept or constitute a small design detail. As an example, consider the persuasive pattern of Scarcity. The whole concept of Groupon.com was formed around offers being available only for a limited time, whereas travel sites such as Booking.com use the principle of Scarcity thoroughly in small design details designed to nudge us to book, letting us know that a hotel has been “Booked 2 times today” or that “8 users have been looking at this hotel in the last hour”.
As you mentioned, persuasive patterns work on multiple levels, and you need to adjust your approach accordingly.
I like to say that persuasion is like a relationship. And that you need to treat persuasion like a relationship. The ultimate question then becomes: Do you want a one night stand or a lasting relationship? Your approach to engaging users should be appropriately adjusted to the relationship you have with them. Have you just met them, are you trying to get to know them, or are you building on an already long-lasting relationship?
And could you walk through an ideal project where the Persuasive Patterns were used?
There are no ideal projects, but some projects are easier to comprehend for designers than others. When I conduct workshops for companies developing enterprise business applications, they often ask me to skip through all of the “selling stuff”. They explain how they don’t face the same problems as B2C companies do. I agree. They don’t face the same problems, but the same patterns can easily help better both B2C and B2B products.
Persuasive design can both nudge and help facilitate easier decision making. Persuasive Patterns like the Status-Quo Bias, Recognition over Recall,Intentional Gaps, Limited Choice, Tunneling, Reduction, Sequencing,Completion, and Social Proof will all help reduce cognitive load in both B2C and B2B products.
I’ve seen that it tends to be harder for designers to apply persuasive design to business applications. Instead, they start with concrete UI design patterns and work their way up to their persuasive pattern counterpart:
The Good defaults UI Pattern is supported by the Status-Quo Bias persuasive pattern
The Progressive Disclosure UI Pattern is supported by the Limited Choice persuasive pattern
The Wizards and Steps Left UI Patterns are supported by the Completion and Sequencing persuasive pattern
The Forgiving Format UI Pattern is supported by the Reduction persuasive pattern
The Fill in the Blanks UI Pattern is supported by the Intentional Gaps persuasive pattern
The Autocomplete and Calendar Picker UI Patterns are supported by the Recognition over Recall persuasive pattern
The list goes on.
I am yet to discover a design problem where persuasive design failed to enlighten the discussion. Over the last 2 years, I’ve helped companies implement persuasive design for:
Landing page and conversion rate optimization
Reducing cognitive load in business applications
Redesigning onboarding flows
Facilitating learning through gamification-systems through badges, points, and building communities
Helped stimulate content creation and moderation in community sites
I’ve understood that you tested the scientific insights and their translations into the cards extensively. Having scientific sound insights is one thing, but it’s another thing completely to translate those insights in business smart experiments.
How do you see the relationship science, business and design?
Together with my teams, we’ve conducted a multitude of A/B tests that have both proved the success and failure persuasive patterns. Whether they work or not completely depends on the context in which it is used and the context of the user. All the patterns in the Persuasive Patterns card deck have proven to work in various situations, but none for all situations.
The science of psychology or persuasive design will help suggest informed and qualified best guesses. Insight into psychology and UI design patterns will help spark new and useful ideas that you can then test.
If you fail to test the features and ideas that persuasive patterns can help you dream up, you might as well not use them at all. Then they will be a waste of time for your business.
However, once you’ve applied the same persuasive patterns in similar situations and seen it work, your confidence in that pattern grows. If we’ve seen a solution work hundreds of times and have the data to back it up, we just implement it. If we’re not certain enough, we conduct A/B tests and experiments to either increase our certainty or make us drop the idea.
If you want to get into Persuasive design, you can get hold of your own Persuasive Pattern card deck at the UI Patterns web shop.
Do you like what you have read? Get a quarterly update of what I am busy with. | https://medium.com/i-love-experiments/how-persuasive-design-is-not-a-process-with-anders-toxboe-210617f0c83e | ['Arjan Haring'] | 2016-08-26 03:47:05.241000+00:00 | ['Persuasive Design', 'Conversion Optimization', 'Design', 'Behavior Change', 'UX'] |
Syrg — Future Employee Relationship Management (FERM) Solution | Syrg — Future Employee Relationship Management (FERM) Solution
Allowing F500 corporations and franchisees to quickly hire new hourly workers from untapped applicant pools and reduced turnover
The Problem
Hourly workplaces have notoriously high turnover rates, which for quick service restaurants is around 132% per year, according to Quick Service Magazine. Hourly employers are therefore constantly hiring new people and spending thousands of dollars to train them, only to see their employees leave a few weeks or months later. This has been a problem for decades, but fear of infection by COVID-19 and high unemployment benefits has only made things harder, even as unemployment rates rise.
What The Company Does
Syrg helps hourly employers get ahead of their hiring needs and stabilize their workforces. They do that by creating engaged, personalized, and growing talent communities for each customer, getting customers great hires today and a place to easily source from in the future. Talent communities consist of customers’ best former, seasonal, and student employees, as well as past applicants and referrals, and receive regular updates ranging from open opportunities to exit and insight surveys. The results are simple: customers no longer need to scramble for hires every time someone leaves and instead can focus on growing their business for the long-haul.
The Market
Today, Syrg focuses mostly on helping franchise operators, which represent a $960M opportunity in the U.S. alone. Syrg is the only software company that proactively cultivates relationships with potential future hourly employees, and as a result, has never gone head to head with another software vendor. Syrg’s prospects evaluate Syrg against hiring recruiters and using temp agencies, against which Syrg has countless points of differentiation. In addition to being significantly cheaper than either option, Syrg cultivates relationships with and generates candidates from tens of thousands of potential future employees simultaneously. This means that Syrg is able to produce a more consistent flow of higher qualified candidates at a fraction of the cost of any alternative.
Business Model
Syrg is a software-as-a-service business that charges an annual subscription fee. The fee is determined by the number of employees a customer has at each one of their locations and how many total locations they have. If you’re interested in a specific quote, please reach out to [email protected] to set up a brief intro call.
Traction
Syrg publicly launched Future Employee Relationship Management in June 2020 and has since signed almost 15 new customers in just 4 months. Over the same period, Syrg participated in OneValley’s (fka GSVlabs) Pioneer Pitch Night: Black Founders Edition, where it won first place. Syrg is also a certified technology partner of Ultimate Kronos Group.
Founding Team Background
CEO Rahkeem Morris and COO Rob Snyder met at Harvard Business School in 2017. Before getting his MBA and working at Google, Rahkeem was a high school dropout working two to three hourly jobs at a time. As Rahkeem bounced from job to job, he realized his employers were always short-staffed and scrambling for hires. Rahkeem knew he’d be willing to work for any of his previous employers under the right circumstances and found it strange they never reached out to him proactively. Syrg’s FERM is the first step in Rahkeem’s vision for transforming hourly work and recruiting. Before business school and meeting Rahkeem, Rob was a McKinsey consultant focused on frontline operational transformations and turnarounds. He became interested in the gig economy and the future of work after experiencing it firsthand as a consultant with Boston-based Catalant (fka Hourly Nerd).
What They Need Help With
If you’re involved in an hourly workplace that struggles with hiring — particularly if you’re a quick service restaurant franchisee — Syrg would love the opportunity to understand your business and explore if they can get you similar results to their rapidly growing customer base. Syrg is also always interested in hearing from hardworking, talented individuals passionate about helping hourly workers and their employers. Syrg loves working with great people and is open to interviewing across all levels and functions, with a particular interest in hardcore engineers considering manager roles and customer success professionals. The company is also always looking to partner with great investors passionate about its vision. Connect With The Syrg Team.
Subscribe To The Buzz To Get More Startups In Your Inbox | https://medium.com/the-startup-buzz/syrg-future-employee-relationship-management-ferm-solution-41ae51c08965 | ['Bram Berkowitz'] | 2020-10-27 19:02:40.743000+00:00 | ['Gig Economy', 'Startup', 'Restaurant', 'Venture Capital'] |
No, Having a Niche Is Not Important on Medium | No, Having a Niche Is Not Important on Medium
What you should do instead
Photo by Denis Oliveira on Unsplash
Every blogger who hopes to be successful has to choose a niche. It’s the first essential step in becoming successful. Choosing a niche isn’t simple. It has to be something you are interested in writing about — imagine having to spend hundreds of hours writing about something you simply didn’t care about. It should center around a popular topic while being ‘different’ enough for others to want to read your blog over another one.
Even though choosing a niche doesn’t guarantee success, not choosing one makes it difficult to market and monetize your blog. That’s why when writers begin to write on Medium, many continue to believe that a niche is the key to success. However, as someone who reads and writes a lot on Medium, I believe writing on Medium is vastly different from running your own blog.
The situation here at Medium
Unlike a personal blog, Medium already has an audience available for all writers. And every reader on Medium is different. Some spend all their time reading stories about writing better. Others might choose to read about politics. Another group could spend their time on Medium reading articles about astrology.
Fiction, non-fiction, climate change, mental health, psychology, technology, societal issues, productivity hacks, lifestyles, sports — the list of available topics people read about on Medium goes on and on.
This means that no matter what you decide to write about, there will likely be an audience already out there waiting.
Unlike a blog, where you have to maintain a niche to encourage readers to return regularly, Medium’s existing audience means you don’t have to box yourself in and force yourself into a certain mold.
Instead of building a targeted audience group from scratch, Medium writers can easily tap into various publications to reach readers who might be interested. And after building up a large enough reader base, Medium writers will begin to stand out and become recognizable, even if they do not stick to specific topics.
Here on Medium, instead of forcing yourself into a popular niche, become your own niche.
How to succeed on Medium
On Medium, a fixed niche doesn’t guarantee success. Instead, passion is the key. Write about things that matter to you, subjects you care enough about to spend hours reading up on, and ideas you hope to be able to spread to all your readers.
With so many writers here, the Medium audience has become a discerning bunch. They can tell if an article was forced out simply because the topic is popular. This means you’ll have to make sure that you write from your heart to build a brand here.
Let your voice shine through consistently, and let that become your brand and your niche.
Without the limitation of a niche, Medium is a place where you can tell any story you like. Don’t tie yourself down by forcing yourself into a niche. Share all the stories you want and do so compellingly. And by continually writing from deep within, you’ll eventually find success as a good writer that loves to write.
So the next time you click on that “write a story” button, move away from the niche that’s holding you back. Write that story that you’ve been itching to tell and enjoy the process. Because no matter what you write about, there will be someone waiting to read. | https://medium.com/write-i-must/no-having-a-niche-is-not-important-on-medium-716b4d8bdaae | ['Jerren Gan'] | 2020-12-11 12:48:44.175000+00:00 | ['Writers On Writing', 'Ｍedium', 'Writing', 'Write', 'Niche'] |
How To Get People To Do Stuff #6: Hot drinks, soft pillows & heavy objects | Joshua Ackerman and John Bargh (2010) conducted research where they had candidates for job interviews hand in their resume one of three ways. One candidate handed in her resume on regular printer paper. Another candidate handed in her resume on regular printer paper, but had it clipped to a light clipboard. A third candidate handed in her resume on regular printer paper, but had it clipped to a heavy clipboard. Then they had interviewers rate which candidates were the best for the job. The interviewers gave higher ratings to candidates whose resume they were reading while the interviewer was holding a heavy clipboard.
Holding a heavy object while looking at a resume makes a job candidate appear more important. In fact, any idea you’re considering while holding something heavy (for instance, a book) you will deem to be more important. The metaphor of an idea being “weighty” has a physical corollary.
There are two terms that are used for this. Sometimes it’s called “haptic sensation” and sometimes you will find it referred to as “embodied cognition.” We are very influenced by the meaning that our sense of touch perceives.
You may be surprised to find out all the ways that these haptic sensations affect our perceptions and judgments. Besides the effect for a heavy object, people also react to these other haptic sensations:
• When people touch a rough object during a social interaction, for instance, if they’re sitting on a chair with coarse wool upholstery, they rate the interaction more difficult than if they touch a soft object.
• When people touch a hard object, they rate a negotiation as more rigid than if they touch a soft object.
• When people hold a warm cup (for example, a warm cup of coffee), they judge the person they’re interacting with to have a warmer personality than if they’re holding a cup of cold liquid.
You can use these haptic sensations to get people to do stuff. If you want people to have easier interactions with others, then you might want to have soft furniture, not hard chairs, in your conference room, and use a soft fabric covering for them rather than a scratchy tweed. If you have an important client coming to your office, and you want her to feel warmly about you, get her a cup of hot coffee or tea in a mug that will transmit the heat before you start.
Ackerman, Joshua M., Christopher Nocera, and John Bargh. 2010. “Incidental haptic sensations influence social judgments and decisions.” Science. 328 (5986): 1712–1715. DOI: 10.1126/science.1189993 | https://medium.com/theteamw/how-to-get-people-to-do-stuff-6-hot-drinks-soft-pillows-heavy-objects-173f1f9573c | ['The Team W'] | 2016-09-21 22:11:52.462000+00:00 | ['Decision Making', 'Research', 'Persuasion', 'Influence', 'Psychology'] |
14 Simple Ways to Be More Productive Every Day | How to Master Productivity in 14 Simple Steps
To improve productivity and achieve work efficiency, remember to “focus on the process.” This also means spending time on improving areas that are under your control as opposed to worrying about the impact of things beyond your control.
These 14 steps, while independent, build upon each other. Their power comes from small improvements done consistently over a period of time. By applying previous learnings and exploring new and better ways of doing things, their benefits compound over time, freeing up more energy to do meaningful work with less effort.
It sets the motivation flywheel in action by doing and seeing tiny progress in work that matters.
1. Build a simple habit of prepping the previous night
Photo by the author.
Spend a few minutes at the end of each day to prioritise what you want to accomplish the next day. It’s also extremely useful to divide work into chunks of different activities and outline a few bullet points on what you want to achieve in each chunk.
The simple act of writing things down not only boosts confidence by making us feel in control, but it also frees our mind to actually get the work done and avoid reacting to urgent work as opposed to doing important work.
While prioritising tasks, pick the ones that will move you in the direction of your goals while avoiding time-wasting activities that do not add value. Applying the Pareto principle, the 80/20 rule that states that 80% of results are achieved from 20% of our work, is a good way to achieve optimal prioritisation.
It’s also useful to employ the Eisenhower decision matrix for organising tasks in a four-quadrant view for easy reference and prioritisation.
This answers the What. We also need to determine the most optimal time during the day for each activity, which brings us to our next step for determining When.
2. Use your most energetic hours to do your best work
Each work unit demands different physical and mental resources. Learn what kinds of tasks need what kind of energy and organise for optimal utilisation by matching energy with the physical and mental demands of work.
Use your peak productivity period to do some of your most important work. It could be in the morning, afternoon, evening, or at night — whatever works for you. Different people work best at different times of the day.
I am an early riser, and a few hours after I wake up is the best time to do deep work. Knowing deep work compounds over time, it’s best utilised in doing work that matters.
Now that we know the What and the When, we have established a clear intent. Achieving work efficiency and productivity requires turning it into action, starting with creating optimal working conditions. But why?
3. Create a comfortable workspace suited to your personal needs
A lot of people do not realise that a poor working environment (something as simple as a cluttered work desk) can be a big source of stress and a productivity killer. Designing a comfortable workspace is not intuitive and seldom explored because we learn to live with it.
I understand that not everyone is privileged to get the best working environment of their choice, but consciously making an effort to improve it — even in small ways — can make a dramatic difference in the quality of your work.
Ask yourself these questions:
Do you have comfortable seating?
Are you using the right headphones?
Does the temperature suit you?
Are you hydrated while working?
Do you sit with a humped back or straight up?
Is your desk too cluttered?
Are you wearing comfortable clothing?
Our mind gets easily distracted and annoyed under suboptimal working conditions, thereby impacting productivity and taking away our ability to focus. Take control of your surroundings by making a conscious effort to observe what around you diverts your energy or makes you uncomfortable while doing your work. Then find ways to get rid of it.
Once you have the right working conditions, how do you achieve the right levels of thinking?
4. You may not need a break, but your brain and body do
Photo by the author.
At times when we are at an impasse, we get overwhelmed and anxious. This state limits our thinking to existing solutions and does not seek new ideas, causing us to stop making progress or take too long to finish a task.
What do we do when we reach this state? Most of us continue working while feeling frustrated, annoyed, and even angry with ourselves for not finding the right answers. Continuing to work when dealing with a mental block is not only less productive, but it may lead to bad decisions with severe consequences in the future.
We experience new ideas or better ways of doing things when we are not actively thinking about them. Giving space to our brain to form new connections is essential to discover new solutions and get over a hurdle.
Stand up, go grab a snack, get some fresh air, simply walk around without actively thinking about the problem. Great ideas will flow without requiring you to make an effort.
As a practice, it’s very useful to recover and refuel our mind and body with a small break after every slot of deep work (e.g. 60 mins, 90 mins) to re-energise and develop better levels of thinking for improved productivity.
Developing a better way of thinking is a great way to improve work productivity, but unless we also learn to control our thoughts, we may waste a lot of time with automatic mind wandering and distracted thinking that breaks our flow of work. How can we learn to control our own thoughts?
5. Don’t let your thoughts control you
David Rock sums this up beautifully in his book, Your Brain at Work:
“I decide to be responsible for my mental state instead of a victim of circumstances. In the instant that I make this decision, I start seeing more information around me, and I can perceive opportunities for feeling happier. This experience is one of finding a choice and making that choice, and it shifts what and how I perceive in that moment.”
The most important thing is to be aware that we all have the tendency to get carried away in thoughts, and while it may seem they are beyond our control, we can learn to manage them with practice.
Consciously make an effort to catch yourself in the act and bring the focus back to your work. Our mind will not stop running this constant stream in our heads unless we learn to recognise it and take charge of it.
It’s good to feel powerful by learning to say no to our own thoughts. But what about saying no to others? Isn’t saying yes to everything highly unproductive?
6. Learn to say no to things that don’t add value
Being productive demands saying no to work that does not align with your goals and mission. It’s not easy to say no, but it’s the only way to focus on work that really matters. Remember, productivity is not just doing the thing right. It is also about doing the right thing.
Once you get over your fear of saying no or desire to be liked, you will stop worrying about how your no may upset some people or how they might react. A carefully crafted no can enable both parties to engage in a more productive discussion and align on the outcomes.
There’s something else that can still derail you from your intended course. Emails?
7. Your emails can wait only if you resist the urge to open them
Photo by the author.
Emails can suck productivity if not managed well. Even one tiny peek can distract us from doing important work. The quick look is never really quick, as the image of those emails in our mind leaves less room for doing meaningful work.
It’s best to not open emails until noon. Schedule a 30-minute or one-hour slot before lunch and then another slot in the evening or at the end of the day. Whatever works best for you. It does require a lot of self-control to do what you had planned the previous night and prevent the false sense of urgency from taking over your day.
If something is really urgent, people in the organisation will find a better means to reach out to you.
Emails managed. What other distractions can hamper your productivity?
8. Eliminate distractions if you cannot resist the temptation
Focused work requires time away from distractions. Checking social media, looking at your notifications, and browsing the phone are time-wasting activities that can prevent you from achieving your goals.
Being aware of your distractions is the first step to removing them. If you are someone like me, adding friction can be a powerful way to block them out. I log out from all my social media accounts while doing important work. The simple barrier of having to provide login information prevents me from checking them too often.
Another strategy that works well for me is to schedule time to do these activities as part of my daily planning. Knowing that I have blocked time makes me more tolerant of the idea of not checking them out during my other planned work.
Once we manage to keep all distractions away and actually get down to work, what’s the best way to get started?
9. Develop simple routines to start your activities
The anxiety from the task’s unknowns may prevent us from getting started. “What if I am not able to complete it on time? Will I do it well? Am I competent? What if others don’t like my work?” The constant chatter in our mind even before we have started can prevent us from making any progress.
Breaking the barrier to getting started requires motivation. It doesn’t come from waiting for the right time or the right idea to click. Motivation is in small acts that move us in the direction of our goals.
Establishing routines can activate patterns in our brain that can overcome the mental block to begin. For my writing, I always begin by outlining key points for a new article or read what I wrote the previous day. This simple act kicks in the motivation to continue and do more.
Identify what routines work for you and use them as a hack to get started.
Once you actually get down to work, can you shift between multiple tasks and still be productive?
10. Doing many things at once is as good as doing nothing
Photo by the author.
For important activities that require deep thinking, we need to have the mental discipline to focus on that task alone. Doing many things at once convolutes our brain with too much information, preventing it from generating useful insights to get the work done.
Shuffling constantly between tasks may be a great excuse for leaders and managers to keep busy, but without learning to manage their time, they cannot do work that will advance their organisation and its people.
A useful trick that improves productivity is to schedule similar activities one after another. It enables much faster progress, as learning and inputs from the previous activity feed directly into the new tasks without requiring extra mental effort to form new connections.
For instance, if I have to write job descriptions and create a pitch for a hiring strategy, I prefer to schedule job descriptions right after developing a hiring strategy. The inputs from the hiring strategy can then be directly applied to writing job descriptions.
Once we learn to discipline our minds to focus on one task at a time, we also need to discipline our minds to give up the urge to do everything ourselves.
11. Doing more is doing less yourself
There are multiple benefits of delegation: achieving scale, building trust, and empowering others. Productivity just happens to be a side effect of making the right choice to delegate. Focus on what you can do best and delegate everything else.
Delegation is not easy and can only be effective if you set clear expectations and align on the outcomes. Initially, it may seem like a lot of work and your mind may tempt you to do it yourself, but you can learn to give up the temptation in favour of achieving long-term productivity.
As you learn to delegate more, you will get ample free time on your hands to do things that matter.
It’s also not wise to engage humans in doing work that can be achieved through automation. Technology has enabled some beautiful solutions. Identify what part of your repetitive work can be automated and make the best use of technology to automate them.
Setting up your own productivity routine is great, but will it work without cooperation from others?
12. Establish communication guidelines
If your work demands communication and collaboration with others, you cannot be productive unless you establish communication guidelines with them. Without knowing the best way to reach out to you or when to do so, others may continue to interrupt you during your most productive hours.
Define the medium to use for different kinds of issues (email, phone, chat) and broadcast blocks of time when you will be available to answer queries, help coworkers, or whatever else your work demands.
Plan the unplanned work by consciously dedicating a few hours in a day to avoid reacting every time someone comes up with an ask.
You may have a great memory, but trying to remember everything is not the best use of your mental resources.
13. Don’t waste your brainpower in trying to remember things
Photo by the author.
Our brains can’t hold too much information at the same time. Keeping the mental stage free for processing information that needs it is the easiest way to improve productivity.
The best way to keep yourself organised and do the right things at the right time is to schedule your responsibilities. Send calendar invites, set up reminders.
Scheduling also saves energy by removing decision-making from the process. Once it’s on your calendar, you don’t need to decide anymore.
This is the beginning of unleashing your full potential.
14. Good habits sow the seed of productivity
Good habits are the key to unlocking your maximum potential. Examine which habits energise you and which ones contribute to your inefficiency. Unless you learn to change your habits, you may get the work done, but it will be far less than what you are capable of.
Learn more about yourself. Healthy eating, sleeping well, constant learning are some of the most basic needs for a healthy mind and body. Also, doing what you really enjoy — even if for a small duration every day — can renew your focus and unleash a creative outflow. | https://medium.com/better-programming/14-steps-to-productivity-66bd4f4e0c93 | [] | 2020-06-22 14:50:24.693000+00:00 | ['Personal Development', 'Self Improvement', 'Productivity', 'Life Lessons', 'Programming'] |
AES Implementation in Python | Implementing AES in Python
Fortunately, we don’t have to implement AES from scratch, but you can give it a try if you’re feeling spicy. In order to avoid doing so, we first need to install the pycrypto library, which can be done via pip with the following command:
pip install pycrypto
which should run without errors. Once pycrypto is installed, create a python file and write the following to import everything we need:
Now we are going to create a class for our AES cipher with the following constructor:
Lets take a quick walk through the constructor, it receives a key which can be of any length. Then we proceed to generate a 256 bit hash from that key. A hash is basically a unique identifier of a given length, 32 characters in this case, of any input no matter its length. This enables that you can pass the constructor your name or a phrase, and it will generate an unique 256 bit key for your cipher. We also set block_size to 128, which is the block size of AES.
Before we proceed to define the encrypt and decrypt methods for our AESCipher class, lets first create the __pad and __unpad methods.
Pad
The __pad method receives the plain_text to be encrypted and adds a number bytes for the text to be a multiple of 128 bits. This number is stored in number_of_bytes_to_pad . Then in ascii_string we generate our padding character, and padding_str will contain that character times number_of_bytes_to_pad . So we only have to add padding_str at the end of our plain_text so that it is now a multiple of 128 bits.
Unpad
In an opposite manner, __unpad method will receive the decrypted text, also known as plain_text and will remove all the extra added characters in the __pad method. For that we first must identify the last character and store in bytes_to_remove how many bytes we need to trim of the end of plain_text in order to unpad it.
Encrypt
With these two methods out of the way, lets implement the encrypt method
The encrypt method receives the plain_text to be encrypted. First we pad that plain_text in order to be able to encrypt it. After we generate a new random iv with the size of an AES block, 128bits. We now create our AES cipher with AES.new with our key , in mode CBC and with our just generated iv . We now invoke the encrypt function of our cipher , passing it our plain_text converted to bits. The encrypted output is then placed after our iv and converted back from bits to readable characters.
Decrypt
Now that we can encrypt text, but I’m sure we would like to be able to decrypt it:
In order to decrypt, we must backtrack all the steps done in the encrypt method. First we convert our encrypted_text to bits and extract the iv , which will be the first 128 bits of our encrypted_text . Much like before, we now create a new AES cipher with our key, in mode CBC and with the extracted iv . We now invoke the decrypt method of our cipher and convert it to text from bits. We remove the padding with __unpad and that’s it!
The AESCipher class should end up looking as the following:
All the code shown in this tutorial is publicly available in this Git Hub repository. If you want to play with AES or just check that your implementation was correct, try out this online AES Cipher. If you are intrigued about how AES works, feel free to pass by my article about it.
I hope you all found this tutorial useful, happy encryption!
Thank you. | https://medium.com/quick-code/aes-implementation-in-python-a82f582f51c2 | ['Pablo T. Campos'] | 2020-05-03 00:53:23.915000+00:00 | ['Python', 'Security', 'Good Practices', 'Tutorial', 'Cryptography'] |
Crush! | He still remember his school crush.
He talked her a lot.
Without even speaking a single word. | https://medium.com/3-lines-story/crush-92433b04ed4 | ['Pawan Kumar'] | 2017-09-06 14:18:46.009000+00:00 | ['Crush', 'Storytelling', 'Relationships', 'Poetry', 'Love'] |
Gardening made me the grumpy old man on my block | Tales from a grumpy new gardener
Gardening made me the grumpy old man on my block
My childhood neighbor would be so proud
Photo credit: Paul Streltsov/Unsplash
I scowled outside at the birds, having the times of their lives stabbing the grass. I hate organic fertilizer. The birds and rabbits are enjoying our lawn and bringing friends over, social isolation be damned. And all I keep thinking about is how long the grass will grow with them in it, eating up all the garden treatment. I’m going back to good old-fashioned traditional fertilizer and weed removal. I know I guest-write for Earth & Me, but going green is too stressful.
Then came the tenant who was determined to let her kids run in the grass. With every step they run — regardless of the condo association bylaws — I wince, thinking of how long it’s going to take for grass to grow if it’s getting trampled over all the time. And they leave popsicle plastic behind. The audacity. Even after lecturing the parents to stop letting the kids play in the grass, I came outside on Friday and the same couple put their lawn chairs in the lawn — and brought friends. When I said, “Please don’t sit in the grass,” I didn’t mean go get a lawn chair to sit higher up in the same grass.
I never expected to be in my 30s and invested in grass. How one goes from having a black thumb for 36 of her 38 years to standing in Home Depot for an hour, reading the backs of all the fertilizer and weed removal bottles, is beyond comprehension. I thought I was fancy for keeping an arrowhead plant alive on my coffee table, but never did I ever humor the idea of guarding an actual garden.
Somehow, some way, I’ve become my childhood best friend’s grandfather — the guy who lived next door and mowed one foot of the grass on his side of our gate, along with his entire yard. (Imagine how goofy it looked to have uncut grass on our side, but one foot of it neatly trimmed.) I’ve become the person who mumbles, “Don’t you see the sidewalk!” whenever some maniac cuts through the grass to leave the building. I’ve become the person who tests out garden hoses to find out their weight — and knows to wear clothing that I can prepare to be drenched by sprinkler water. And as much as I hate to admit it, if I see people standing outside the condo too long, I ponder on whether I should go get the sprinklers and turn them on “high” to get them to move to the next building.
My grandfather, another gardener who grew enough fruits and vegetables in his backyard to the point of never really needing grocery store produce sections, got along splendidly with our crabby neighbor. I had to respect the man because he was my friend’s grandfather, and of course because my parents would’ve yanked me up by neck for being disrespectful. But never in my life did I expect that hiring gardeners and then getting involved in the garden would turn me into this duo.
I stare in the mirror every day, waiting for a beard to grow in or to start buying flannel shirts and khaki vests — or even worse, for me to start uncontrollably pointing my finger and yelling at kids. (Right now I’m only lecturing adults.) I already have the floppy hat to keep the sun out of my eyes while I size up Hostas and Echinacea purpurea magnus and Diervilla Kodiak Orange plants, wondering if they’re really in the right spot for Chicago’s heat.
And I catch myself mentally staring at the amount of hardwood mulch and compost, to make sure it still looks as neat as when it was first installed. My grandfather and my neighbor (who have passed on) would be so proud to see I joined their Grumpy Gardener crew. Meanwhile I’m wondering why 2020 has turned me into someone I don’t recognize. Don’t judge me though, unless you’re prepared for my water hose, locked and ready to spray at a moment’s notice. And I’m not scared of you crazy kids either! | https://medium.com/tickled/gardening-made-me-the-grumpy-old-man-on-my-block-766a4ad1ee | ['Shamontiel L. Vaughn'] | 2020-09-28 00:54:15.806000+00:00 | ['Humor', 'Gardening', 'Storytelling', 'Green Thumb', 'Plants'] |
“It is just an animation” | “It is just an animation”
Many junior analysts are struggling to sell their minimal slides to a boss who is insisting on combing slides 2, 3, 4, and 5 into one “to save time”. Here is one line of thought you can try: convincing here that slides 2, 3, 4, and 5 are in fact already one and the same slide, it is just an elaborate animation that you spread out over a couple of pages to make editing easier. “The audience will perceive it as one slide”.
You can make your argument more credible by making sure that there are connecting elements on your slides that stay the same as you switch pages. | https://medium.com/slidemagic/it-is-just-an-animation-792b5baa5bb6 | ['Jan Schultink'] | 2017-02-07 07:09:08.312000+00:00 | ['Design', 'Presentations', 'PowerPoint'] |
Understanding Kubernetes Multi-Container Pod Patterns | Ambassador
The ambassador pattern derives its name from an Ambassador, who is an envoy and a person a country chooses to represent their country and connect with the rest of the world. Similarly, in the Kubernetes perspective, an Ambassador pattern implements a proxy to the external world. Let me give you an example — If you build an application that needs to connect with a database server, the server configuration, etc, changes with the environment.
Now, the official recommendation to handle these is to use Config Maps, but what if you have legacy code that is already using another way of connecting to the database. Maybe, a properties file, or even worse, a hardcoded set of values. What if you want to communicate with localhost, and you can leave the rest to the admin? You can use the Ambassador pattern for these kinds of scenarios.
So, what we can do is create another container that can act as a TCP Proxy to the database, and you can connect to the proxy via localhost. The sysadmin can then use config maps and secrets with the proxy container to inject the correct connection and auth information.
Ambassador
For the demo, we will use a simple NGINX config that acts as a TCP proxy to example.com. That should also work for databases and other back ends.
If you look carefully in the manifest YAML, you will find there are three containers. The app-container-poller continuously calls http://localhost:81 and sends the content to /usr/share/nginx/html/index.html .
The app-container-server runs nginx and listens on port 80 to handle external requests. Both these containers share a common mountPath . That is similar to the sidecar approach.
There is an ambassador-container running within the pod that listens on localhost:81 and proxies the connection to example.com , so when we curl the app-container-server endpoint on port 80, we get a response from example.com .
Let’s have a look: | https://medium.com/better-programming/understanding-kubernetes-multi-container-pod-patterns-577f74690aee | ['Gaurav Agarwal'] | 2020-11-20 10:32:17.522000+00:00 | ['Technology', 'Kubernetes', 'DevOps', 'Containers', 'Programming'] |
Why is Golang better than other languages? | In the world of programming languages, there is no stoppage for innovations and breakthroughs. Programmers always look for a sophisticated and easier way of writing code for development. To fulfill their need, Google launched Golang in 2007 for increasing programming productivity.
Since the launch, Go has been one of the most trending languages by overcoming the shortcomings of the other programming languages. And it has become a game-changer in the programming industry.
Let’s start with the introduction:
Golang is a compiled and statically typed programming language. It was developed by Ken Thompson, Rob Pike, and Robert Griesmer at Google.
Golang is very similar to the C programming language that helps to develop reliable, simple, and efficient software easily. And this has made many businesses look for Golang development company for developing their future projects.
Golang offers a full pack of solutions that makes it a wondrous programming language. From the time of its arrival, Go language has taken the entire programming industry by surprise.
Now, let’s reveal many of these surprises that make Golang better than other programming languages.
“Go will be the server language of the future” Tobias Lütke, the CEO of Shopify, has given this statement which shows that Golang will dominate the future of programming languages.
The Core Capability
Golang is the only programming language that integrates all the sought-after capabilities;
Ease of coding
Efficient code-compilation
Efficient execution.
Thus, Golang is considered special as all these capabilities are integrated into this single language.
Go language is a robust system-level programming language among large distributed systems and large-scale network servers. Hence, Golang has emerged as an alternative to programming languages such as Java and C++.
Golang offers various solutions such as;
A boost to code readability and documentation
Allows development with multiple languages
Facilitating easy versioning of the program
Allows easier maintenance of dependencies
Fast-paving compilation and execution
Goroutines
Goroutine is at the heart of the popularity of the Golang. This is the major feature that makes it popular worldwide. This feature allows experts to connect the facilities and features of multicore machines in the best possible way.
Memory Management
Golang is excellent at managing software’s memories. Golang has been created after a detailed analysis of all the other programming languages such as C, C++, Java, and more for integrating their useful features in it.
Error Handling Like a Pro
Golang is considered a pro at handling errors in a diverse way than other programming languages. It handles the error in its unique manner by using conventional features.
Multi-threading And Concurrency
In today’s time, hardware requires a programming language that can scale up the performance and support concurrency better as there is a continuous increase in added cores.
While working with multiple threads, most of the programming languages lack at concurrent execution. And thus, many times, they decrease the speed of execution, compilation and programming.
Here, Golang is an ideal choice for supporting both concurrency and multi-threading environments. Golang can concurrently handle a considerable number of tasks at one go as it works with Goroutines and not threads.
Offers unmatched Simplicity
Simplicity lies at the core of the Golang that is a highly sophisticated language & full of features. Because of its simplicity and straightforward approach, Golang stands taller among other programming languages.
No Generics:
Golang developers made things very simple by deciding to go without generics. Because these generics usually increase the difficulties of understanding and obscurity.
Single Executable:
Golang is able to develop a single executable code that can be deployed by just copying as it does not have any separate runtime library. This removes all the worries related to making mistakes on versions mismatch or dependencies.
Built-in Framework for Testing and Profiling
While developing any application, you might have faced the complexities of selecting a testing framework through a series of analysis.
And most of the time, you would have failed to use more than twenty percent of the chosen framework. When you need good profiling for evaluation, even then there is the same difficulty.
So, to take you out of these difficulties, Golang provides a built-in profiling and testing tool. It helps you to test the app efficiently and easily.
Easy Learning Curve
The lower learning curve is one of the most significant positive aspects of Golang. Don’t be surprised to know that you can learn the features of Golang in just few hours.
Once you learn these core features, there will be a requirement to learn the programming practices for standard library and particular requirements. Still, to start with this language, a two to three-hour sitting can be enough. So, it is easy to learn Golang compared to other languages.
Garbage Collection
This is a significant feature of Golang. This feature offers control over memory allocation.
The latest version of the garbage collector has drastically decreased latency, with average latencies under 1ms. It does this by using a tricolor mark-and-sweep algorithm and running simultaneously with the program.
Offers Great IDE Support and Debugging
Whenever a mobile application developer switches programming language, IDE support is one of the essential things. Because 80% of the developer’s coding time can be saved by a comfortable IDE.
Conclusion
Golang has been proven as a massive value addition to the programming world for developing large scalable enterprise software systems and complex apps. No wonder why Golang is better than other programming languages! | https://medium.com/devtechtoday/why-is-golang-better-than-other-languages-158c06c63d1b | ['Bharti Purohit'] | 2020-06-18 06:29:37.355000+00:00 | ['Programming', 'Golang', 'Go', 'Technology', 'Development'] |
Religion As a Political Tool: A Case Study of the Ayodhya Verdict | Religion As a Political Tool: A Case Study of the Ayodhya Verdict
How a largely communal conflict was politicized at the national level
On the 16th of October, the five-judge constitution bench, headed by Chief Justice Ranjan Gogoi, concluded the hearing on the Ram Janmabhoomi-Babri Masjid case, the judgement for which was delivered on 8th November. According to the Supreme Court verdict, the entire 2.77 acres of the disputed land is to be used for the building of the Ram temple, while an alternate 5 acres of land will be awarded to the Sunni Waqf Board for the potential construction of a mosque. The court cited evidence of Hindu temple remains under the masjid as an important factor that led to this verdict.
Illustration by Pooja Nagaraj
This judgement is being hailed as one of the most important judgements in recent times due to the long, conflict-riddled history of Babri Masjid which will soon give way to a Ram Mandir. On the surface, this issue might seem like a religious one. After all, the three main parties involved are the Sunni Waqf Board, the Nirmohi Akhara, and ‘Ram Lalla’, represented by the Hindu Maha Sabha. However, contextualise the debate in a political timeline, and it becomes clear that the developments around the case were rooted in politics.
It all started in the mid-19th century when a Ram worshipping Hindu sect, called the ‘Nirmohis’, claimed that the Babri Masjid was built upon the ruins of a Ram temple. This claim gave rise to some minor communal skirmishes, but it was largely ignored by most. In the year 1949, some people claimed that they witnessed a mysterious child recognised as ‘Ram Lalla’ or infant Ram appear within the gates of the masjid with a sudden flash of lightning. This child then allegedly transformed into an idol of Ram which was found inside the masjid. Later investigations revealed that there was no proof of there ever being a child involved. In reality, three Hindu activists belonging to the Akhil Bhartiya Ramayana Mahasabha (ABRM), an offshoot of the Hindu Mahasabha (a right-wing Hindu Nationalist party), had broken into the mosque and placed the idol of Ram inside. Thus began the politicization of what was until now a largely communal conflict.
Upon the orders of the Supreme Court, the mosque was declared as a ‘disputed site’ and the gates were locked to both Hindu and Muslim worshippers. In the 1980s, the Vishwa Hindu Parishad (VHP), an offshoot of the Rashtriya Swayamsevak Sangh (RSS) and affiliated with the Bharatiya Janata Party (BJP), spearheaded a campaign to ‘liberate’ Lord Ram’s alleged birthplace. This claim is unsupported by historical evidence, not that absence of evidence has ever been a hindrance in this issue. The campaign, aimed at demolishing the mosque and constructing a temple on the spot was led by BJP minister Lal Krishna Advani, who later went on to become the Home Minister. At this time, a relatively non-powerful party like the BJP started being included in the mainstream narrative of Indian politics. They also organised a whole hoard of rallies, marches and congregations to find support for their agenda.
Although this issue had mainly been used by right-wing ‘Hindutvavadi’ (parties which seek to establish the hegemony of Hindus) until now, the Congress joined the wagon soon after. After facing flak for the appeasement of Muslim fundamentalists in the infamous overturning of the Shah Bano judgement, the Rajiv Gandhi government decided to play the ‘Hindu card’ for the upcoming elections.
On the 1st of February 1986, the padlocks of the gates were opened to allow Hindu priests to worship inside the structure. Although the opening of the gates was a court-ordered decision, the promises made by the government to the Hindu communalists before the court decision and failure to appeal against the judgement revealed the hand of the Congress government.
Not to be left behind, the BJP in 1989 consolidated the Ayodhya Movement and made the building of the temple a substantial part of their manifesto; a move that proved to be extremely fruitful as BJP witnessed a steep rise in the number of seats secured in the next elections. In 1989, the Rajiv Gandhi government, having faced backlash for the Bofors scam ( weapon contract political scandal between India and Sweden), allowed the VHP to perform the Shilanyas ceremony (laying of the foundation stone on the mosque property) in November 1989, despite it still being a legally disputed site.
What had until now been covertly used by political parties to sway public opinion, soon became a full-blown political issue.BJP leaders like Lal Krishna Advani, Murli Manohar Joshi, and Uma Bharti organised a congregation for kar sevaks (volunteers for the religious cause) from all over the country, allegedly for a grand pooja in Ayodhya. However, things descended into chaos when the kar sevaks broke through the mosque gates and began to manually destroy the mosque using hammers, rods, and even their bare hands. By mid-afternoon on 6th December 1992, the mosque had been completely destroyed, and all that was left of it was rubble. Cases were filed against the BJP leaders involved but they all maintained that the demolition was completely spontaneous and not the intended outcome of the congregation.
However, later investigations into the apparent police inaction and testimonies of present journalists and camera people paints a clear picture of there being a significant amount of planning involved in the destruction of the mosque. One does not need to look further than the BJP leaders’ recent comments on the nearing resolution of the issue to understand how the ‘apparently’ religious debate stopped being (assuming if it ever was) a religious one and stands today as a purely political one. While the courts favoured the Hindutvavadi parties in their judgement, it does not seem like the last we will see of Hinduism being used to promote the current right-wing, near-militant, political agenda. | https://medium.com/thecontextmag/religion-as-a-political-tool-a-case-study-of-babri-masjid-6e77da68957f | ['Kayva Gokhale'] | 2019-11-18 08:30:06.564000+00:00 | ['India', 'Conflict', 'Babri Masjid', 'Politics', 'Religion'] |
I’ve Only Taken One Photo With The New Nikon D850. And It’s Love. | My judgment on the quality of an image is not about what it looks like in its raw format, it’s about the subtlety and flexibility with which I can make changes to it in Lightroom. In fact, from a “straight out of the camera” aspect, my cheapest camera (the Fujifilm XT2) is far and away the best of the lot. I don’t need to touch a Fujifilm image — and as odd as it sounds, their in-camera film looks are absolutely stunning. But when it comes to the nuanced work of post-processing, there’s just nothing like working on a Nikon file. It’s truly transcendent. Until you’ve done it, it’s hard to appreciate it. Watching the image adjust ever so slowly as you slide your shadows up, for example, offers a seemingly endless range of control and subtlety that no other file I’ve worked with does as well. And for some reference, I was a Quantel Paintbox artist for years, doing high-end photo retouching.
I’ve been using the Nikon D800 since it came out, but have been looking forward to advancements announced in the D850 (backlit sensor, higher pixel count, articulated screen, wifi, higher frames per second, to name a few). I finally got the call, rushed down to Samy’s Camera and grabbed it. By the time I got home, it was dark and I couldn’t take a photo — instead, I just slowly attached the new grip, my leather Tanner strap and downloaded Nikon’s SnapBridge app and paired it to the camera. Needless to say, sleep was difficult.
There’s a lot to know about the camera — and I’m happy to get into it, if anyone cares to hear the same info that is on a few hundred other sites from my perspective — but all I really cared about was getting a decent image into Lightroom and playing around with it. We are beginning some construction in our back yard, so I took the thing with a 24–70mm f/2.8 on there out to the back. It’s a perfectly overcast day here in L.A., so light was perfect. I just wanted one shot. As usual, I was cautiously underexposed — my priority is always to keep my ISO way down for the purest image possible. I grew up shooting ISO 100 film, so I’m always starting there. This was at 160. Wide open at 2.8 (it’s probably worth noting that the image would be even sharper at f/5.6), to let the most light hit that backlit sensor and a middling 1/500th of a second. My meter showed that I was underexposed, but I had a feeling it was all there. Here’s how the Raw file looked: | https://joshrose.medium.com/ive-only-taken-one-photo-with-the-new-nikon-d850-and-it-s-love-99063b025c31 | ['The Art Of Photography'] | 2018-03-12 17:57:52.607000+00:00 | ['Creativity', 'Nikon', 'Photography', 'D850', 'Art'] |
3 Semesters Through & A Year in Review | 1. Of Rest & Sabbath
Perhaps one of the most radical changes I’ve made this semester would be keeping to a weekly sabbath. In essence, this would mean setting aside a day of the week to simply REST (sounds simple enough… but not quite). This practice originates from Christianity and Judaism, but there are also secular variants like corporate sabbatical leave (extended leave for employees to recharge personally or professionally).
Going back to the idea of “resting” for a day once a week, it soon becomes apparent that keeping to such a practice is easier said than done. Being a typical “kiasu” (meaning “afraid to lose” in colloquial Singlish) Singaporean, a recurring thought I had was whether I would be “losing out” to my peers by studying/working a day less. Having a truckload of weekly tutorials/assignments/labs as a d̶r̶o̶w̶n̶i̶n̶g̶ Computing student meant that even if I wasn’t afraid of losing out, the sheer workload made such a practice tough. With good reason, a friend who heard about my peculiar practice also asked whether I was worried that others would “overtake” me academically as I rested (oh the fear of the mighty bell curve is real).
These thoughts occasionally led me to question if having a sabbath was really worth it, but for faith-based reasons, I (largely) kept to it and didn’t regret my decision. While that same friend suggested the alternative of having a smaller break daily, I knew I could still be worrying about work even during those moments of rest. There just seems to be something unique about having a full day to rest. It meant being able to choose to do what I felt recharged me, whether it meant going for long walks or however long I wanted on the piano. Saturdays thus became a day without worries that I looked forward to each week and where I took a slower pace of life amidst the hecticness around me.
why are they all frowning ded
That being said, what I allowed myself to do on my Saturdays was also something I had to consider carefully. Would packing my Saturdays full with events and social gatherings/activities or bingeing Netflix for hours on end honour my Sabbath? These activities would not usually be associated with work but neither did they give true rest. While all academic work was understandably a no-go, there were grey area activities like prep work for student exchange programmes or internship applications. I soon realised these activities could also potentially bring stress or worry (especially over their outcome) and decided that I had to be careful if I chose to do them on Saturdays. They were only to take up a set amount of time at most, and nothing more.
Moving forward, having a sabbath is something that I’ll definitely continue to do. It may be hard to put aside work for a day, but doing so has proven its worth to me and perhaps it may do the same for you also.
Before I end off my observations on rest, I’ll share an excerpt from an article by the late Clayton Christensen titled “How Will You Measure Your Life”. It speaks volumes.
I played on the Oxford University varsity basketball team… It turned out the championship game was scheduled to be played on a Sunday. I had made a personal commitment to God at age 16 that I would never play ball on Sunday. Every one of the guys on the team came to me and said, “You’ve got to play. Can’t you break the rule just this one time?” In many ways that was a small decision — involving one of several thousand Sundays in my life. In theory, surely I could have crossed over the line just that one time and then not done it again. But looking back on it, resisting the temptation whose logic was “In this extenuating circumstance, just this once, it’s OK” has proven to be one of the most important decisions of my life. Why? My life has been one unending stream of extenuating circumstances. Had I crossed the line that one time, I would have done it over and over in the years that followed. The lesson I learned from this is that it’s easier to hold to your principles 100% of the time than it is to hold to them 98% of the time. If you give in to “just this once,” based on a marginal cost analysis…you’ll regret where you end up. You’ve got to define for yourself what you stand for and draw the line in a safe place.
2. Of Plans and Ambition
Having been in university for more than a year, I had the opportunity to observe the mindsets of my peers as I interacted with them. Some knew full well what they wanted to do upon graduation (and were perhaps even clearer on the lifestyle they wanted). Some were aware of the stiff competition (both from locals and from abroad) and decided to start preparing themselves for their careers early. Others were l̶u̶r̶e̶d̶/̶m̶i̶s̶l̶e̶d̶/̶h̶a̶d̶ ̶s̶u̶c̶c̶u̶m̶b̶e̶d̶ into multi-level marketing (MLM) schemes and their equivalents. And of course, there were those who just wanted to enjoy their time in university and worry about the things ahead later on.
While I had previously written about my own experience in chasing after internships, it took me a while to acknowledge that I had done so out of fear. I had believed that by stacking these work experiences on my CV I would be able to make up for my academic weakness and stay ahead of the pack. Yet doing so had only caused me to worry each time things didn’t go as planned.
We often think that the trajectory of our lives has to follow some kind of upward linear trend. Few people would like to think that they would be worse off than they currently are five years from now. But perhaps things will not always go as planned no matter how hard we try. Perhaps we should stop to consider what is it that we are really working towards? A materialistic pursuit can only go so far.
the higher you climb the harder you can fall
Not too long ago, I came across two articles that shared some wisdom and insight and are worth a read (via the links). One gave a stark reminder on the fragility of life (the author being a 26-year old woman diagnosed with Stage 3 breast cancer) while the other gave practical handles on handling money (from a Christian perspective).
With that, I found myself reevaluating what my own goals were. Was it really necessary to land a job that pays a five-figure salary so as to have a fancy home/car? Will I be contented with a simpler lifestyle? Would you?
3. Of Envy
Strangely, I only knew of the difference between jealousy and envy this year. All along I assumed they were one and the same but they actually describe different situations.
Envy occurs when we lack a desired attribute enjoyed by another. Jealousy occurs when something we already possess (usually a special relationship) is threatened by a third person. Envy is a two-person situation whereas jealousy is a three-person situation. Envy is a reaction to lacking something. Jealousy is a reaction to the threat of losing something.
Just like pretty much everyone else, there were definitely moments in my life where I had wished I was more like … / could have … (which I now know describes envy). These thoughts were simply brushed aside as normal and countered with the general wisdom of not comparing myself with others. Yet when I saw for myself how envy had manifested, it hit like a truck (not that I have actually been hit by a truck n̶o̶r̶ ̶w̶o̶u̶l̶d̶ ̶I̶ ̶e̶v̶e̶r̶ ̶w̶a̶n̶t̶ ̶t̶o̶ ̶b̶e̶ ̶h̶i̶t̶ ̶b̶y̶ ̶o̶n̶e̶).
the sash should have been labelled envy btw
Over the past few months, some friends had shared with me about their successes/achievements, many of which were things that I had also hoped for myself. While I congratulated them upon hearing the news, deep down my initial thought was usually, “God, why couldn’t that have been me?”. It’s scary to think that I struggled to be genuinely happy even for close friends if they had first achieved what I hoped to have. As I reflected further, I decided to observe how did others handle the good news of people around them even if they were still waiting for similar things themselves.
In particular, I noticed the (consistent) response of a friend as she handled the good news of another friend getting married. While she also desired for herself to be attached and eventually married, it was visibly clear (at least to me), that she was genuinely happy for her friend’s upcoming wedding. It was clear that her actions spoke louder than the congratulatory words she responded with, as said friend then organized collective wishes and a group gift for her friend. She was present at the wedding and was probably there to help out as well.
While I didn’t ask her directly how she dealt with envy, she indirectly shared that it boiled down to where her anchor and assurance was found. While she may/may not have wrestled with envy, for her it was her faith that “God knew what was best for her at each season of her life”. That gave her the option to place her trust and joy in something/someone that was unwavering.
Having observed this, my own takeaway was that perhaps we each need an anchor. This could mean anything to anybody, but I roughly knew what it meant to me. The human “heart” is deceitful above all things and perhaps my initial thought upon hearing the good news of others might still contain a tinge of envy. But having been made aware of envy, I would know better to hold fast to my own anchor and to genuinely rejoice for those around me.
Closing Thoughts:
I previously alluded to life being like a race or like balancing on a tightrope, and perhaps now I’ll also describe it as a garden. We each have a “garden” (̶y̶e̶s̶ ̶e̶v̶e̶n̶ ̶i̶n̶ ̶s̶q̶u̶e̶e̶z̶y̶ ̶S̶i̶n̶g̶a̶p̶o̶r̶e̶)̶ and we all want to bear fruit. Some of us have gardens that resemble orchards or plantations while some barely have a tree growing, but there’s no need to compare with your neighbour’s garden.
literally trashy neighbours
Yet, if your neighbour’s garden has weeds or is on fire (̶h̶o̶w̶ ̶e̶v̶e̶n̶ ̶r̶i̶g̶h̶t̶?̶)̶, let him/her know in a manner that is helpful. If we are constantly stressed, lack rest or don’t even know what fruits we are growing we won’t make good farmers. If we focused on weeding what’s in our garden and on bearing our own fruits maybe we will all be a step closer to having joy. But even then, this garden is not forever.
“Hold everything earthly with a loose hand, but grasp eternal things with a death-like grip.” Charles H. Spurgeon
S.D.G.
JG | https://medium.com/age-of-awareness/3-semesters-through-a-year-in-review-aa5a99223d85 | ['Joshua Goh'] | 2020-12-28 00:15:17.239000+00:00 | ['University', 'Self Improvement', 'Reflections', 'Self-awareness'] |
MuZero 101: a brief introduction to DeepMind’s latest AI | MuZero 101: a brief introduction to DeepMind’s latest AI
After several breakthroughs such as AlphaGo or AlphaZero, researchers from DeepMind have published their latest effort, MuZero. What is it all about?
Back in 2016, DeepMind introduced AlphaGo, the first computer software able to defeat professional Go players, included the world champion. The games between Lee Sedol (18 Go world titles) and AlphaGo were even immortalized in a documentary, available now in Youtube.
Why was AlphaGo so relevant? Until then, computer programs were only able to play Go at an amateur level, as traditional Machine Learning methods such as search trees were simply not capable of evaluating all possible moves, board positions strengths, etc. Actually, Computer Go is described as the field of artificial intelligence (AI) dedicated to creating a computer program that plays the traditional board game Go, and Go programs exist since 1968.
AlphaGo combined search tress with deep neural networks, with the intention of capturing the intuitive aspect of the game, as its creators confirm.
AlphaGo is based on two deep neural networks:
One for evaluating the policy: which should be the next move?
which should be the next move? Another one for evaluating the value: predicting the winner of the game
By using reinforcement learning, AlphaGo began learning from its own mistakes while playing against different versions of itself, until finally being able to beat the world champion 4 games to 1. Also, it demonstrated that by using those two deep neural networks, AlphaGo was able to come up with policies that seemed bad when watched from a human perspective, but turned out to be maximizing the value, as in the famous move 37 against Lee Sedol:
In order to achieve its performance level, AlphaGo was feed with human data, domain knowledge and a known set of rules.
Following Deepmind research results, AlphaGo Zero (2017) and AlphaZero (2018), improved the original algorithm by learning to play on its own, without any human data or domain knowledge, or even by mastering three different games (Go, Chess, Shogi) with only one algorithm being only provided with the known set of rules.
Now, December 2020, Deepmind has published its latest research paper in which they present their latest computer program, MuZero, capable of mastering four games (Go, Chess, Shogi, Atari). So, what is the big difference between AlphaZero and MuZero? It might seem like a simple evolution of the former which has been able to “just learn” another game, but there is one big difference. MuZero has learnt to master these games without being provided known rules.
DeepMind claims that MuZero is a big step forward towards general-purpose algorithms. And they think so because they have introduced a novel method to approach planning in unknown environments, a very relevant problem to AI. | https://towardsdatascience.com/muzero-101-a-brief-introduction-to-deepminds-latest-ai-a2f1b3aa5275 | ['David Pereira'] | 2020-12-29 13:12:36.673000+00:00 | ['Deep Learning', 'Artificial Intelligence', 'Deepmind', 'Reinforcement Learning'] |
Why Nick Cave is wrong about Fairytale of New York. | If we switch the F-Word for the N-word, would we still be having the same discussion?
Photo by Zeeshaan Shabbir from Pexels
It’s not Christmas for me until I hear the opening lines of Fairytale of New York.
I can’t remember the first time I heard the song, but its jarring and desolate story has become synonymous with December in my mind. I’ve sung my sober and drunken heart out to this cheerless festive tune, year after year. Which means I’ve also sung the word.
The F-word.
Even as I belted the lyrics out without thought or care, I knew that taboo always lurked within the lines:
You scumbag, you maggot You cheap lousy faggot Happy Christmas your arse I pray God it’s our last
Culture is not stagnant. It evolves. And so it does not upset me that the BBC has recently decided to play a censored version of Fairytale on some of its stations. Somebody that it does seem to have upset is Nick Cave, who’s railed against the decision on his website:
The changing of the word ‘faggot’ for the nonsense word ‘haggard’ destroys the song by deflating it right at its essential and most reckless moment, stripping it of its value. It becomes a song that has been tampered with, compromised, tamed, and neutered and can no longer be called a great song. It is a song that has lost its truth, its honour and integrity — a song that has knelt down and allowed the BBC to do its grim and sticky business.
I’m sorry but Cave is simply wrong. If Fairytale is a great song then it can survive the replacement of a single word.
Shane McGowan has previously justified his use of the F-word in his song. It’s spoken from the mouth of a character, he explains, and ‘not all characters in songs and stories are angels’. As a writer, I agree. If we only ever wrote from our own perspective, the literary world would be dull and one-dimensional.
But are the nuances of character and authorial intent clear in the heat of the moment? Is that clarity carried through when the listener is buoyed along by the exuberance of the song? How can the listener fail to be influenced by the position the word occupies within the lyrics, bookended as it is between a series of slurs?
Even if the listener understands that the F-word is being used to denote a specific time and place, the word itself is always dragged into the present. It’s dragged into us.
‘Words are things’, as Maya Angelou said.
As a gay woman, the F-word is not mine to reclaim. But neither is it Nick Cave’s. As a heterosexual man, he approaches the use of the word from a position of relative privilege. He admits:
I am in no position to comment on how offensive the word ‘faggot’ is to some people, particularly to the young — it may be deeply offensive, I don’t know’.
He doesn’t know. But as an artist, he should recognise the danger in the repeated positioning of the word as a slur.
I love Fairytale. It’s Christmas to me. When I hear Kirsty McColl use the word ‘haggard’ as a substitute, I don’t find that the song is lessened. I recognise that the song is alive and sensitive to the fluctuations of culture. I find it to have progressed.
The furore over insisting the F-Word remains says more about the people that want it to stay in than the people that want it out.
There’s a simple test to gauge how we should treat this. If we switch the F-Word for the N-word, would we still be having the same discussion? How would we describe those that fought for its inclusion in the song? How would we consider their demand to sing it and have it sung?
As writers, we’re trapped within the prism of our current perspective and place in unfolding history. But that doesn’t mean that we can’t look back on our stories with hindsight, humility, and respect for others. This should be especially true for the stories that we revere the most.
Just switch out the damn word. A Fairytale of New York will survive without it. | https://medium.com/illumination/nick-cave-is-wrong-about-fairytale-of-new-york-b50646a4becd | ['Vee Uye'] | 2020-12-03 13:39:21.614000+00:00 | ['Songs', 'Writing', 'Christmas', 'Culture', 'LGBTQ'] |
Dear Neighbors, Please Don’t Judge Me By The Condition of My Yard | When we moved into this house 19 years ago, the yard was in pristine condition. The former owner — her name was Marie — lived here for 30 years.
The older couple (now long gone) who lived next door told us that Marie spent most of her time outside working on the yard because the house was crowded with three generations of her family, plus several elderly tenants that Marie took care of.
It’s no wonder Marie took solace in the yard. The inside of the house was dark, cluttered, and neglected. We didn’t love the house, but it was large and cheap and had potential. There was room for my husband’s drum studio and, with a bit of hard work, up to four bedrooms for our growing family.
I was eight months pregnant with the baby who would die almost 16 years to the day that we moved in.
Marie had planted flowering bushes randomly around the yard, all of them covered with sturdy, unyielding thorns. My husband, haunted by the image of our baby toddling into one of those bushes, took it upon himself to remove them one by one. That was pretty much the extent of the yard work we accomplished that first year, beyond a few feeble attempts to mow. The baby arrived on May 16th, about a month after we moved in. Both of us were exhausted.
The lawn had looked wonderful during the first few weeks of spring, but quickly became overgrown. Our property is just under half an acre. It may not seem that large, but when you are tired, unused to owning your own home, and trying to figure out how to take care of a new baby, yard work becomes a low priority task.
We bought our first lawn mower soon after we moved in (we’re now on our third) and barely used it. I have a vague memory of two neighborhood kids offering to mow the overgrown yard and giving up halfway into the task, defeated by the many obstacles (mostly thorns) that made their job nearly impossible. We paid them $20 anyway. | https://jacquelinedooley.medium.com/please-dont-judge-me-by-the-condition-of-my-yard-b1989d04cd64 | ['Jacqueline Dooley'] | 2020-05-31 03:15:45.926000+00:00 | ['Life', 'Outdoors', 'Parenting', 'Mental Health', 'Family'] |
The State of Piping in JavaScript | The Pipeline Operator Proposal
To better support chained function calls in a readable and functional manner, there is an open TC39 proposal to introduce a pipeline operator. The pipeline operator |> is inspired by equivalent operators in functional languages like F# and Elixir and also regular Unix pipes.
Using the proposed pipeline operator, we could fix the readability issue of our previous example:
Much better — now we have our reusable components, and we can compose them in a readable manner from left to right (or top to bottom, as in the above example).
This was a very basic example using a single argument for each function. Let’s look at another example with multiple arguments.
This is not too bad. Our example using the pipeline operator is more readable than the equivalent example at the last line without the pipeline operator. Still, the arrow functions necessary to handle multiple arguments in the pipeline add noise and make the code more difficult to read.
To solve the readability issue, we turn to yet another TC39 proposal — namely, the partial application proposal. In programming, partial application is the act of pinning a number of arguments to a function, producing another function with fewer arguments. According to the proposal:
This proposal introduces a new syntax using the ? token in an argument list which allows you to partially apply an argument list to a call expression by acting as a placeholder for an argument.
Using the partial application syntax, we could improve the readability of our multi-argument pipeline like this:
That is satisfying — removing the arrow functions greatly improved the readability of our example. Now the multi-argument pipeline reads almost as easily as our single argument pipeline.
This all looks very promising, but you might think, what’s the catch? Well, as it turns out there are a few, unfortunately.
First of all, the proposals for both the pipeline operator and partial application are still at Stage 1, which by the TC39 process means, “The committee expects to devote time to examining the problem space, solutions and cross-cutting concerns.” At this stage, there is no guarantee whatsoever that the proposal will ever reach the standard.
Further, for the pipeline operator, there are currently two competing proposals — F# Pipelines and Smart Pipelines. This article will not cover the details of how the two proposals differ. Both proposals share some basic commonalities like using the |> as the pipeline operator. They also overlap on how to handle the most basic use case of synchronous unary functions, like our first example with functions taking a single argument. However, the proposals do differ on many things, like how to handle asynchronous functions, partial application, etc.
Looking at the proposals, they have been around for almost three years without reaching Stage 2. One of the problems is that the committee hasn’t been able to agree on what proposal to move forward with. There is a long-running discussion on GitHub on the idea of moving forward with a minimal proposal, but unfortunately, that too seems to be stuck for now.
The standards committee operates by consensus, which is reasonable considering the impact decisions have on the language. As one TC39 committee member puts it on GitHub:
Our style in TC39 tends to be a bit more conservative, where we try to hold off until a feature is “done.”
The flip-side of this is that proposals (like the pipeline operator) risk getting stuck when the committee fails to reach a consensus on how to move forward.
Personally, I hope that the committee can agree on going forward with a minimum proposal to get the ball rolling. The minimum proposal would be a great contribution to the language on its own, and having something to build on with future proposals could hopefully aid in moving the development of the feature forward. | https://medium.com/better-programming/the-state-of-piping-in-javascript-ccbead18b971 | ['Christofer Eliasson'] | 2020-10-27 15:21:52.370000+00:00 | ['JavaScript', 'React', 'Functional Programming', 'Nodejs', 'Programming'] |
Day Twenty-Two: My Journey to Zen | Day Twenty-Two: My Journey to Zen
Who gets to tell the story?
Photo by Peter Pryharski on Unsplash
As my creativity unleashes itself from all the knots and tangles in my brain, I am reveling in the stories that have been set free. I can write unimpeded by social demands and obligations. My work space is neat and tidy and my fingers are unfettered on the keyboard.
But….
I am now facing the dilemma of whose story can I tell. Where does my story end and the other person’s begin. I’m writing fiction and I do love to tell a story but there are dangers in someone thinking I am writing about them when I am just telling a story.
Whoa!
The brain is a devious thing. I have been carving out the time and space and creating the mindset I need for writing as I wish to write. There are no impediments.
So what does my brain do?
Throws up an invented roadblock that could appear logical or at least reasonable. It is using my habits and personality in its battle against change. The brain is like that.
It is one of the reasons that really small changes are more successful. Simply explained, the changes are so small they sneak past the change-resistant brain.
Family Stories
I have been intrigued since childhood with family stories. Not the whitewashed ones we were meant to accept but the human side of our grandparents and their grandparents. I wanted to understand what their lives might have been like.
The arrogant figure of my grandfather with his deep booming, and rather scary, voice is what I witnessed. When I was ten or eleven, my grandmother was in the hospital and apparently very ill. Grandpa and my parent and I went to the hospital. I had to stay in the car.
The pertinent part of the memory is that I was sitting alone in the back seat and cranky old Grandpa was in the driver’s seat, crying. What would he do without his beloved Bert? That was his plaintiff wail. I did what I had learned was best in those days and faded into the upholstery, barely breathing. I had to absorb the reality that the old guy had feelings.
He was strict and I knew that he disliked his step-father immensely. That was all I knew. Years later, I found the tombstone for his mother. She had been buried with her second husband, the much-despised stepfather. She had been born in 1864 and the second husband had been born in 1885.
Yes. A 21 year difference. Oddly enough, I looked for the tombstone on Billion Graves and found it but it notes that she was born in 1884. The transcriber must have edited the date thinking that 1864 was an error. Her first child was born in 1892 so an 1884 birthdate is highly unlikely.
She and her second husband died the year after I was born so I never knew her. At a family funeral, an elderly cousin remembered her and was great at telling me about her. She was a warm, funny person who liked to laugh and enjoy life.
There are wonderful records available for anyone who wants to explore their ancestry. I found her marriage certificate for her first marriage. And birth certificates for her four children. I was interested to learn that she married six weeks before her first child was born.
Her first husband was a man who lived next door to her family and who was five years younger than her. He died 10 years later in a farm accident. Within five months, she remarried. She was 39 and the new husband was 18.
Everyone Has a Story
We all have stories, all of us, our parents, our grandparents. The chances of writing a story straight from our brain and not hitting on someone’s story are slim.
I have to be more mindful than ever imagined to keep an eye on my self-sabotaging brain. Doing this for 30 days just sets a pattern. Maintaining it is another thing entirely. I have to be mindful of that too. | https://medium.com/narrative/day-twenty-two-my-journey-to-zen-8709761f04dd | ['Joanne Reid'] | 2020-09-05 15:25:44.215000+00:00 | ['Mindfulness', 'Genealogy', 'Writing', 'Writing Tips', 'Habit Building'] |
5 Things You Should Know About the Evolution of Books | 5 Things You Should Know About the Evolution of Books
There was once a time when books couldn’t be purchased by everyday people, this is how they have evolved
Photo by Natalia Y on Unsplash
Nowadays, when it comes to books, we’re very lucky. At the click of a button, we can buy books at a reasonable price and have them conveniently delivered to our door.
And that’s not all. With 2,320,318 books published this year alone, we’re never short on choice. Books are now available in every genre you can imagine.
And after all that the digital age has thrown at us — books are still big business. But they have a more humble beginning than you probably think.
Here are 5 things you should know about books and how their long history influenced their current form.
Hopefully, after reading this, you’ll never take a book for granted again. | https://medium.com/books-are-our-superpower/5-things-you-should-know-about-the-evolution-of-books-4f93e958e061 | ['Violet Daniels'] | 2020-11-13 12:40:17.797000+00:00 | ['History', 'Books', 'Reading', 'Education', 'Culture'] |
[Chapter-2] Fine-Tune Your Machine Learning Model — Part-5🧙🏻♂️ | Grid Search 🕵️♂️
One way to tune your machine learning model that would be to fiddle with the hyper-parameters manually 👷, until you find a good combination of hyper-parameter values. This work will be very tedious and you may not have time to explore plethora of combinations. Sounds very hectic right🥴.
Instead you can use Scikit-Learn’s GridSearchCV to do that for you😍. All we need to do is tell which hyper-parameters we want to experiment with, and what values to try out, sounds very cool right. Let’s have quick look at the small example.
from sklearn.model_selection import GridSearchCV param_grid = [
{'n_estimators': [3, 10, 30], 'max_features': [2, 4, 6, 8]},
{'bootstrap': [False], 'n_estimators': [3, 10], 'max_features': [2, 3, 4]},
] forest_reg = RandomForestRegressor() grid_search = GridSearchCV(forest_reg, param_grid, cv=5,
scoring='neg_mean_squared_error',
return_train_score=True) grid_search.fit(housing_prepared, housing_labels)
When you have no idea what value s hyper-parameter should have, a simple approach would be much better😮. Here, param_grid tells Scikit-Learn to evaluate 3 x 4 = 12 combinations of n_estimators and max_features hyper-parameter values specified in the dictionary. Do not worry about what these hyper-parameters mean as of now. After that try all 2 x 3 = 6 combinations of hyper-parameter values in the second dictionary, however this time with bootstrap hyper-parameter set to False instead of True🤗.
All in all, the grid search will explore combinations of Random Forest Regression hyper-parameter values, and it will train each model five times😲.
>>> grid_search.best_params_
{'max_features': 8, 'n_estimators': 30}
We can find out about best parameter by trying out. Below will be to find about best estimator.👩🏫
>>> grid_search.best_estimator_ RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,
max_features=8, max_leaf_nodes=None, min_impurity_decrease=0.0,
min_impurity_split=None, min_samples_leaf=1,
min_samples_split=2, min_weight_fraction_leaf=0.0,
n_estimators=30, n_jobs=None, oob_score=False, random_state=None,
verbose=0, warm_start=False)
And of course the evaluation results are also available:👨💼
>>> cvres = grid_search.cv_results_ >>> for mean_score, params in zip(cvres["mean_test_score"], cvres["params"]):
>>> print(np.sqrt(-mean_score), params)
63669.05791727153 {'max_features': 2, 'n_estimators': 3}
55627.16171305252 {'max_features': 2, 'n_estimators': 10}
53384.57867637289 {'max_features': 2, 'n_estimators': 30}
60965.99185930139 {'max_features': 4, 'n_estimators': 3}
52740.98248528835 {'max_features': 4, 'n_estimators': 10}
50377.344409590376 {'max_features': 4, 'n_estimators': 30}
58663.84733372485 {'max_features': 6, 'n_estimators': 3}
52006.15355973719 {'max_features': 6, 'n_estimators': 10}
50146.465964159885 {'max_features': 6, 'n_estimators': 30}
57869.25504027614 {'max_features': 8, 'n_estimators': 3}
51711.09443660957 {'max_features': 8, 'n_estimators': 10}
49682.25345942335 {'max_features': 8, 'n_estimators': 30}
62895.088889905004 {'bootstrap': False, 'max_features': 2, 'n_estimators': 3}
54658.14484390074 {'bootstrap': False, 'max_features': 2, 'n_estimators': 10}
59470.399594730654 {'bootstrap': False, 'max_features': 3, 'n_estimators': 3}
52725.01091081235 {'bootstrap': False, 'max_features': 3, 'n_estimators': 10}
57490.612956065226 {'bootstrap': False, 'max_features': 4, 'n_estimators': 3}
51009.51445842374 {'bootstrap': False, 'max_features': 4, 'n_estimators': 10}
In this example, fortunately we obtain the best solution by setting the max_features hyper-parameter to 8 and the n_estimators hyper-parameter to 30. The RMSE score for this is 49,682. Congratulations, you have successfully fine-tuned your best model!👨🚀 | https://medium.com/analytics-vidhya/chapter-2-fine-tune-your-machine-learning-model-part-5-%EF%B8%8F-8704fdabbdc0 | ['Vishvdeep Dasadiya'] | 2020-12-25 16:14:31.781000+00:00 | ['Scikit Learn', 'Machine Intelligence', 'Data Science', 'Machine', 'Python'] |
Powerful Plots with Plotly | One of the finest data visualization tool
(Note: Some phones may not be able to show the interactive plots. Please use a laptop if that’s the case.)
Why Do I Use Plotly?
I have written a series of articles on data visualization, including “Pandas-Bokeh to Make Stunning Interactive Plots Easy”, “Use Seaborn to Do Beautiful Plots Easy”, “Create Beautiful Geomap with Plotly” and now this one. My goal is to assist you to produce data visualization easily and proficiently. I have written articles on a variety of data science topics. For the ease of use, you can bookmark my summary post “Dataman Learning Paths — Build Your Skills, Drive Your Career” that lists the links to all articles.
So why plotly? I am thrilled to tell you that with the following summary:
Elegant: Plotly is built on top of visualization library D3.js, HTML and CSS, which makes it one of the finest data visualization tools. D3.js is a flexible library for rendering and animating SVG in the web browser. If you have seen D3.js or are using it, you will agree it has a very powerful visualization. Plotly incorporates D3.js. It lets you create interactive visualizations without even having to know D3.js. Interactive: It is created using Python and the Django framework. One can choose to create interactive plots online or use the libraries that plotly offers to create these visualizations in the language/ tool of choice. Multiple languages: It is compatible with a number of languages or tools: R, Python, MATLAB, Perl, Julia, Arduino. If you are an R user or MATLAB coder, you will still take the full advantage of Plotly. Easy: The syntax for creating interactive plots is very simple.
Typical Types of Plots in Exploratory Data Analysis
As we know “a picture is more than a thousand words”, a graph can convey the interactions of variables effectively. That’s the idea in exploratory data analysis. If we summarize the typical tasks in exploratory data analysis, the following types of exercises are probably the most:
Showing the distribution of X, and Showing the distribution of Y by another categorical variable X, and Showing the interactions of two or three variables.
So in this article I will present you the plots according to the above categories. I will walk you through different types of plots. The code notebook is available at the end of this article.
I am going to use the red wine quality data in Kaggle.com to do the analysis. use the same dataset in “Explain Your Model with the SHAP Values” and the above mentioned articles so you can compare and contrast. The target value is the quality rating from low to high (0–10). The input variables are fixed acidity, volatile acidity, citric acid, residual sugar, chlorides, free sulfur dioxide, total sulfur dioxide, density, pH, sulphates and alcohol. All the variables are numeric. There are 1,599 wine samples.
I create two categorical variables. I use Pandas’ quantile-based discretization function pd.qcut() to cut each variable into two equal-sized buckets.
Type 1: Showing the distribution of X, and
(1.1) Bar Chart
How to export your plotly figure? You can use plotly.offline.plot() and turn the auto_open=False as below:
(1.2) Pie Chart
(1.3) Histogram
Type 2: Showing the distribution of Y by another categorical variable X
(2.1) Box Plot 0
(2.1) Box plot 1
(2.1) Box Plot 2
(2.1) Box Plot 3
(2.2) Violin Plot 0
(2.2) Violin Plot 1
(2.2) Violin Plot 2
(2.2) Violin Plot 3
(2.3) Ridgeline Plot
(2.4) Two Subplots
You may need to produce several subplots, one for count, one for average, and so on. Plotly lets you specify multiple subplots and craft each plot individually.
Type 3: Showing the interactions of two or three variables.
(3.1) Bubble Chart
(3.1) Scatter Plot
(3.2) Density Plot
The code notebook is available in this github. | https://medium.com/analytics-vidhya/plot-with-plotly-114ac106e25f | ['Dr. Dataman'] | 2020-10-01 17:06:54.402000+00:00 | ['Analytics', 'AI', 'Data Visualization'] |
Buddhist Parable to Reduce Suffering | The Buddhist parable about the second arrow explains the cause of much unnecessary suffering.
The story goes that if an arrow strikes you as you wander in your garden, naturally, it hurts. But this initial injury might not signal the end of your distress. Even if the person who attacked you flees without casting another blow, you might still experience yet another bout of pain which Buddhists call the second arrow.
The second arrow, unlike the first, is controllable. It is anguish caused by your thoughts after a hurtful blow takes place. So, you might add pain (another arrow) to the initial cause of angst.
How you deal with a problem causes the second arrow to flow or disappear
We all suffer from the impact of that second arrow at times. Maybe we say something inappropriate at a business meeting, or to a friend, and pierce ourselves with torment by reciting the event in our minds rather than sort out the problem.
Or, perhaps an adverse incident occurs, we lose our job for instance, and we tell ourselves we are no good. The second arrow, self-defeat, continues injuring us and we lose courage.
The arrow may strike when we attempt to be productive too. My friend often berates herself when she neglects to use positive thinking. Her second arrow is self-imposed punishment for contemplating what she sees as an improper thought.
So, you see, we may injure ourselves with that second arrow for a variety of reasons, and even when we embark upon self-mastery arrows can flow.
How to stop casting the arrow
Self-awareness is the first step. You must recognize when you pull back the imaginary bow before you let the arrow fly. It’s difficult to change, since piercing our psyches with suffering is habitual, but it is possible to alter our behavior.
One step at a time
Initially, when you take steps toward change, you’ll find the arrow has been cast. You will notice it installed in your mind. But you’ll stem the pain by extricating it with determination and resilience.
You’ll tell yourself you are in charge of your pain and can control what happens next.
You will realize you couldn’t govern the first arrow, but now’s your chance to make a difference to how it affects your well-being.
After much practice, you’ll stop the arrow flying before it strikes you. In a moment of wisdom, because your brain is changing to accommodate this new behavior, you’ll take a different route.
At such a time, you’ll take stock, and release your resistance to the first arrow.
Accepting the first arrow is the key
The second arrow never flies unless you resist the first. By resisting, you don’t help the situation. Your rage or worry about what’s befallen you changes nothing for the better. It is a weapon you turn inward.
The way to stop hurting yourself is to accept whatever’s upset you as having happened. This doesn’t mean you think it is okay or agree with it. It means you stop fighting, living in the past with your sorrow, and shift into the present to deal with what ails you.
When you take positive action, whether it’s to self-sooth, find help, or manage the problem another way, you stop focusing on casting the second arrow and the difficulty lifts.
Let go of the need to change what has occurred, and be philosophical. An adverse event happened, but it has passed. What’s left is the fallout, and it’s time to sweep up and cope without yelling about the injustice you’ve suffered, aloud or in your imagination.
Mostly, second arrows take place in the dark. They roughhouse in people’s heads, thumping against the sides until a headache ensues. But, they exist, just the same as when people rant and rave about a situation they can’t change.
Angry outbursts, worries, and self-defeating talk are second arrows. Once you recognize them, you will see how much they harm you and stop creating them in favor of greater peace of mind. | https://medium.com/recycled/buddhist-parable-to-reduce-suffering-5dfce7b94897 | ['Bridget Webber'] | 2019-11-02 15:45:27.537000+00:00 | ['Personal Development', 'Lifestyle', 'Mindfulness', 'Psychology', 'Meditation'] |
Change sucks. Until it doesn’t. | Change sucks. Not the kind of change you planned for and now it’s here. The kind of change you didn’t see coming and it disrupts your normal. You are expected, known, beloved normal. I know maybe two people who will tell you they love that kind of change. Either it’s true there’s a really weird, really small splinter group of people in the world who actually embrace uncontrolled change or this same splinter group of people is lying. I say they lie. But even if they don’t, they have to be talking about unexpected changes for good. Like the meh boyfriend deciding to move to Italy. Or being offered the job of your dreams and you get to move to Italy.
Not the unexpected change you know could have decent eventual outcomes but only after a lot of work you really don’t want to do. Not the change you know will make your work more difficult. Not the change you want to undo but can’t. Not the change that takes the foundation of how you’ve always lead the people who depend on you and turns it on its head.
Change is hard.
Before I became executive director at my current organization, I’d been there two years in a department director position. There were two other directors I worked the most closely with and we’d become close, both personally and professionally. We trusted each other’s judgment, we respected each other and we were good joint decision makers. We had compatible personalities and liked each other. Apart from work, we went on annual girls’ weekends. We were friends in addition to being one of the best teams ever. We were the proverbial well-oiled machine.
This is working, let’s not change it.
After I took the executive leadership role, we continued to grow steadily as we developed new programs and grew the senior management team. We were able to hire more employees and serve more people. The original core three of us had worked our butts off to establish a bigger footprint in our community. We had been successful. It never occurred to me our three-person force for good would ever not exist.
It didn’t occur to me until one-third of us closed the door to my office one day and told me she was leaving for a different position. I literally couldn’t talk. It was awful. I cried, then I stopped crying, then we talked, then I said I understood, then I didn’t say I felt betrayed, hurt and a little scared.
Change sucks.
It’s easy to be more comfortable with the known and expected, the things that don’t change. So much of life, whether it’s work or personal, has a certain built-in rhythm. We sleep during particular hours. We have morning routines and evening routines. My mornings have a kind of ritual from when I get up to when I leave for work. After work I go to the gym unless it’s Thursday. Thursday is my night to cook for our neighborhood supper club. The rhythms of my life.
The rhythms at work are different but they are still distinct. Schedules change, tasks and meetings vary, and any given day will bring an unexpected adventure. But even when calm turns into chaos, routine turns into “nope, we have to deal with this now”, we know those things are going to happen. Its predictable in its own way. And the bones of who we are and what we do remain constant. The people and the infrastructure we’ve built ensure we know what we’re doing until this thing you didn’t see coming rocks your world.
What if the worst thing happens?
When one-third of us resigned, I knew in my heart things would never be as good as they had been. The kind of synchrony we had couldn’t be replicated. There would be fall out. How could I close the gap? Could I close the gap? Oh. my. god. This was the litany running through my head.
I was afraid of what was to come. I have made myself crazy more than once waiting on the worst thing. It took a long time to learn to separate fear from reality when change hits me like a tidal wave, meaning without my permission and beyond my control. Although instinctively I wanted to go hide for a while, some decisions had to be made. I was still expected to lead. I was still expected to be a good leader. We promoted a fully capable mid level manager who was ready to take on more. As she learned the position, nothing exploded and nobody went screaming into the night. Our team became different, not worse.
The worst thing didn’t happen.
Change can gradually become your non-enemy. The other third of our original trio retired a few years ago. The entire organization held its breath. Nobody thought we would actually implode the day after she retired, never to exist again, but we were fighting the feeling that life as we knew it was over. The irony is life as we knew it was over. It’s what change does. It, well, changes things. The super irony was, we were fine.
We missed her like crazy. There was anxiety to get past. The waiting game…will the new guy be OK? Will he want to change everything? Is he as nice as he seemed in the interviews? Yes. No. Yes. Look at us. Working our way through change. Even me.
Maybe I’ve been giving this change thing too much power. Can change be good?
We opened an art reuse retail store five years ago as a social enterprise to support our mission. We sell donated art and crafting materials and supplies, new and gently used, at deeply discounted prices. We employ people with and without disabilities. We have two store managers, one who has been there from day one and the other four years. The work of the store is to both supplement the bottom line of the larger organization and to provide employment and employment training for people with disabilities.
Our managers came without experience in the world of disability and yet their immediate embrace of equality, accommodation, independence, respect and support was instant and remarkable. I have unending admiration for them both.
I found out the other day we’re losing 1.5 of them. One is moving on to a different job. One is cutting back to have more time for personal life. I didn’t cry. I didn’t feel betrayed or angry or hurt. I felt so grateful for what they have given, so proud of what I know they will do in these new chapters. I’m a little anxious about this change, but I know at least three things. The worst thing won’t happen. I will be watching for the good changes coming. I will learn from this change and without a doubt will be better for it.
Change sucks until it doesn’t. The only way to get to “it doesn’t” is to let go of the fear. Maybe you even embrace it eventually. Maybe, like me, you never love it but you learn to ride its wave. Big gnarly change may always piss you off a little, but in the end, you have to know who’s gonna win. It’s you. | https://medium.com/publishous/change-sucks-until-it-doesnt-3da3c0e1bbb | ['Donna Goodaker'] | 2019-04-03 21:24:42.773000+00:00 | ['Change', 'Leadership', 'Entrepreneurship', 'Leadership Development', 'Leaders'] |
The Unusual Perks of Mental Illness | The Unusual Perks of Mental Illness
Sometimes, a major weakness is a major strength.
Photo by Viktor Forgacs on Unsplash
Mental illness is a major component of my life. It affects so many aspects of how I live that it’s become an integral part of who I am. Huge life decisions are made taking into account my bipolar.
And, like any illness, there are major downsides that have a huge impact on a given day. Bouts of depression can wreck multiple months of my life. A run of mania can lead to bad decisions. The ever-present anxiety makes simple tasks hard.
These symptoms can cause me great difficulty and pain. Depression sucks, and there’s no varnishing that truth. However, not all aspects of my illness are negative.
That’s not to say that my symptoms are good. Being anxious, depressed, manic, or neurotic are all crappy feelings, and I don’t particularly care for them. However, they all inform the wiring of my brain and are all a part of me in some way. Because these symptoms affect my brain functioning, they naturally inform how I work in unique ways.
I am a neurotic person. I tend to fixate on past mistakes and negative feedback and events, and they tend to ruin my wellbeing when this fixation flares up. I also obsess over every possible outcome of relatively simple conversations, driving myself to madness over a hundred hypothetical scenarios that will never happen.
These same traits assist me at work. My fixation on mistakes and aversion to negative feedback translates to attention to detail in my work, meaning I make fewer critical mistakes in my writing. My obsession with every possible outcome translates to complex processes and mechanisms that guide my work process and help me succeed at my job.
Other traits that negatively affect my life have positive outcomes in other ways. Being late to anything tends to make me panic, and I tend to be hyper-fixated on time. It certainly helps me get to meetings and appointments on time if not early, which reflects well on me as an employee and a professional.
I also operate using strict routines that drive my daily life. For example, I am fully conscious of how much time it takes to get to work and how the traffic signals operate on my whole route. This fixation on routine can ruin my day if I’m running late, but also helps me manage multiple deadlines using standardized practices I’ve developed.
If I’m in the midst of a manic streak, I tend to act recklessly and make poor life choices. I am also incredibly productive with my work and have launched several side projects from manic states.
Not all symptoms of my bipolar can be made unequivocally positive. Depression is depression, and it’s hard to put a shine on it. However, often when coming out of my depression, I feel a sense of renewal that helps me jump-start new ideas or better self-care practices. The deep, dark introspection of the depressive state often leads to major revelations when you pull out of it.
Not everyone is capable of putting a positive spin on their symptoms. I am fortunate that my symptoms are well-controlled by my meds, which allows me to apply their positive effects at work and in life. However, others may not be able to control their symptoms well enough to reach this point, and not all symptoms have positive sides.
Everyone is wired differently, and for some, it’s hard if not impossible to achieve an upside. That’s not a personal failing on their part in the same way that mental illness is not a moral failing. Sometimes, your illness prevents any potential upside, and sometimes, your illness doesn’t have an upside at all. That’s okay.
For me, I do my best to pull value from my illness. It’s a blessing and a curse in that way — my biggest weakness feeds my greatest strengths and enables me to be better.
It’s not always sunshine and roses, though. When I’m hyper-productive during a manic phase, I still have self-destructive tendencies that affect my life negatively. My drive to be timely is fantastic, but not when I’m having a panic attack in my car because I’m running five minutes late.
It’s just a matter of give and take in my mind. I was born with my bipolar and am forced to live with it, but I’ll be damned if it doesn’t give me something in return.
I’ve always been a bit more self-aware than normal, and years of therapy and forced introspection have helped build on that skill. I’m also an analytical person, preferring logic and processes and systems. That may or may not be a product of my neurotic side, but it’s part of me either way.
So, being self-aware, introspective, and analytical, I’ve dedicated a lot of time in my life to finding ways to exploit the complex system that is my brain. How can I make my negative qualities work for me as well as against me? How can I channel my problems into solutions for other things?
This may seem like trying to draw blood from a stone. I’m cursed with a mental illness, after all. How can that be good? As I’ve described above, there’s a lot of good to be had from my more self-destructive tendencies. It’s just a matter of finding it.
Turns out there’s plenty of blood in this stone. | https://medium.com/invisible-illness/the-unusual-perks-of-mental-illness-18d250c614e2 | ['Matthew Maniaci'] | 2020-01-24 16:34:39.869000+00:00 | ['Life', 'Bipolar Disorder', 'Depression', 'Anxiety', 'Mental Health'] |
Technology is a tool, not a human barrier. | Technology is a tool, not a human barrier.
Technology, like bows and arrows, exists to make our lives easier. They help to improve the quality of our lives.
Photo by Mimi Thian on Unsplash
We should not view technology as a firewall between human interaction.
As our productivity level increases, we need to make a conscious effort to spend time with the people we care about.
That should be the way. | https://medium.com/technology-hits/technology-is-a-tool-not-a-human-barrier-c37440706726 | ['Aldric Chen'] | 2020-12-20 09:01:06.399000+00:00 | ['Self Improvement', 'Productivity', 'Life Lessons', 'Thinking', 'Technology'] |
We present the first project of Crypterium Family — InstaChange | To bring mass adoption of cryptocurrencies closer, Crypterium is creating the whole range of products that fundamentally change the cryptocurrency usage experience. We are on the way of making any crypto-related activities easier, more secure, and accessible. This week, we’ve launched the first of many coming services. It allows anyone to buy crypto at best rates within seconds. Without further ado, please meet InstaСhange.
From the beginning, Crypterium has been comprehensively addressing the issues making anything related to crypto a bit of a pain. Today, everything is so complicated, time consuming and expensive, that only very tech-savvy people can find their way around.
We want to change that, by building a whole new ecosystem, that we like to call Crypterium Family, and that is making any operation with crypto easy. From buying crypto and exchanging it to spending crypto it in everyday life and pledging it to get loans — we’re building satellite products that will make dealing with crypto at least as easy as with traditional financial products.
The first project we’d like to introduce to you is called InstaChange. It offers you the easiest and most convenient way of entering the crypto world. The idea is simple: before you can start using crypto in everyday life, you need to get some crypto. InstaChange lets you buy Bitcoin, Ether, Litecoin or Tether straight from your bank card or bank account — instantly and with minimal fees.
Crypterium Family concept
Crypterium is currently working on a bunch of new products that would ideally fit in the world’s first cryptobank ecosystem — Crypterium Family. They won’t immediately go inside the Crypterium App — we’ll first run them in parallel and properly test them — but eventually, they’ll be merged together.
“We are going to do a lot of additional services, and within that range of services we have an Innovation Team that is working on launching new products and one of them — InstaChange — is finally launched” — says Austin Kimm, COO at Crypterium.
Maybe it’s time to give our first satellite a try while crypto prices are the lowest they’ve been for a whole year. To make the deal even sweeter, we offer you 0% for any purchases made by November 25. Let’s celebrate the first CF project together!
About Crypterium
CCrypterium is building a mobile app that will turn cryptocurrencies into money that you can spend with the same ease as cash. Shop around the world and pay with your coins and tokens at any NFC terminal, or via scanning the QR codes. Make purchases in online stores, pay your bills, or just send money across borders in seconds, reliably and for a fraction of a penny.
Join our Telegram news channel or other social media to stay updated!
Website ๏ Telegram ๏ Facebook ๏ Twitter ๏ BitcoinTalk ๏ Reddit ๏ YouTube ๏ LinkedIn | https://medium.com/crypterium/instachange-the-first-project-of-crypterium-family-e3424fb8ca41 | [] | 2018-12-25 12:48:44.174000+00:00 | ['Cryptocurrency', 'Bitcoin', 'Mobile App Development', 'Exchange', 'Technology'] |
Star Wars Symbology: The Plot Twist George Lucas Kept Secret For Forty-Five Years | Of All Things, Why No Force Flight? A Theory.
Why? My theory explaining why we haven’t seen canonical Force Flight yet, is that we simply haven’t seen a Jedi powerful enough yet.
Hard to believe considering what Luke pulled off in The Last Jedi.
But in The Last Jedi, we were exposed to “The Prime Jedi.”
The Prime Jedi shown in the mosaic on the floor of the Jedi Temple on Ahch-To depicts a Jedi that is equal parts light and dark. Like a yin and yang.
This surprised some Star Wars fans, myself included, considering that we were told all along that it is a battle of Light versus Dark — not the integration of Light and Dark.
But we know that the Star Wars Canon has a precedent called the Gray Jedi.
The Gray Jedi was “a term used by Jedi and Sith to describe Force-users who walked the line between the light and dark sides of the Force without surrendering to the dark side.”
There is a strong indication from the trailers that The Rise of Skywalker is heading toward the Gray Jedi.
Rey’s survived the darkside’s call to mirror cave in TLJ, but will it be enough dark to balance her light?
And when we analyze the symbolism of the Gray Jedi it makes sense.
True power comes not from naively cleaving to the notion of Good given to us by other. That form of Good is said to be a form of cowardice.
True power comes from knowing your own capacity, as a human, for malevolence, evil and destruction — but choosing to be an agent for Good instead. Some may argue that that choice, that free will, is what Life is about.
This is the core essence of “strength in balance.” The Gray Jedi will be the most powerful we’ve ever seen. Part of his power will be from his self-belief. | https://medium.com/jung-skywalker/star-wars-symbology-the-plot-twist-george-lucas-has-kept-secret-for-45-forty-five-years-c859476fc9e | ['Brian Deines'] | 2020-04-29 16:09:13.096000+00:00 | ['Philosophy', 'Star Wars', 'Cinema', 'Storytelling', 'Film'] |
Garden of Eden Revisited | Hey you
Demon days are over
Relax
Stop looking over your shoulder
It’s okay now
I know it’s hard to trust
A change that’s been so long coming
And let go of the fears
That were so long plumbing
The depths of the subconscious
But really they were just monsters
Of your imagination
Time to let them go
And explore the rest of creation
This world is the garden of Eden
It’s all just a matter for perception
Just adjust your focus
And let the light
Play on your pupils
and rest. | https://medium.com/poets-unlimited/garden-of-eden-revisited-19dc214953c | ['John Horan'] | 2017-03-21 16:42:11.632000+00:00 | ['Life', 'Change', 'Christianity', 'Poetry', 'Mental Health'] |
Where Everybody Knows Your Face | Sign up for Taking Stock of COVID: New comics from The Nib
By The Nib
New comics from Matt Lubchansky, Keith Knight, and Niccolo Pizarro. Plus How Seniors Are Living With COVID-19. Take a look | https://medium.com/the-nib/where-everybody-knows-your-face-f444bd071af0 | ['The Nib'] | 2020-12-10 15:55:19.848000+00:00 | ['Racism', 'Privacy', 'Comics', 'Capitalism', 'Amazon'] |
‘No sweat’ said my EarPods, just before they drowned | ‘No sweat’ said my EarPods, just before they drowned
Making identical products for different conditions is a bad idea
Touchscreens and wet fingers don’t gel (Photo by Vanderlei Longo from Pexels)
Most of the electronic gadgets I get in India seem to be designed for countries with cooler climates where humidity is not a factor. Looking at it that way, I would say there is still room for growth in the stagnating mobile phone industry if phone makers focus on customizing their phones for different markets. In India, I would say price sensitivity is a key factor affecting sales. In fact, Apple has recently accepted that its high pricing strategy won’t get it anywhere in India and is beginning to look into assembling iPhones in India in order to avoid the high import taxes.
However, weather conditions don’t seem to be getting much attention. The recent models of iPhones may be waterproof, but touchscreens don’t work too well with sweaty fingers. As for the bundled accessories like headsets, they aren’t even waterproof, and this is an issue.
Dew point and muggy days
Indian Summers are not something you look forward to in India. It’s not yet peak summer and I stay in a city located around 400m above sea-level. But temperatures are already nudging 40℃ (100℉). However, what really gets me is the oppressive humidity, which leaves me drenched in sweat and wreaks havoc with my gadgets.
Here’s what the weather was like this evening at my place. The key figure is the dew point as explained by the linked website, which I quote below. That 22℃ (72℉) gives an idea of how muggy the weather is in India.
…once the dew point gets between 55 F and 65 F, the NWS says the outdoors will feel “sticky with muggy evenings.” Anything above 65 F means there’s a lot of moisture in the air, and most people will start to feel uncomfortable. Once that dew point temperature hits 70 F (21 C), things are getting oppressive, if not outright dangerous. High dew points are uncomfortable because the air’s moisture is slowing down the rate at which our sweat evaporates off our bodies. It’s how we cool down. So, if you’re in some place with a very high temperature and a low dew point, your body is going to sweat and that sweat will evaporate. It’s also very easy to become dehydrated in this situation.
Killer Sweat
Normally, I use my gadgets indoors, where ACs and fans take the sweat out of the equation. However, this changed when I began jogging a few years ago. Like lots of amateur joggers, the music on my headphones is what keeps me going during my 5km morning run. Though a simple headset without a mic can do the job, a headset with remote controls allows me to easily control the sound volume, and skip tracks until I find a song that matches the cadence of my footsteps pounding the pavement. That’s a lifesaver on days when I lack energy as the beat motivates me to maintain my pace and complete my run.
The Apple Earpods that come with iPhones is perfect for this task. They allow me to control music and volume and even take calls while jogging. Unfortunately, Apple’s Earpods are not waterproof. I found this out the hard way when its buttons stopped working midway through a sweaty run. The sound cut off as I fumbled with the remote to change a song. Once I got home, I tried to fix the Earpods by cleaning them as best as I could but to no avail. My guess is sweat must have entered the remote, and corroded the electrical points inside. I even tried burying the headset in a sack of rice as it’s supposed to be an effective way to absorb the moisture. No luck. After waiting patiently but unsuccessfully for a couple of months, I mournfully dropped the dead headset in the dustbin.
After that, I switched back to my old iPod’s EarPods, which come without remote controls. The Earpods are not waterproof either but they kept going, probably because there aren’t too many sweat glands inside the human ears. But I missed my remote. It was a pain to pull my phone from my pocket while jogging every time I needed to change a song or increase/reduce the volume.
Around this time, Bluetooth headsets started becoming affordable. I tested the waters with a cheap Chinese brand from Amazon. It lasted around a year before it stopped working, and I assumed that it was due to its poor quality. On hindsight, it might have again been the sweat. A little while later, I spotted a JBL set going for under ₹2000 ($29), I gave in to temptation and got myself a pair, conveniently ignoring the fact that it didn’t mention anywhere that it was sweatproof or even waterproof. That was in December 2017.
In South India where I live, the major difference between winter and summer is a slight drop in temperature and humidity. Basically, you sweat less in winter. Moisture, however, didn’t seem to be an issue for my new Bluetooth headset. I relished the freedom of once again going wireless, though the JBL’s remote could only control the volume. Maybe it was luck or maybe the summer of 2018 wasn’t that humid, and the JBL survived a year of running.
Just as I was beginning to think sweat was no longer an issue, my wireless world came crashing down. Or up, to be precise. It happened midway through a sweaty run a week ago. The volume on my JBL headset suddenly started shooting up to deafening levels. I thought maybe a volume button was being pressed accidentally somewhere. But the buttons on my phone and the headset’s remote didn’t seem to be pressed. However, pressing the buttons on the JBL’s remote got me no response. So I pulled out my phone and used the physical volume button on the phone’s side. It worked. But as I was about to put my phone back, the volume began to rise again, and I could see the volume slider on my phone’s lockscreen beginning to move right again as if being pulled by some invisible hand. So I unplugged my ears, turned off the Bluetooth, and resigned myself to a run sans music. The next day, I went back to my mic-less Earpods but it was cumbersome as I was used to being wireless.
However, I had almost lost three headsets to sweat. Something had to change.
I say ‘almost’ because somewhat surprisingly, the JBL headset regained its full powers after a few days, with all buttons becoming functional. However, I didn’t want to push my luck by taking it out jogging again.
A case of mistaken identity
It’s not just headsets which have an issue with sweat. Take Apple’s Touch ID. This is a wonderful innovation but it too doesn’t work too well when it’s humid and your fingers are sweating. The thing is unlike wet fingers, sweaty fingers won’t get perfectly dry by wiping them as sweat is being constantly generated. This can lead to a maddening experience where my phone stubbornly refuses to believe I am me. If I’m lucky, my phone will sometimes relent and give me the option of entering a passcode to unlock it. If it’s a bad day, I will have to keep jabbing away at the home button and onscreen buttons for a while till you get the passcode option, and then hope you can type out your passcode with your sweaty fingers.
In fact, one of my sad memories is the day I beat my personal best timing for a 5 km run by going under 28 minutes. But my clock says I crossed 28 minutes, simply because the phone refused to recognize my sweaty finger desperately swiping away at the ‘stop’ button on my running app.
Of course, there are workarounds. There always are.
Saving a wet fingerprint sometimes works. But ‘sometimes’ is not a good solution.
Upgrading to a phone with Face ID is another option. This can get you past the lockscreen even if you have sweaty fingers. But you still have to deal with the maddening ‘swipe-that-won’t-swipe’ that happens with sweaty fingers. Besides, Indian taxes make iPhones cost more than anywhere else on earth. Like a 64GB iPhone XS Max costs ₹106,999 ($1547), and that’s after a recent price cut of ₹4000. There’s also the fact that my iPhone 6S+ is fine apart from this sweaty issue. It should easily last another couple of years with the battery change I made a couple of months ago. It just doesn’t make sense to upgrade my iPhone so I can get past my lockscreen when my fingers are sweaty.
AirPods are another option as you can bypass the touch function, by using voice control with Siri. But again AirPods are an expensive solution as the new AirPods 2 with standard (wired charging) casing costs ₹14,900 ($215).
The need of the hour is an economical solution.
Sports headsets
Earlier this year, I made a resolution to avoid adding to the world’s growing electronic junk by cutting down on my consumerism. But this is a genuine need. So I ordered a sports headset from Xiaomi for a tenth the price of the AirPods at ₹1499 ($22). What makes the Mi a sports headset is its IPX4 splash and sweat-proofing.
Xiaomi is a Chinese company that has a reputation for making relatively good quality, value-for-money phones and accessories under the Mi, Redmi and Poco brands. My short personal experience, sort of backs that up. I have been using my sweat resistant Bluetooth headset for three days, and I’m quite impressed with it.
The Mi is more suited than the JBL for jogging, and it’s not just the IPX4 rating. The JBL has two components, a battery and the remote component attached to its headset apart. These bounce around uncomfortably on my neck when I go jogging, and the sound of them bouncing off my neck clashes with the music. It takes some practice to tune out the disturbance.
On the other hand, the Mi’s short cable and light remote unit (which combines battery and remote) mean it doesn’t bounce around on my neck like the JBL, and interfere with the music. Xiaomi has also made track skipping work by a long press on the volume buttons, something which normally won’t happen on cheaper, non-Apple devices.
Mi also claims the battery will last for 9 hours. I doubt this but I think there will be substantial savings in power consumption with the Mi as they have done one simple thing. Unlike the JBL and most other Bluetooth brands, which have a bright ‘always on’ light to serve as an indicator that the device is turned on, the Mi has a lower power ‘on’ light that blinks once every ten seconds or so.
Making things better
In that sense, a lot of what Xiaomi did with the sweat-proof headset is just common sense. But I like the thought process by which they fine-tuned the gadget for a jogger’s needs.
I must add that the Mi products do have their disadvantages. Xiaomi has to cut corners to keep the price down on its products. Like I have a Mi Band which I got in Sep 2016 for ₹2000 ($29). I use it to keep time, count my steps, and vibrate to alert me when my phone is in silent mode and someone calls. It has done this job well for two and a half years. But the screen began to go dim a couple of weeks ago, and I will have to replace the whole gadget. In comparison, my Apple products seem to last forever, with even my old clickwheel iPod still in sound working condition. But the way I look at it, $29 for a product that needed to be charged just once a month, yet delivered non-stop service for 30 months is pretty good value for money.
Coming back to the Bluetooth headset, the Mi’s cable doesn’t feel as sturdy as the JBL’s flat, tangle-free one. The JBL also has magnets in the earpieces to hold them together around your neck, unlike the little clip that the Mi uses. In fact, the overall build quality of the JBL just feels better. But the reality is the flimsier Mi is actually far better at the job of being a jogger’s headset.
In that sense, Xiaomi has taken a page out of Apple’s Playbook of taking an existing product and creating a better version of it. But Mi has an edge on Apple as it manages to do this while keeping its price lower than competitors instead of higher. This explains why Apple has only a 1% share of India’s cellphone market while Xiaomi is the market leader with 28.9%.
To sum up, if Apple and other phone makers apply Xiaomi’s simple principles, and tackle the specific needs of customers in different locations, I see no reason why they can’t reverse the current global trend of falling phone sales. | https://medium.com/hackernoon/no-sweat-said-my-earpods-just-before-they-drowned-5ab3719a3fa5 | [] | 2019-11-11 08:31:42.594000+00:00 | ['Sweat', 'Electronics', 'Water Damage', 'India', 'Apple'] |
Remove duplicates from two Pandas DataFrame | During my research on Indeed for classifying the job types, found an issue regarding multiple labels for the same job i.e. temporary and part-time are assigned to the same jobs. I used the following code to remove common descriptions of jobs with different labels.
import pandas as pd part_time = pd.read_csv("part_time.csv", index_col=0)
temporary = pd.read_csv("temporary.csv", index_col=0) # find common jobs using description column with isin() function.
# A intersection B
A = part_time[part_time.description.isin(temporary.description)] # remove common elements from both part_time and temporary jobs.
# temporary - A
# part_time - A temporary = temporary[~temporary.description.isin(A)]
part_time = part_time[~part_time.description.isin(A)] # now concat these two data frames and save.
total = pd.concat([temporary, part_time])
total.to_csv("indeed_jobs.csv", index=False)
Hoping it will help those who have the same issue. | https://medium.com/nlp-for-noobs/remove-duplicates-from-two-pandas-dataframe-3d8b18d0f0b0 | ['Muhammad Irfan'] | 2019-04-17 07:46:57.606000+00:00 | ['Python', 'Data Science', 'Remove Duplicates', 'Pandas'] |
Saving Indus River Dolphins — Pakistan | The role
My name is Hamera Aisha and I work as Wildlife Manager at WWF-Pakistan. My job involves working very closely with communities, governments, and other relevant partners to protect the Indus river dolphin and many other vanishing wildlife species of Pakistan and their habitats. My work also includes raising awareness about the planet and country’s biodiversity and engaging with the public, especially children (students), to help develop an association with nature, the need to care for wildlife, such as the rare dolphin, and the need for its protection.
New hope for the Indus River dolphin
The Indus river dolphin is the second most endangered freshwater dolphin in the world, due to many threats it faces including fragmentation of their habitat after barrages were built across the Indus River, entanglement in fishing nets, reduced flow of water in the river, and pollution from surrounding industries.
In 2001, WWF conducted the first Indus River dolphin survey and the chances of survival for this endangered river dolphin were bleak. That year, only 1,200 individuals counted were identified. What followed were concerted efforts by WWF, provincial wildlife departments and communities to protect the dolphin.
Fifteen years later, a survey showed that these conservation efforts were successful and 2,000 dolphins were counted, showing the population is on the rise for the first time. This conservation work also involves physically rescuing dolphins trapped in shallow waters and carefully moving them to be released into the Indus River where they can survive. The 2017 survey is a sign of hope and a reminder that we need to continue our efforts to protect dolphins and explore innovative solutions for their conservation.
Who are the rangers who help with dolphin rescue?
© WWF-Pakistan / Uzma Khan / WWF
The Indus dolphin habitat is spread across three provinces and hence wildlife departments from each province play an important role in the conservation of the species.
The role of wildlife rangers is a good example of the commitment of communities towards saving the wildlife of Pakistan. Nazir Mirani is one such inspirational ranger who has spent well over 25 years serving the Sindh Wildlife Department, rescuing dolphins in the lower Indus River, which hosts the highest dolphin population.
Rescuing dolphins can be a tricky job. Nazir’s hands are full of dolphin bites from various rescue operations he has been a part of. But according to him, he cannot imagine a life without dolphins around him. He is an excellent swimmer and is joined by a team that locates dolphins in shallow water bodies. After a dolphin is spotted, Nazir ensures its safe translocation to the main river channel in a soundproof ambulance.
Over the years, WWF along with communities and the Wildlife Department has devised a standard mechanism to rescue these dolphins, which Nazir and his team follows. More than a hundred dolphins have been rescued in the past few years.
There are other community watchers who facilitate the Wildlife Departments in monitoring and dolphin rescue operations. But these rangers work with limited resources, without proper equipment and necessary training; most are unfamiliar with work-related safety measures. Their jobs involve patrolling along the Indus River to look for illegal fishing activities, which can entangle dolphins in nets and lead to mortality. The rangers involved in dolphin rescue operations need proper swimming gear and training on how to use it.
Are there different roles for other partner organisations?
© François Xavier Pelletier / WWF
The dolphin conservation initiative is a great example of participatory conservation which integrates research and community-based conservation while addressing socio-economic challenges as well as effective coordination and enforcement.
WWF works in close collaboration with the wildlife departments and other relevant organisations to facilitate dolphin conservation, especially the rescue efforts.
A dolphin monitoring network including the representatives of Sindh Wildlife Department (SWD), Sindh Irrigation Department (SID), Sindh Forest Department (SFD), Sindh Fisheries Department (SFD), Sindh Environmental Protection Agency (SEPA) and local community activists, who monitor the Indus River as well as its adjacent canals and tributaries to rescue any stranded dolphins.
How do rescue teams find out about a trapped dolphin?
WWF has helped set up a hotline where communities report strandings of dolphins to WWF and the Wildlife Department. The rescue team then arrive on a dolphin rescue ambulance.
Stranding of Indus River dolphins in canals is amongst the key threats this endangered species is facing. When canal gates are closed, the level of water drops and these canals turn into small ponds. The dolphins are trapped in these ponds, unable to swim back with only limited food supply.
WWF-Pakistan has set up a 24-hour phone helpline for fishermen and the rest of the community to report any incidents of stranded river dolphins to WWF-Pakistan and the Sindh Wildlife Department for timely reporting and rescue of the dolphins.
What does a dolphin rescue look like?
© François Xavier Pelletier / WWF
All rescues are conducted by the Wildlife Department with the help of community members. Once the team arrives on the dolphin rescue ambulance, they spot the dolphin. The dolphin is then manually lifted out of the canal, placed on a stretcher and taken to a soundproof ambulance.
The Indus dolphin is a sensitive species and is vulnerable to stress during the rescue and translocation process. If standard procedures are not followed, the Dolphins face a high probability of injuries.
The Indus dolphin relies on its highly developed sonar and hearing to navigate. The sound and movement of vehicles can cause additional stress. If the team is not careful, the dolphin could even die. This all makes it extremely critical to keep the animal calm once it is outside the water.
WWF-Pakistan’s specialized ambulance is equipped with necessary equipment including a portable stretcher specifically designed for dolphins, to ensure the dolphin’s safety during the rescue. The ambulance is completely soundproof and has a water tank with an electric motor on the roof to carry river water used to moisten the dolphin’s skin through a shower.
My inspiration
The Indus river dolphin is my inspiration and I have an emotional association with it due to its continuous struggle for survival despite various conservation challenges, and because of its calm and lovable nature. I take pride in the fact that I am one of the few women of Pakistan who was able to join the survey and count dolphins throughout their habitat in the remote areas of Pakistan. In order to do so I, along with a number of other colleagues, spent close to a month living and working on the Indus River, making camp on the shore to sleep at night. This work has been very helpful to understand that dolphins have increased in number as an outcome of the continuous conservation efforts in Pakistan and also for future planning for their better conservation.
This experience has inspired me to connect more with communities and to better understand the challenges which dolphins face. I am fortunate enough to have been able to continue to work close to the dolphin habitat in the south of Pakistan to address direct and indirect threats to the species and its habitat. My current work for example on the poaching and illegal trade of freshwater turtles also focuses on the Indus river system within the same habitat. | https://wwf.medium.com/saving-indus-river-dolphins-pakistan-ff1bb290aef | [] | 2019-07-31 00:01:01.252000+00:00 | ['Environment', 'Dolphin', 'Conservation'] |
numpy.random — Generating Random Numbers In Python | Random number have lots of application like, cryptography, game and Neural Network. Here we’ll discuss and identify different methods for generating random numbers in NumPy module in python.
import numpy as np
numpy.random.random (size)— Return random floats in the half-open interval [0.0, 1.0), ie 1.0>x≥0.0
>>>np.random.random((2,2)) array([[ 0.66032591, 0.91397527],
[ 0.63366556, 0.36594058]])
numpy.random.randint(low, high, size, dtype=int) — Return random integers from low to high from the “discrete uniform” distribution.
>>> np.random.randint([1, 3, 5, 7], [[10], [20]], dtype=np.uint8) array([[ 8, 6, 9, 7],
[ 1, 16, 9, 12]], dtype=uint8)
numpy.random.randn(d1,d2,..dn) — Return a sample/s from the “standard normal” distribution, for given dimension.
>>> np.random.randn(2,2) array([[ 0.6762164 , -1.37066901],
[ 0.23856319, 0.61407709]])
numpy.random.rand(s1,s2,..sn) — Random values in a given given dimension and ranges between (0,1).
>>> np.random.rand(3,2)
array([[ 0.14022471, 0.96360618],
[ 0.37601032, 0.25528411],
[ 0.49313049, 0.94909878]])
random.rand and random.randn are similar function, both in functionality and syntax and can be confused easily.
random.rand — Returns sample according to standard normal distribution.
random.randn — Returns numbers ranging between (0,1)
numpy.random.normal(loc=0.0, scale=1.0, size=None) — Draw random samples from a normal (Gaussian) distribution.
loc : Mean (“centre”) of the distribution
scale : spread or “width” of the distribution >>> np.random.normal(3, 2.5, size=(2, 4))
array([[-4.49401501, 4.00950034, -1.81814867, 7.29718677],
[ 0.39924804, 4.68456316, 4.99394529, 4.84057254]])
These are the most methods used for creating random number. More details can be searched in documentation. To get broader knowledge on numpy check my article. | https://medium.com/analytics-vidhya/numpy-random-generating-random-numbers-in-python-fc5fff3a1594 | ['Om Rastogi'] | 2020-06-11 16:07:14.607000+00:00 | ['Machine Learning', 'Data Science', 'Numpy', 'Python', 'Random'] |
Using Tableau for Live Dashboards on Event Data | Live dashboards can help organizations make sense of their event data and understand what’s happening in their businesses in real time. Marketing managers constantly want to know how many signups there were in the last hour, day, or week. Product managers are always looking to understand which product features are working well and most heavily utilized. In many situations, it is important to be able to take immediate action based on real-time event data, as in the case of limited-time sales in e-commerce or managing contact center service levels. With the realization of the value businesses can extract from real-time data, many organizations that have standardized on Tableau for BI are seeking to implement live Tableau dashboards on their event streams as well.
Getting Started
In this blog, I will step through implementing a live dashboard on event data using Tableau. The events we will track will be recent changes to various Wikimedia projects, including Wikipedia.
For this project, we will need:
Ingesting Data from Wikimedia Event Stream
I will first create a collection, to which we will write the events originating from the Wikimedia stream.
Once we have the collection set up in Rockset, I can run a script that subscribes to events from the Wikimedia stream and writes them to the wiki-events collection in Rockset.
While we are using Rockset’s Write API to ingest the Wikimedia event stream in this case, Rockset can also sync data from other sources, like Amazon DynamoDB, Amazon Kinesis, and Apache Kakfa, to power live dashboards, if required.
Now that we are ingesting the event stream, the collection is growing steadily every second. Describing the collection shows us the shape of data. The result is quite long, so I will just show an abbreviated version below to give you a sense of what the JSON data looks like, along with some of the fields we will be exploring in Tableau. The data is somewhat complex, containing sparse fields and nested objects and arrays.
Connecting a Tableau Dashboard to Real-Time Event Data
Let us jump into building the dashboard. I’ll first need to connect to Rockset, as a new data source, from my Tableau Desktop application. Follow the steps here to set this up.
We can create a first chart showing the number of changes made by bots vs. non-bots for every minute in the last one hour. I can use a custom SQL query within Tableau to specify the query for this, which gives us the resulting chart.
That’s about 1,400 events per minute, with bots responsible for the majority of them.
Wikimedia also tracks several types of change events: edit, new, log, and categorize. We can get an up-to-date view of the various types of changes made, at 10-minute intervals, for the last hour.
Finally, I plotted a chart to visualize the magnitude of the edits made within the last hour, whether they were small-, medium-, or large-scale edits.
Recap
In a few steps, we ingested a stream of complex JSON event data, connected Tableau to the data in Rockset, and added some charts to our live dashboard. While it may normally take tens of minutes, if not longer, to process raw event data for use with a dashboarding tool, using Tableau on real-time data in Rockset allows users to perform live analysis on their data within seconds of the events occurring.
If you wish to adapt what we’ve done here to your use case, the source code for this exercise is available here. | https://medium.com/rocksetcloud/using-tableau-for-live-dashboards-on-event-data-926b33c81980 | ['Haneesh Reddy Poodutoori'] | 2020-01-28 23:12:54.265000+00:00 | ['Json', 'Dashboard', 'Tableau', 'Kafka', 'Real Time Analytics'] |
The definitive guide to Neural Networks | The most simplest and appropriate way to define a Neural Network, more precisely, an Artificial Neural Networks(ANN) was proposed by the inventor of one of the first neurocomputers, Dr. Robert Hecht-Nielsen. He defines a neural network as:
“…a computing system made up of a number of simple, highly interconnected processing elements, which process information by their dynamic state response to external inputs.”
An Artificial Neuron Network (ANN) can be understood as a computational model based on the structure and functions of biological neural networks. It is very similar to an artificial human nervous system that receives, processes, and transmits information in terms of Computer Science.
In order to explore more on Neural Networks, Let us see what builds a neural network. To get a more clear idea, let us know in detail about a neuron.
Neurons, the building blocks
A neuron is a unit that takes inputs, does some math with them, and produces an output. Here’s what a 2-input neuron looks like:
Let us see what happens here.
First, multiply each input by a weight
Next, we need to add all the weighted inputs together along with a bias b
Finally, we pass the sum to the activation function.
This is the basic structure of a neuron. Neurons are the building blocks of neural networks. Now, you might have got an idea about a single neuron. Now let us understand a network of neurons.
NEURAL NETWORKS
A neural network is a bunch of neurons connected together. Here’s what a simple neural network might look like:
We have 3 different layers in a neural network:-
Input Layer Hidden Layers Output Layer
Input Layer
The Input layer deals with the communication between the external environment and the neural network. This input is transferred to the hidden layers. Every input neuron will represent an independent variable that has an influence over the output of the neural network.
Hidden Layer
The hidden layer is the intermediate layer which is a collection of neurons having activation function applied to it. It is found between the input layer and the output layer. Its job is to process the inputs obtained by its previous layer and extract the required features from the input data.
Output Layer
The output layer of the neural network collects and transmits the information in a way it has been designed to give. The pattern presented by the output layer can be directly traced back to the input layer. The number of neurons in the output layer should be directly related to the type of work that the neural network was performing.
How to implement a Neural Network?
A neural network can be executed in 2 steps :
1. Feedforward:
Talking about feedforward neural networks, we have a set of input features and some random weights. This sets up the base of a neural network.
2. Backpropagation:
During backpropagation, we calculate the error between the target output and the predicted output. After that, we can use different algorithms such as gradient descent to update weight values.
Why do we need backpropagation?
While designing a neural network, we should train the model and assign specific weights to each of the inputs. That weight that we assigned decides how vital that feature is for our prediction. The higher the weight, the greater the importance. However, we do not know the specific weight required by those inputs in the initial period. So random weights are assigned to our inputs, and our model calculates the error in prediction. Thereafter, we update the weight values we choose and rerun the code. This is known as backpropagation. After individual iterations, we will get lower error values and higher accuracy.
The gist of Neural Networks
Take required inputs Add bias to it Assign weights randomly to the input features Run the code for training and note the error in prediction. Update the weights by means of an algorithm. Repeat the training phase with the new weights. Get the predictions.
How to train a Neural Network?
Design the model and calculate the error in prediction See if the error is acceptable If an error is acceptable, the model is ready for prediction Else, update the weights and train again
Photo by Ilya Pavlov on Unsplash
Conclusion
The neural network is the real stimulus behind the existence of the field of deep learning. Here, ’deep’ refers to the multiple layers of a neural network. It also dominates the contemporary research literature in machine learning and many other fields involving data classification and prediction. Understanding neural networks play a vital role in your first step to deep learning. The magical field has yet many more mysteries to be unsolved. Hope you will dive deeper into the deep learning field. | https://medium.com/nerd-for-tech/the-definitive-to-neural-networks-7fe02ff49e9 | ['Nandana Sreeraj'] | 2020-07-13 17:06:19.515000+00:00 | ['Neural Networks', 'Machine Learning', 'Deep Learning', 'Artificial Intelligence', 'Artificial Neural Network'] |
How to: customise Looker Actions & make them work for you | This is a tale of a hero visualisation platform, Looker is its name. Making insights actionable is its fantastic game!
As Looker says, with Looker Actions, you can analyse and act on your data. You can do everything from engaging with your users to updating records in any of the applications you use. From Slack to Segment.io, Looker has a set of amazing third party actions right out of the box. These actions are NOT limited to the ones provided out of the box, and you can build your own.
Actions primarily help users in 3 areas:
Act: The use case could be to create a marketing campaign based on a filtered user list.
Communicate: Send data to a subordinate or different department to be actioned
Deliver: Delivery/tracking of reports in an automated fashion
Let’s look at the power of the magic wand provided by Looker. Using Looker’s action hub, a package to develop custom Looker Actions, we took the Looker user’s experience to another level. Let’s be more specific here. In simple terms, we have built a simple set of microservices to send data from Looks ( click here to understand about Looks in detail) into Salesforce in seconds, in a simple and straightforward way.
So why is it a big thing?
Let’s deep dive into the whole process!
Looker Action
We have a built a custom action for Looker to enable data ingestion capabilities at the Looker end. For discussion’s sake let’s give this action a name. Let’s call it MyAction. MyAction enables client-specific data delivery endpoints to Looker. These actions are automatically attached to the dashboard gear menu in Looker — making it easily accessible for the user. For example, we have an action enabled to send customer churn data into a Salesforce instance which will help the marketer to target/track these churners in Salesforce. After applying relevant filters user can trigger the data delivery job.
See, it’s so simple! Let’s look into the building blocks.
Looker Action Hub provides Looker Action API. This is useful to build an action directly into an existing web service in any language, you can easily have your server implement Looker Action API. The Action API is a simple webhook-like API to accept actions from Looker. Your server will provide a few endpoints that Looker can connect to list and execute actions. By implementing the Action API your server becomes an “Action Hub” that users can connect Looker to. An Action Hub can expose one or more actions.
Going with the serverless approach you can create your own Actions Hub on an AWS Lambda exposed through API Gateway. Looker will send an HTTP POST request with an empty body to this API Gateway endpoint. This endpoint should return a JSON payload with an “integrations” key containing a list of actions.
Here’s an example of a list containing a single integration:
{ "integrations": [ { "name": "my_action", "label": "My Great Action", "supported_action_types": ["cell"], "url": "https://example.com/actions/my_action/execute", "icon_data_uri": "data:image/gif;base64,R0lGODlhEAA....", "params": [ {"name": "api_key", "label": "API Key", "required": true} ] } ] }
You can find the full data structure of the payload here.
When the user clicks on the action from the Looker Dashboard, Looker will run the relevant query and then POST the data to execute endpoint.
For more information check out Looker Action Hub.
Looker Actions can also be implemented directly into views using LookML which makes it sit closer to your dashboard and in case you want to pass liquid parameters or any dynamic values from LookML to your endpoint, this makes it possible.
Here’s an example of implementing action within a view:
view: view_name { measure: field_name { action: { label: "Label to Appear in Action Menu" url: "https://example.com/posts" icon_url: "https://looker.com/favicon.ico" form_url: "https://example.com/ping/{{ value }}/form.json" user_attribute_param: { user_attribute: user_attribute_name name: "name_for_json_payload" } param: { name: "some_auth_code" value: "abc123456" } form_param: { name: "title" type: select | string | textarea label: "desired label name" description: "description text" required: yes | no default: "value string" option: { name: "name string" value: "value string" } } } } }
Check out this link for more on this.
And so, what about when you want to get some input from the user when they click on a Looker Action? You might have noticed in the LookML snippet above there is a key called “form_param”. That’s for opening a pop-up form to get more inputs from the user based on your use case. Forms can be served from your own Action Hub as well.
Salesforce
Dumping data straight into custom Salesforce objects wouldn’t add much value. Instead, we have been using Salesforce API to perform specific tasks, for example, creating a campaign. However, if you still need just the data, it will be delivered to its destination in a disciplined manner.
Moral of the story is that you now have the data for potential churners in Salesforce which you can use to target these customers through personalised messages, monitor their activity, built customer journeys, etc through Salesforce. The Looker dashboard provided the insight which can now be used to take the relevant action and that helps you improve business outcomes.
On the closing note, it is amazing how Looker has provided the tools to close the end to end loop from data, insights to actions. There are a number of use cases to enable using this capability. Combining this with serverless power of AWS you can turn your insights into actions seamlessly and with no management of server code, you can focus more on your users. We have built a suite of Looker Actions interacting with the third-party applications, to know more and understand how we can help you turn insights into actions that you can actually act upon, please contact us. | https://medium.com/acrotrend-consultancy/how-to-customise-looker-actions-make-them-work-for-you-b579c12fee8b | ['Acrotrend Consultancy'] | 2019-08-10 11:06:25.433000+00:00 | ['Looker', 'AWS', 'Data Analytics', 'Data Science', 'Data Visualization'] |
Why you should write something no one will read. | Why you should write something no one will read.
Join me in a writing adventure that will take you underground.
Have you ever lived through a moment when your writing went dormant? One morning, you wake up and realize you need a sweater. Next thing you know, you’re unpacking your parka and wondering where all the wildflowers went. Or if they’ll ever be back.
Photo by Nikola Jovanovic on Unsplash
The worst part? There’s no calendar where you can neatly cross off squares until you reach the first day of spring. I know because I once went 5 years without sharing my writing. The leaves dropped off like this:
No more hosting writer’s groups with clinking pitchers of iced white tea and the latest chapter of my novel. No more family writing circle with critiques traversing oceans with love, encouragement. No more blog about life in a new country, nourishing old friendships and forging new ones. No articles submitted for publication, no guest posts. No more reaching out to others through writing.
Today, each time I send a shoot up to the light and share my writing, it feels like a romp through the dandelions. Part of me just wants to forget all about that dormant period.
But when we struggle, we learn.
You see, all that time underground, I was writing — for an audience of one. Myself. And I’ve learned this:
As much as I need to send the hopeful, green shoots of my writing into the world, some of my writing has to take place underground.
So I’d like to invite you below, to this place I’ve discovered and still need to visit from time to time. (And let me be the first to reassure you, the wildflowers will be back.)
Writing only for myself changed me (and my writing) forever.
Sharing your writing matters. But writing in public is like shuffling barefoot over loose, black earth. It’s delicious and all-consuming — an experience not to be missed. But sometimes it’s worth diving in a little deeper. An entire world is wriggling, pulsing and expanding just under your toes.
The world opens up under your feet when you start writing alone. When you keep a tiny bit of your writing only for you — without publishing, sharing or showing.
I get it, the time you spend on your journal is time you can’t spend on the writing you need to share: your blog, your memoir, your poetry, your book. It’s time you’re not cultivating relationships with other writers, readers, adventurers of the written word.
But the time you spend writing only for yourself changes what you do share in ways that are hard to believe until you try.
And the relationship you cultivate with yourself? I learned that’s worth something too.
When you write alone, you surprise yourself.
You write anything that comes to mind, in any order. Sometimes you will babble. The craziest ideas will bubble up — ideas that would have been strangled if you’d been following an outline or wondering if your audience cares about your subject.
The writing you do by yourself is your place to write it all. Even if you think it’s weird. Especially if you think it’s weird.
Personally, I think that’s some of my best writing. The stuff that I discovered while rummaging around underground, that I once thought I’d never share.
You will write what (really) terrifies and embarrasses you.
You know, I see a lot of people sharing some amazing things online. Things I’ll admit I wouldn’t be brave enough to put out there. But all my reading plus my underground writing habit leads me to suspect something:
We all stop somewhere. There is no such thing as a ‘tell-all.’
Think of ‘reality’ TV — one of the most artificial experiences you can delve into. If an author (or TV personality) tells me I know everything about her or that her life is an ‘open book,’ it leaves me scratching my head. In the end, there’s always something held back, unexplored.
My underground writing habit has taught me that sometimes we’re blind to our own inner workings. But they are there.
My most beloved writers (fiction, non-fiction, famous and the still ‘under-the-radar’) are the clearest on who they are — who have honed the art of being themselves and accepting the occasional ugly, crusty truth. Is that really something you can do in the limelight?
Writing alone develops your honesty and teaches you about yourself. And it will make you a different kind of writer in public too: the kind of writer you want to be. Not the one who knows everyone is watching.
Writing alone gives me permission to write more than I actually share. I think you deserve the same. Which brings me to the next point…
Writing for yourself teaches you to filter (but not discard.)
If doing some of your writing in secret allows you to tap into an earthy world of honesty, it also helps you make wiser choices about what shoots you send up to the surface.
I’m not saying you have something to hide. Or that some of your stories must remain underground because they aren’t worthy.
Sometimes subjects that pull at us and intrigue us are simply not ready to see the light of day. (Yet) Or not right for our current audience. (For now.)
That’s the beauty of writing only for yourself. Any of the ideas that unearth in that quiet, dark place? You’re not banishing, burying them, or letting them rot. You can turn them over, explore them, let them develop into something you might share later.
Or they might remain below the surface and develop into something even more powerful than the shoots you send up and into the light…
When you write alone, you create an entire universe — and you invite your reader to create alongside you.
If you’re reading this, you probably love books. (Seriously, that’s likely to be how we met — book club? Voracious reading habit…I know you from somewhere).
You’ve probably already thought about the way the best books make every reader the accomplice, the co-author.
How do books work this magic?
With everything the person who penned the story leaves out.
The spaces in between. The meanings she never spelled out, but that exist in a dark, earthy space under the surface. Sinewy depths we can’t ignore. Ideas that grow, twist, curl, take on new shape and meaning underground — long after the text has reached its final form.
All of these underground spaces invite you, the reader, to complete the ideas with your own imagination. And the meanings and truths you create alongside the author? Well, they become the most powerful part of the story.
Where does that subtext come from? From the lessons learned, wisdom, moments captured. The unsaid, the unwritten transform your writing — lets your reader meet you halfway and fill in what you left out.
But that underground world can’t nourish your writing if you don’t plant it under the earth, spill the water and let it grow.
So grab your pen and notebook, sink your feet deep into the soil and create an unspoken universe that will change your writing forever.
And in that silence? You’re not alone. I’ll be there too. Because if there’s one thing I’ve learned it’s that writing alone is something I can’t do without.
Want to meet me halfway?
Help me create meaning from the muddy stuff of words and tell me if any part of your writing takes place underground. Or if you’re about to grab a pen and spill a little ink just for yourself. | https://trishatraughber.medium.com/why-you-should-write-something-no-one-will-read-758099beb0fc | ['Trisha Traughber'] | 2019-08-29 12:54:52.704000+00:00 | ['Writing', 'Creative', 'Journal', 'Life Lessons'] |
How to Make Your Website Perform to Its Fullest Potential | For years, Hibbert Sports Inc. suffered for not developing a great website.
Within two months of re-launching an appealing website, the company site’s ranking on Alexa jumped from around 200,000 to 78,000.
Web traffic to the site also increased by 45%.
Your website is the digital face of your company. The place where most people interact with your business.
Whether it’s to learn more about your products or services, or talk to a customer service rep, or make a purchase, your website is where most people go.
You may have a great marketing campaign, but if your website doesn’t convert because it’s not appealing to customers, you’ll lose potential sales.
Fifty-five percent of visitors spend less than 15 seconds on a web page.
If your goal is to acquire customers through your website, you need visitors to spend more than 15 seconds on your pages.
The more time people spend reading your web pages, the more they are likely to convert into customers.
It shows they are interested in your product or service.
I assume your business has a website. How can you increase the number of customers you get through it?
Here’s what you should do:
Do a/b testing on a regular basis.
A/B testing is comparing two versions of a web page to see which one performs better.
The version that receives the most conversions wins.
A/B testing is a powerful marketing technique that helps you improve every aspect of your website.
It helps you get more subscribers, leads, and sales.
How A/B testing works
Here’s what happens when you run an a/b test:
You have a 50/50 traffic split between the original page and another variation.
The original (current) page design is Variation A.
The other variation which shows to half of your visitors is Variation B.
Your goal is to see if Variation B gets more customers than Variation A (the current design).
If your website receives a lot of visits, like 5,000 daily visitors, you could decide to have more than two variations.
You can create and run four variations at a time.
My favorite is running two tests at a time.
Don’t run tests for a few days.
Make your tests run for 14 days. That means two full weeks starting from Monday to Sunday.
There should be no external event within the testing period that might affect the outcome of the test, e.g., Thanksgiving Day.
When you let the test run for over 14 days, the results may start to become meaningless.
Here are the top tools for doing a/b testing:
Optimizely
VWO (Visual Web Optimizer)
Adobe Target
So, what are the things you should a/b test on your site to make it perform to its fullest potential?
A/B test the headline
The headline is an essential element of a web page.
Even though the headline is not more than a sentence, 50% of the time you invest in creating a web page should go to the headline.
Why?
Eighty-percent of people who visit your site will never make it past the headline.
High-trafficked sites like BuzzFeed and Upworthy are successful due to their abilities to create captivating headlines on a regular basis.
Peter Koechley, the co-founder of Upworthy talks to WIRED on the importance of headline.
“The headline is our one chance to reach people who have a million other things that they’re thinking about, and who didn’t wake up in the morning wanting to care about feminism or climate change, or the policy details of the election. The difference between a good headline and a bad headline can be just massive. It’s not a rounding error. When we test headlines, we see 20% difference, 50% difference, 500% difference. A really excellent headline can make something go viral.” — Peter Koechley, Upworthy’s co-founder
Test many headlines to find the one that gets visitors to stay longer on your web page. For example, the website of Nicholas A. Parr, Maryland personal injury lawyer, displays the name of the firm’s principal on the headline. The name could be tested with other headline ideas.
I love Lyft’s driver recruiting landing page. Their headline is captivating. Anyone who likes driving will surely want to “earn more.”
There’s no doubt that the marketing team at Lyft conducted a series of a/b tests before settling on this headline.
Digiboost, a digital marketing agency uses a simple headline on its site: “Web Design for Small Business.” The agency helps small businesses and it made that clear in its headline.
A/B test the subheadings
When you scroll down the Lyft’s landing page example given above, you’ll see these subheadings:
These subheadings keep visitors’ interest alive.
An interested driver will want to know how much they’ll earn.
The next subheading answers a question drivers may have:
“What is Lyft?”
These subheadings keep visitors on the Lyft’s website, increasing their chance of converting them into drivers.
Jumpstone International, a business coaching firm has a great subheading on its website. Here’s the subheading:
“Find clarity, gain confidence and excel in your market!”
Potential clients will want to have these three key things.
A/B test the first sentence
The first sentence also called the description or lede in news writing is an important element of your website.
It should grab a reader’s attention immediately. It should be memorable.
For example, on the homepage of Ahrefs, their first paragraph makes a bold promise to visitors.
“Ahrefs helps you to learn why your competitors are ranking so high and what you need to do to outrank them.”
A/B test your call-to-action buttons
Every website has a goal it wants visitors to complete.
What’s the action you want visitors to take on your website?
It could be that you want visitors to book a demo of your product.
Or, you want visitors to fill out a form.
Or, you want visitors to start a free trial of your service.
MedCognition wants visitors to test-run its realistic patient simulation software.
For example, when you visit Groove’s site, they want you to do is start using their help desk software for free.
They want you to drop your email address and click their call-to-action button which says, “Try Groove For Free.”
You need a call-to-action button that makes visitors act.
The goal of Evernote is to get visitors to sign up. Their call to action is focused on that.
Some websites have two call-to-action buttons.
The reason for that could be that there are two important actions visitors may want to take.
For example, Skype’s website has two call-to-action buttons.
The first button wants you to “start a conversion,” using Skype’s web version.
The second button wants you to “Download Skype” on your computer and start using Skype from there.
Some websites use three call-to-action buttons.
For example, Zendesk gives you three options.
The first button asks you to “sign up for a demo.”
The second button asks you to “start a trial.”
The third button appears on the bottom-right side of the screen and less prioritized, and asks you to “contact sales.”
No doubt all these companies have conducted a/b tests for the positions and number of their call-to-action buttons.
Another important thing you should a/b test in your call-to-action button is its text or phrase.
For example, Netflix call-to-action button text is simple:
“Join Free For A Month.”
A/B test different words and phrases in your call-to-action button.
EmpireFlippers.com increased clicks on their CTA button by 33.10% just by changing its text.
You should also a/b test the color of the CTA button.
OKCupid’s call-to-action button stands out.
Karen Brody offers man coaching service, where he helps men awaken their potentials both as leaders and lovers. Her call-to-action copy is, “learn how.” A prospect would want to click after reading that. But can a different call-to-action copy generate more clicks? We can never know until we test.
A/B test images on your site
Images boost conversions.
For example, the images on the Circle Surrogacy home page would make visitors stay longer.
Some sites don’t use images on their landing pages.
If an image isn’t improving conversion rates on your website, then it has no reason to be there.
After you’ve proved that an image increases conversions on your website, you should test different types of images too.
For example, Harrington Movers, a New Jersey and New York City moving company, tested a generic photo versus an image of their employees.
They also tested an image of the company’s truck.
Conversions improved for both the truck and employees’ images by 45% versus the generic stock photo.
Test different kinds of images.
Test different image sizes too.
For example, test a large image versus a small size.
Images sell online. But some images sell better. It doesn’t matter the size of your business, you should a/b test the images on your site. For example, Cute Smile 4 Kids is a dental clinic in Phoenix, Arizona. They have both stock photos and brand pictures on their site. They could a/b test both to see which performs best.
A/B test videos on your site
Sixty-one percent of businesses use video in their marketing according to Wyzowl.
Various marketing studies have shown that videos boost conversions and sales.
Shoppers who view video are 1.81 times more likely to purchase than non-viewers according to Adobe.
Now, you may want to consider adding a video to important pages on your website.
You should still a/b test video versus non-video on a web page.
For example, MedCognition, a healthcare startup that creates realistic patient simulation software, has a demo video on its homepage.
Elville and Associates, an estate planning and elder law firm in Maryland has videos that feature its principal on the home page. Videos like these turn prospects into clients.
Videos help visitors learn about your product or service quickly. It takes less time to watch a video than to read an article.
A/B test every aspect of your website and watch your revenue soar.
Hit the clap icon if you enjoy reading this article.
Thanks in advance. | https://medium.com/the-ascent/how-to-make-your-website-perform-to-its-fullest-potential-d0bc726529f6 | ['Michael B. Akinlabi'] | 2019-12-02 16:53:59.772000+00:00 | ['Marketing', 'The Ascent', 'Digital Marketing', 'Conversion Optimization', 'Advice'] |
I Thought I Was Done Exploring Uncharted Territory. I Was Wrong. | Photo by Mark Potterton on Unsplash
We stood close together, dressed in our heavy coats and pants atop Haleakalā, waiting for the sunrise. In the frigid air, the snot ran from my nose — a normal reaction, I was told. I incessantly wiped, using the tissues I’d shoved in my pocket, but my tour driver playfully laughed with me, asking, “It just keeps goin’, eh?” When I sheepishly looked up at him, he said, “No worries. It’s a temporary cold you have. You’ll be over it at the bottom.”
He’d picked me up from my hotel in the middle of the night, his last pickup. When I got on, I saw everyone else was paired up. Some were honeymooners and others were members of family groups. I was a party-of-one who’d traveled to Maui for a week-long break from my internship on O‘ahu during the summer after my freshman year of college. At first, I was concerned I was alone because I was far from athletic, I hadn’t been on a bike in years, and I’d considered myself acrophobic. Who'd be there for me if I got hurt or froze up, completely unable to continue? But, I pushed the idea away, convincing myself that I’d regret giving up the adventure if I quit before trying; that there is no time like the present.
So, I chose my seat, and we started the drive, along which, before reaching the apex of the 10,000-foot volcano, we layered up with lots of gear to protect us from the rapid change in temperature and air pressure. Midway through the drive, my ambivalence returned. I concocted a wild story in my head about my own demise and ruminated over how they’d handle my dead body once they found it hanging on the cliff somewhere. At one point I considered staying on the bus so I wouldn’t embarrass — or kill — myself.
Our driver told us that we’d have plenty of time to marvel at the majestic oranges, pinks, yellows, and blues before the bike ride. He also mentioned that the bike ride is considered no more dangerous than any other bike ride, unless the rider is driving recklessly. But, for someone like me, who had little athletic ability, even with recently climbing Diamond Head in Honolulu, this was a risky feat.
The bike leader noticed my apprehension, and said, “Talk to me.” I told him how afraid I was of heights (other than being in a plane) and that I was scared I’d be the victim of some freak accident and I’d be left for dead. He assured me that once I got moving, all would be fine, and he promised he would not let anything bad happen to me.
Still apprehensive, I decided that getting on the bike. A colleague had coordinated my entire Maui itinerary, including this bike tour, and I didn’t want to disappoint her or waste her hard work. But,more importantly, I needed to stand in my autonomy, separate from my family and friends’ expectations of me, and do something for myself.
With a little push of my right foot and then quickly pulling it up to the pedal, I sped down the initial leg of the trail. Unstable initially, my body bounced along, the vibrations, competing with my pounding heart and fearful panting, ricocheted from my toes up through my cheeks, which had already turned flush. The decline, though insignificant, felt immense to me, and my sweaty palms frantically tried to gain control before I fell forward and scraped my face on the asphalt. I concentrated on looking forward, allowing the momentum to pull me so I wouldn’t have to exert energy on pedaling.
Within the first several miles, the leader told us to brake and pull over for our first stop. Relieved, I did as instructed, and hopped off my bike. While the breathtaking view invigorated me — the beautiful shades of greens below and the crystal blue sky above — I also remembered my location, high above flat land below, and I felt frightened. Yet, I encouraged myself with logic: You’ve gone this far already; you’re already a third there.
Some of the honeymooners took pictures, while the rest of us gazed into the scenery, and then we mounted our bikes again, continuing our descent.
I found it hard to enjoy the biking part; I was too engrossed in not flipping over. However, with each stop, I stepped closer to the edge of the new platform and landing, forcing myself to, at the very least, admire the wonder from my new vantage point. By the time I reached the bottom, I felt accomplished and grateful I’d seen the ride to the end without allowing my fear to overtake me.
That’s the funny thing about youth — we do things and take risks that we wouldn’t necessarily do later in life, putting ourselves out there for a thrill or even simple curiosity. Over the next few years I traveled to a few new places, testing myself with new challenges, like traveling to Mexico with friends where I rode a stubborn and slow donkey down a rocky mountainside, or taking a semester abroad. No matter where I went or what new experiences I had, I always recalled the girl on the bike — the one who told fear to “get lost” and did the thing anyway. | https://medium.com/atta-girl/i-thought-i-was-done-exploring-uncharted-territory-i-was-wrong-cb3a7522caea | ['Christine M. Estel'] | 2020-11-30 00:57:53.591000+00:00 | ['Writing', 'Adulthood', 'Travel'] |
Object counting with Kerberos Cloud | Welcome to Kerberos Cloud
The Kerberos Cloud is our SAAS to manage and monitor video surveillance cameras in the cloud (or in your private data center, actually where-ever you want).
The idea is that you connect one or more cameras to our platform, through the concept of Kerberos Agents. Once you connected your Kerberos Agents to the Kerberos Cloud solution, the platform will do all the heavy lifting for you. A limited number of features included are:
Monitoring of devices (when they go offline),
Sequencing and grouping of recordings,
Advanced search capabilities,
Live streaming,
Notifications,
Machine learning,
and, many others.
One of the new features is now object counting.
1. Object counting
Once you have a Kerberos Cloud account, and a gold subscription, you will have access to the object counting functionality. If you have one or more devices connected, you can navigate to the devices tab, and open a specific device of choice.
Once you opened the device detail page, you will see several tabs at the top of the page: Details and Counting. Select the Counting tab, to configure the object counting for the selected device.
The concept: A counting zone
The idea of our object counting function is that you can create multiple counting zones for each device. A counting zone is a set of counting segments, which are connected.
A counting zone is interesting to track the path of a specific object. For example when you have a promotion in your retail store, you could verify the success of the promotion by verifying the walking directions of your customers. A counting zone could also be as simple as counting incoming and outgoing objects.
Two counting zones, who can count incoming and outgoing objects.
The concept: Types of segments
As described previously a counting zone consist of multiple segments. A segment can be of following three types: incoming, intermediate and outgoing.
Three different segment types: incoming (green), intermediate (yellow) and outgoing (red).
The idea of having different segment types is to make different use cases possible.
Entrance Counting: By adding an incoming and outgoing segment, the platform will count the objects when crossing both lines. The sequence of crossing will determine if it’s an incoming or outgoing object. When an object first crosses the outgoing (red) segment and then the incoming (green) segment, it’s counted as an incoming object. When first crossing the incoming (green) segment and then the outgoing (red) segment, it’s counted as an outgoing object. Traject Counting: Instead of simply counting incoming and outgoing objects, you can also measure the traject of a specific object. By adding one or more intermediate segments in between, the platform will calculate if objects are crossing these intermediate segments as well. Multi entrance counting: It’s also possible to have have multiple incoming and outgoing segments in a counting zone.
2. Dashboard and Measuring
Once you’ve setup the preferred counting zones for all your devices. You can start evaluating and monitoring the object counting on a high-level view. When navigating to the counting page, in the left navigation menu, you will land on the dashboard page. On this page insights are given regarding your object counting settings made in the previous step. | https://medium.com/swlh/object-counting-with-kerberos-cloud-a4e7486abf00 | ['Cédric Verstraeten'] | 2020-07-03 14:36:51.905000+00:00 | ['Machine Learning', 'TensorFlow', 'Computer Vision', 'Video Surveillance', 'Counting'] |
You Don’t Have to Know Everything; You Just Have to Know Where to Find It | You Don’t Have to Know Everything; You Just Have to Know Where to Find It
“Just-in-time learning” can make your life a whole lot easier. Here’s how to make use of it.
Photo by Andrea Piacquadio from Pexels
In 1921, an interviewer asked Albert Einstein about the speed of sound. Einstein replied that he didn’t know because he didn’t carry such information in his mind. Instead, he’d look it up in textbooks where the answer is available anytime.
If Einstein lived today, he’d probably memorize even less because the internet enables us to find answers to most questions within seconds.
Yet, education systems all over the globe still force students to learn information by heart regardless of the relevance.
That’s not only ineffective but also worthless as knowledge itself isn’t worth much. The saying knowledge is power might have been true a few decades ago. Today, only applied knowledge is power.
In the 21st century, you can learn almost any skill for free if you know how to browse the web and take enough time.
That’s a blessing and a curse at the same time because most people don’t know how to find relevant information. Yet, that’s one of the most powerful skills of the 21st century.
If you know how and where to look for information the moment you need it, you can master almost any problem with ease.
Instead of wasting your brainpower on memorizing information, you should try so-called just-in-time learning.
“Just-In-Time means learning something only when/because you actually need to use it.” — Kathy Sierra
Learning on the job is also called embedded learning and becomes increasingly relevant. More and more organizations switch from traditional education systems to just-in-time learning because of several reasons.
In the past, companies invested in random educational programs to equip their employees with additional skills. Today, most businesses don’t want to waste time and money on programs teaching random skills that will never be used anyway.
Typically, if you learn a skill without applying it, you’ll forget what you learned pretty quickly. When you need the skill after a few months or even years, you’ll need to relearn everything and realize that the first time was a waste of time.
And as time is our most valuable resource, we shouldn’t waste it on learning skills and memorizing information if we don’t need it.
Instead, leading organizations encourage their members to apply just-in-time learning. That way, employees can educate themselves in real-time when they actually need a skill. | https://medium.com/the-ascent/you-dont-have-to-know-everything-you-just-have-to-know-where-to-find-it-af3545b11832 | ['Sinem Günel'] | 2020-11-24 16:03:30.184000+00:00 | ['Books', 'Knowledge', 'Education', 'Innovation', 'Learning'] |
How to Code and Trade the Commodity Channel Index | Let us create the indicator step by step and then present everything in a full function. Here is the plan of attack:
Calculate the typical price which is simply the average between the current low, high, and closing prices.
Calculate the Mean Absolute Deviation of the price.
Calculate the 20-period moving average of the typical price.
Choose a constant. Generally, Donald Lambert recommends 0.015 for stability.
Suppose we have an OHLC array containing four columns; open, high, low, and close. We want to build the indicator in a simple way. Let us start by calculating the typical price. We should write a block of code that tells the algorithm to loop through the array and for each row, calculate the average between the second (high), third (low), and fourth (close) columns.
# Calculating Typical Price
Data[:, where] = (Data[:, 1] + Data[:, 2] + Data[:, 3]) / 3 # The where variable should be 4. It is the location of the Typical price. Therefore, after four columns of OHLC values, we place the typical price in the next column which is the fifth one and indexed at 4.
The next step is to calculate the Mean Absolute Deviation. But what is that, really? The Mean Absolute Deviation — MAD resembles the Standard Deviation only it takes the absolute distance from the values relative to their mean.
The Standard Deviation takes the square distances and then the square root of the results before averaging them.
takes the square distances and then the square root of the results before averaging them. The Mean Absolute Deviation takes the absolute distances and then averages them.
So, we are clear, the Mean Absolute Deviation is a measure of statistical dispersion from the mean that shares some similarities with the Standard Deviation. In the library pandas, we can find a simple built-in function to calculate this so that we do not complicate the problem with numpy. Only one problem, we are dealing with numpy arrays and not pandas dataframes. This is easily solved by converting. Here is how:
# Taking the Typical Price column to transform it
specimen = Data[:, where] # Transforming the Typical Price column into a pandas series
MAD_Data = pd.Series(specimen) # Looping through the Typical Price series with a lookback of 20 periods to find the Mean Absolute Deviation using the mad() built-in pandas function
for i in range(len(Data)):
Data[i, where + 1] = MAD_Data[i - lookback:i].mad()
The next step is simple. We have to calculate a 20-period simple moving average of the typical price:
def ma(Data, lookback, what, where):
for i in range(len(Data)):
try:
Data[i, where] = (Data[i - lookback + 1:i + 1, what].mean())
except IndexError:
pass
return Data # Calculating Mean of Typical Price
Data = ma(Data, lookback, where, where + 2)
Image by Author.
Now, if we combine what we have calculated into the above formula, we can have a full function like the below: | https://medium.com/python-in-plain-english/how-to-code-and-trade-the-commodity-channel-index-c4350fd9a58e | ['Sofien Kaabar'] | 2020-12-27 21:32:32.031000+00:00 | ['Trading', 'Finance', 'Machine Learning', 'Data Science', 'Artificial Intelligence'] |
Examining the Jealousy of Success through Psychological Concepts | Examining the Jealousy of Success through Psychological Concepts
A high-level overview of how jealousy is shaped through social comparison theory, cognitive bias, and mindfulness.
Photo by Fab Lentz on Unsplash
Jealousy is an emotion that is avoided in conversation and disregarded when shared publicly. Society has painted it to be ugly and abnormal. If that were 100% true, why does everyone experience jealousy to a certain degree? The answer lies in the psychology.
Social Comparison
Photo by Raquel Martínez on Unsplash
We compare things to other things. People to other people. Our current selves to our past selves. Instead of ignoring or shaming human nature, though, let’s navigate the phenomenon of social comparison when jealous thoughts come to the table.
When we feel jealous of a friend’s success, we are following Festinger’s Social Comparison Theory. This is when “individuals determine their own social and personal worth based on how they stack up against others”. This can become a positive (motivation) or negative (depression, jealousy) event.
Kristi Hedges from Forbes published this piece outlining a couple reasons why we get jealous of a friend’s success. As a psych person, I loved this statement:
We try to rationalize envy away, but envy itself isn’t rational. It’s an emotional response.
Top B-schools and corporate execs preach to avoid emotionally driven decisions. This in and of itself is highly irrational. Every decision we make will be influenced by at least one emotion, so let’s make this clear: avoidance builds problems, not solutions.
Decision-making skills improve when you become more aware of your emotions and how they make your body feel. Why? You’re applying rationality to an irrational concept.
Cognitive Bias
Photo by Paweł Czerwiński on Unsplash
“A thinking error that occurs when people are interpreting information around them, which affects the decisions and judgments that they make.” — Kendra Cherry, Verywell Mind
I worked in a research CAT (Cognition, Anxiety, and Treatment) lab that focused on understanding the cognitive biases underlying anxiety disorders, distress tolerance, and obsessive compulsive disorders. In short, we worked on shifting situations from being anxiety-ridden to neutral by reshaping thought processes. Think of it like a factory reset for your iPhone, but for your brain.
In a jealousy context, we exhibit a form of cognitive bias when we assume someone got promoted because they were lucky or knew someone higher up in the chain. But if we were the one to get the promotion, it was due to hard work.
Additionally, the Interactive Design Foundation noted that psychologists Amos Tversky and Daniel Kahneman found that presenting the same information differently can lead to opposing decisions being made. This means that our biases and habits can be unlearned or, in the very least, modified.
Just like our cognitive biases, we can reshape our jealous thoughts over time with some awareness and effort.
If interested in learning more about cognitive bias modification (CBM) and the intersection of anxiety and jealousy, check these out:
Mindfulness
Photo by Aleks Marinkovic on Unsplash
Mindfulness is when we become fully aware of what we are feeling and doing in the present moment. I believe it’s important to feel negative emotions (ex: jealousy, sadness) when they happen instead of pushing them aside. Ignoring our emotions completely can make us feel trapped or encourage them to build up over time, leading to an emotional outburst.
Practicing mindfulness can help us become aware of our jealous thoughts when they happen and the type of situations they occur in. Then, over time, improve our understanding of ourselves and challenging such thoughts.
For example, I practice mindfulness primarily through guided breathwork, yoga, and meditation. I also carve out 5–10 minutes for journaling every day.
And for real life examples, check this out:
Created by CBT expert and clinical psych professor Dr. Robert L. Leahy
Conclusion
It’s normal to be jealous when others succeed. It’s also normal to want to change certain thoughts. My advice to you:
Allow yourself to experience life and the emotions that arise. Question the rationality behind the biases you are holding on to. Pivot your thinking towards the path that betters you as a person.
Note from the Author
Thanks for reading! Please feel free to connect with me on Medium and LinkedIn. I’d love to hear your thoughts and continue the conversation.
-Mo | https://medium.com/the-innovation/examining-the-jealousy-of-success-through-psychological-concepts-97ed417c436b | ['Mo Kaiser'] | 2020-11-12 17:20:24.457000+00:00 | ['Mindfulness', 'Jealousy', 'Personal Development', 'Psychology', 'Success'] |
Context-Centred Design | Designer of digital products, brands, websites and type. Co-founder at akord.com, a tech startup based in Paris. Visit the source at www.blogofpascal.com
Follow | https://medium.com/swlh/context-centred-design-ce095aa39e8b | ['Pascal Barry'] | 2020-11-04 11:45:26.033000+00:00 | ['System', 'Design', 'Thinking', 'Theory', 'Product'] |
Waiting in Limbo and Chaos as a Creative Writing Process | Writing is a similar process. I get to a point where I feel it is heavy or boring, an all over the place mess of a thing, before I can start making it into something that hopefully speaks as a message that I only become conscientious of as part of the process of writing. I might think I know what I am going to write about when I start, or maybe not, but the process follows a formational path of its own, not intended by my direction until the end, when it is. The process of completing a painting is the same as perfecting the details of writing as one pays attention to the flow of the content and the choice of words until it feels right.
Now I am secure that I will not let go of my process in exchange in incorporating other intentions. Stage two of my Medium journey is to follow the advice that I intentionally ignored in stage one.
This post is an experiment in “publishing every day” though I do not know if I will actually take it so far as “every day”. This story is intended to be something written and published in a day. I can have that intention because I am in between writing a story using the chaos process. Once in the chaos process, I do not know that I can switch on the quick post process, but that doesn’t matter today.
My second new goal is to submit to publications with a larger audience on a regular basis. I have built up a collection of published works allowing me to feel the latitude to invest in the waiting period it takes to submit to some larger publications. Previously I needed the closure that publishing brings, but since I have a library of published articles already, I do not need that closure to come at the same rate as previously. Everything is always about process and how each activity or decision interacts with the larger process of writing so that it keeps on flowing at a steady pace.
Photo by jacob-mejicanos-unsplash
One publication with a large readership accepted and published one of my first stories. I did not see a wider readership for that story than my other stories but it targets a niche audience and I had no followers at that time. All stories target a niche audience but as one reads other stories published on Medium, one gets a sense of being part of a global conversation. I also need to make some money to advance my own life goals, and so it is time to embrace popular advice but without letting go of what is valued most deeply.
Since the large publication that published my early story took a long time to respond I had stopped submitting to it, not having the patience required by that process. I didn’t want to let waiting around create a limbo in my work flow, but one day I read an article by its new editor and decided to submit again. This time I got rejected within the space of twenty four hours. My story wasn’t right for their publication, which is the publication’s version of asking “where does a painting fit in the market ?”
I don’t mind a rejection based on not fitting in when it comes within a day of submission. My creative cycle can accommodate a day, and I appreciate that the rejections sound like they are from humans who actually read my story. The most recent one said “ It is an interesting story but..it doesn’t fit”. I place more importance on the first part of the statement than the latter. I know I don’t fit….but the world is evolving, not always the same. I am the world. It is me. We are all One. There is a time and a place for everything. Patience is a virtue, as I always tell my cats but they don’t listen to me. | https://medium.com/an-idea/waiting-in-limbo-and-chaos-as-a-creative-writing-process-ddca71e7a8da | ['Mackenzie Andersen'] | 2020-11-06 21:58:12.290000+00:00 | ['Process', 'Writing', 'Meaning', 'Philosophy', 'Writing Life'] |
The Fundamentals of Object-Oriented Programming for Java Beginners | Object-Oriented Programming (OOP) is a method of coding that uses succinct, modular “objects” to organize data and functions. It is a concept that can be used to build complex and expansive programs and spans multiple coding languages. I first learned of it while learning to code in Ruby, for example.
Recently, I began looking into learning Java — a language with a long history known for its use in OOP. Other newcomers to Java might be surprised to find several references to four “fundamentals,” “principles,” or “pillars” (the language used varies, but they’re all referring to the same concepts) of Object-Oriented Programming inside Java’s documentation. While not often taught outside the context of Java, these four fundamentals are applicable across languages and can help give anyone a greater understanding of OOP.
1. Encapsulation
In Object-Oriented Programming, we use classes to define the types of objects used in our application. For instance, an online marketplace would likely have a “Product” class, with class-specific variables such as “name” and “price.” It might also have class-specific methods, for instance a function that adds an instance of a Product to the user’s cart. Here’s what that class might look like in Java:
public class Product{
private String name;
private double price; public void addToCart() {
//method is defined here
System.out.println("Added " + this.name + " to cart!");
}
}
Notice the keywords “public” and “private.” Public variables and methods can be accessed from anywhere in the program. Private variables and methods can only be accessed by the object they belong to. Encapsulation is the ability to designate variables and methods as private. So, for instance, if we were to create a new Product object called newProduct and then try to access its name…
public class Main {
public static void main(String[] args) {
Product newProduct = new Product();
System.out.println(newProduct.name);
}
}
This would return an error, because price is a private variable and therefore can’t be accessed by Main. On the other hand, if we were to do this…
public class Main {
public static void main(String[] args) {
Product newProduct = new Product();
newProduct.addToCart();
}
}
No error would appear, and running the program would result in newProduct’s addToCart() method being employed. This is because addToCart() is not a private method.
2. Inheritance
In our previous example, our theoretical retail application defined a Product class for the products being sold. But what if we wanted to sell different kinds of products? A clothing product would still be a Product with a name, price, and the ability to be added to a user’s cart, but it may also have other attributes —a size, for instance. Likewise, a food product likely has some sort of sell-by date. To further customize objects of our Product class, we can create subclasses that inherit its properties while inserting some of their own.
public class Product{
private String name;
private double price; public void addToCart() {
//method is defined here
System.out.println("Added " + this.name + " to cart!");
}
} public class Clothing extends Product {
private String size;
} public class Food extends Product {
private String sellBy;
}
In the above example, “extends” is Java syntax that defines Clothing and Food as subclasses of the Product class. This means that any Clothing or Food option will have access to the name and price attributes as well as the addToCart() method, in addition to the variables defined within the classes themselves. Inheritance is the ability of a class to inherit properties from a different class.
3. Abstraction
Let’s continue using the code above. Imagine the online store being created is a very specific type of marketplace, and it only sells clothing and food. In this case, we never want to see an object of the Product class being created — we only want objects of the Clothing and Food classes. To ensure that this is the case, we can make the Product class abstract.
abstract class Product{
private String name;
private double price; public void addToCart() {
//method is defined here
System.out.println("Added " + this.name + " to cart!");
}
} public class Clothing extends Product {
private String size;
} public class Food extends Product {
private String sellBy;
}
Notice that the keyword “public” in the first line of code has been replaced with “abstract” — this ensures that Product objects cannot be created. Instead, the Product class may only be inherited from by other classes. Attempting to create an object of the Product class will now result in an error.
public class Main {
public static void main(String[] args) {
Product newProduct = new Product();
//This code won't run!
}
}
Abstraction, like encapsulation, controls access to certain aspects of code. Both classes and methods can be made abstract.
4. Polymorphism
Up until now, the addToCart() method has been identical for both Clothing and Food objects, but this doesn’t have to be the case. Polymorphism means that the same method can have a different implementation for each subclass.
abstract class Product{
private String name;
private double price; public void addToCart() {
//method is defined here
System.out.println("Added " + this.name + " to cart!");
}
} public class Clothing extends Product {
private String size; public void addToCart() {
System.out.println("Added " + this.name + ", size " + this.size + ", to cart!");
} } public class Food extends Product {
private String sellBy;
}
Now, performing the addToCart() method on a Clothing object will result in both the name and size of the added object being printed. This differs from the Product and Food versions of the same method, but retains the same name for ease of use. To change the implementation of any subclass’s inherited method, all we have to do is re-define the method inside of that subclass. | https://reedanna.medium.com/the-fundamentals-of-object-oriented-programming-for-java-beginners-d9c8b4a07ad | ['Anna Reed'] | 2020-12-26 22:25:44.813000+00:00 | ['Software Development', 'Coding', 'Object Oriented', 'Java', 'Programming'] |
Getting Started with Anaconda for Machine Learning | Installation
There are two ways you can install conda . The first way is to download and install the full Anaconda package. It comes with Python and a set of libraries that will eventually take up several gigabytes in the disk. The other way is to install a relatively lightweight Miniconda which only contains the essential data science tools. Whatever you will install, Conda CLI tool is shipped with both ways so you don’t need to worry about what should be installed. If you want to save disk space, install Miniconda otherwise, install Anaconda.
After downloading Anaconda/Miniconda, you can start using it through the Anaconda Prompt. Note that the newer Anaconda versions don’t automatically add itself to the PATH. Therefore if you want to use conda command in the Windows terminal, you need to add the following to the path. Replace <anaconda_directory> with the directory name where Conda is installed on your computer. | https://towardsdatascience.com/getting-started-with-anaconda-for-machine-learning-e8eb7e12199d | ['Pavindu Lakshan'] | 2020-06-22 02:57:23.909000+00:00 | ['Programming', 'Machine Learning', 'Data Science', 'Python', 'Anaconda'] |
Randomly wired neural networks and state-of-the-art accuracy? Yes it works. | In an effort to discover even more powerful networks, Google AI’s research team developed a technique for having an algorithm search for good networks. This is known as Neural Architecture Search (NAS).
Neural Architecture Search
NAS is an algorithm that searches for the best neural network architecture. Most of the NAS algorithms work quite similarly.
Start off by defining a set of “building blocks” that can possibly be used for our network. Then try training a network by putting together different combinations of those blocks. Through this trial and error, the NAS algorithm will eventually figure out which blocks and which configurations work best.
As it turns out, this method works quite well and achieves state-of-the-art accuracy. It discovers some pretty peculiar looking architectures too, as you can see from the official paper below:
NASNet blocks discovered by the NAS algorithm. Source
The great thing about NAS is that we get to see some neural net architectures that we might otherwise not have looked for at all.
It does however come with its own limitations. Since we’re sampling from a fixed set of blocks, we won’t ever discover any new blocks. It’s the same case with the skip connections: NASNet only allows for a skip connection across 1 downsampling level, but what if other types of connections might work?
Facebook’s AI Research team took a new stab at NAS by trying out randomly wired neural networks. The intuition was that if the general idea of architecture search worked well for the fixed blocks and connections, perhaps running it with a much wider search space (i.e random connections) will lead to some unexplored configurations.
Randomly Wired Neural Networks
Since the authors of the paper wanted to focus specifically on the wiring i.e the connections of the neural networks, they make a few restrictions on the network structure:
The input is always an image of size 224x224
The blocks are always a ReLU-Conv-BatchNorm triplet. The convolution in the triplet is always a 3x3 separable convolution from Xception.
The aggregation of multiple tensors (for example when fusing a skip connection with other tensors) is always done in the same way — via weighted sum. These weights are learnable and always positive.
The stages of the network always remain the same. The general CNN structure of a multiple convolution blocks, followed by downsampling, repeated over and over until a final softmax classifier, has become standard in state-of-the-art network designs. This paradigm is also taken in these research experiments.
General structure of state-of-the-art CNNs, also adapted for this research. Source
An input size of 224x224 is quite standard in ImageNet competitions — the ImageNet dataset was used for benchmarking against other handcrafted networks as well as NAS-generated networks. The triplet block of ReLU-Conv-BatchNorm is also common and has been extensively proven to deliver state-of-the-art results when repeated throughout deep CNNs.
The way the tensors are aggregated is a small twist. Most state-of-the-art networks do a direct addition or concatenation without weights — but it’s a minor addition that shouldn’t affect performance too drastically. The staged design of CNNs shown in the table above has also become a go-to network structure used in ResNets, DenseNets, and NASNets.
Notice how this isn’t an entirely a “randomised neural network”. There’s no randomisation from complete scratch. Instead, a single component of the CNN design, namely the wiring, is explored while the other components remain fixed.
This is a significant point that the authors really tried to send home with the reader. We’re not yet doing completely randomised neural networks. But we are beginning to do some deep exploration of the component search space — one step at a time.
With these constraints set in place various classical random graph models were used to generate the random wiring of the networks.
And by random, I mean random.
Some of the randomly wired networks. Source
Randomised Networks open the gates of Deep Learning Exploration
The significance of this research lies in its main exploratory idea: expand the search space of NAS algorithms to find new, better network designs. While researchers have discovered some great designs on their own, it’s not practically feasible to go through the entire search space manually.
This also extends to another point: if we are expanding our search space, we’ll need an algorithm, in this case our network generator, to be really good at searching. It has to know what to look for or at the very least have some tendency to move towards it by design, much like optimisation algorithms like gradient descent work.
Architecture Search is the next frontier for Deep Learning research. It allows us to use an algorithm, rather than trial and error, to discover optimal architectures.
For now, it makes sense to fix some network components while searching for others (in this case wiring). This simplifies the problem to something bite-sized, easier to handle. NAS algorithms should have some degree of randomisation, since that’s the only way we can discover truly novel architectures.
The next step will of course be further expanding the search space and random nature of our search algorithms. That means expanding to more and more components of the network until the whole thing is designed algorithmically.
What happens when we let NAS go loose? Does it select non-square convolutions like 2x4? Does it propose to use feedback loops? Does the network become more simple or more complicated?
Neural Architecture Search presents an exciting new field of research. The hope is that letting a search algorithm loose becomes a way of using randomization to our advantage, discovering creative and never before thought of architectures. | https://towardsdatascience.com/randomly-wired-neural-networks-and-state-of-the-art-accuracy-yes-it-works-9fb3cedc8059 | ['George Seif'] | 2019-10-16 11:07:52.273000+00:00 | ['Machine Learning', 'Education', 'Deep Learning', 'Artificial Intelligence', 'Technology'] |
I Converted a Closet Into an Office | I Converted a Closet Into an Office
And I found freedom in a tiny space
Photo by Malin on Unsplash
I have a big family. I’m married and my husband works from home. I also have six kids at home, ranging in ages from 4 to 12. Daily I spend time homeschooling, working on my own classes as I finish up my degree, working on finances for our business, writing, or doing goodness knows whatever else needs to be done.
I have a desk in the living room in the basement with all of the kids’ desks and school stuff, which is a great place to sit during homeschool hours. During every other time, kids are running around, building pillow forts, screaming, making videos, making up games, or turning off lights to do a neon-light dance party. It’s not the best space for getting things done.
When the pandemic hit, suddenly things became real. We were all home all the time; our regular baby sitter couldn’t come anymore, and I was left to try to get more done in less time, with more people around. My solution was to sit at the kitchen island, on a comfy stool with my laptop. Well, it turns out the kitchen island isn’t much better than the basement, as far as noise levels go. Barely two minutes went by where someone didn’t come in and ask me a question or strike up a conversation, just because I was sitting there.
Besides that, there was the issue of my stuff. Files, books, planner, computer, power cord — it’s asking a lot of the island to hold all that stuff PLUS be, you know, the place where we make meals three times a day for eight people.
My mental health was suffering. As an introvert, I rely on quiet to get stuff done. The constant interruptions were hard at first; after two months, they became unbearable. I needed a place where I could shut the world out, just for a few hours a day, to work in peace.
But there was nothing to be done. All the rooms in our house have a purpose already, other than the storage room which is bare concrete. I was nearly desperate enough to give that a try until I realized that I had a perfectly good space that was the right size for what I needed — just enough room for a little desk and a bookshelf.
It was the closet!
The Space
It’s about 4'x8', certainly not very large but bigger than a lot of closets. It’s right off the garage, which means it’s to the side of the house rather than smack in the middle like the kitchen. It’s cedar-lined; our house was built in the 60s and I guess this was common then. I imagine it was filled with warm winter coats or clothes for special occasions sixty years ago; we used it as a hodge-podge holding place for seasonal decorating items.
There aren’t any outlets in it, but there’s an access panel to the laundry room. I have no idea what the original purpose of this panel was, but by going through the panel, I can plug a power strip into the laundry room and feed it into the closet. I found a floor lamp for some extra light and used a sewing table as my desk.
Photo by Author
I ended up buying myself a cushy office chair with moveable arms for a little bit more freedom of movement; otherwise, this conversion cost me nothing besides a bit of time clearing out the closet and moving some things around.
I hung a few pictures on the wall and squeezed a bookshelf in to hold my crafty stuff and planners and books. Ok, SOME of my books.
The Result
I adore my tiny office. I share it with some Christmas decorations, bike helmets, running accessories, and 90% of the time my dog is snoozing at my feet. Frequently there is at least one child in here, but they usually aren’t yelling so I count that as a win.
My dog, photo by the author
My “stuff” has a home, and I’m not in fear of it being spilled on or being in the way of the rest of the family trying to live their lives. I can stand up and walk away instead of having my work following me around the house.
I feel much calmer on the inside, too. I can shut the door when I need to make a phone call. If I need to quit working to do parenting stuff, I shut the office door and walk away without having to put every single thing where I know it won’t be bothered. | https://medium.com/home-sweet-home/i-converted-a-closet-into-an-office-d157b61f8226 | ['Brynn Mahnke'] | 2020-08-17 20:35:25.628000+00:00 | ['Work From Home', 'Mental Health', 'Home', 'Work Life Balance', 'Parenting'] |
Null is your friend, not a mistake | I’ve been programming in Java for a long, long time. I’ve learned what it takes to write and maintain big (as in million-lines of code) software in Java and I’ve witnessed industry-wide struggle to avoid and contain the dreaded NullPointerException (NPE) that seemed to plague any reasonably-sized Java codebase. This realization of the danger of the null reference had dawned on the industry way before its inventor, Tony Hoare, had admitted in 2009 that null reference was his “Billion Dollar Mistake”.
It was not that obvious back in 1996 when Java 1.0 was released. Let us take a look at just one, now famous, example of a typical Java API from that era: File.list() method. It can be used to list contents of a directory like this:
for (String name : new File("directory").list()) {
System.out.println(name);
}
It works only if the directory exists. Otherwise it throws an NPE, because list returns null . But who would ever write code like that? Not only list documentation clearly states that it returns null when directory is missing, but modern IDEs flag this particular code with a warning right away.
However, people do these kind of mistakes all the time when programming in Java. By now there is a big body of research showing how it happens. It turns out that most of the time our API functions are not supposed and are not expected by other developers to return null. In corner cases, like the absence of something, it is a convention in Java to return some “null object” (empty collection, unfilled domain object, etc) or, as a lesser evil than returning null, to throw an exception. That is how Files.newDirectoryStream — a modern version of File.list , is designed — no nulls anywhere.
So when null does appear as a result of a function in some special case, as performance optimization, uninitialized reference field, etc, it often catches other code off-guard, unprepared to deal with it. Not only it is rare to be having to deal with nulls, but the code you have to write for dealing with nulls in Java is long and verbose:
String[] list = new File("directory").list();
if (list != null) {
for (String name : list) {
System.out.println(name);
}
}
No wonder you’d rather choose not to write it, unless absolutely necessarily (necessity often happens when your customer discovers NPE in production).
Fear of null leads to some extreme practices. There are Java coding styles that forbid null completely, forcing heinous workarounds upon developers. Have you seen a codebase where every domain object implements a special Null interface and must provide manually coded “null object” instance? Lucky if you have not, but I bet you’ve seen Optional<T> wrappers that pollute Java code only for the sake of avoiding nulls.
There are collection-like APIs that forbid null elements out of extra caution and there are members of core Java team that consider support for nulls in Java collection framework to be a mistake. This is extremely sad.
The truth is that the concept of null is not mistake, but Java’s type-system, that considers null to be a member of every type, is. See, "abc" is a valid String in Java and null is a valid String , too, yet you can use all the string methods like substring on the former, but attempts to use them on the latter universally fail at runtime. Is it type-safe? Not really. It is normal when some operations fail at runtime on certain values of a type (like division by zero), but when a value leads to runtime failure of all operations, it shows that the value should not belong to this type in the first place. All those NPEs in Java programs indicate an obvious flaw in Java type-system.
More type-safe programming languages, like Kotlin, fix this flaw by properly incorporating the concept of null into the type system. Adding inspections and warning does help, too, but it is not enough. Clearly, a sound type system must only allow such values of type String that can support its operations, so in Kotlin putting null into a variable of type String is not just a warning, but a type error akin to assigning an integer value 42 to a variable of type String .
Proper null support in type system is a game-changer for API design. There is no reason whatsoever to fear null anymore. Having some functions that return a nullable type String? side-by-side with others that return non-null String is as normal as having some functions that return String side-by-side with others that return Int . They are simply different types with different and safe sets of operations available for them.
Type-safe null is better, more efficient, and less verbose alternative to represent all sorts of “missing results”. Look at Kotlin standard library for inspiration, for example. There is String.toIntOrNull() function that parses a string to an integer, if possible, or returns null if not. It is joy to use. Writing a command-line application that takes an integer parameter and properly complains on its absence is easy:
fun main(args: Array<String>) {
val id = args.getOrNull(0)?.toIntOrNull()
?: error("id expected")
// ...
}
Embrace null in your API design. Null is your friend with Kotlin. There is no reason to fear it and no reason to work around it with “null object” pattern or wrappers, let alone with exceptions. Proper use of nulls in your APIs results in more readable and safer code, free from boilerplate.
Further reading
If you liked the topic and want to know more details from the standpoint of language design then consider reading the companion to this story— “Dealing with absence of value”. | https://elizarov.medium.com/null-is-your-friend-not-a-mistake-b63ff1751dd5 | ['Roman Elizarov'] | 2019-03-02 21:49:30.446000+00:00 | ['Java', 'Null', 'Kotlin'] |
Principal Component Analysis | Principal Component Analysis
Step by step intuition, mathematical principles and python code snippets behind one of the most important algorithms in unsupervised learning
This animation shows how the covariance matrix of the projected points (A…E) get diagonalized when the rotating direction reaches two special directions (being the solution to the eigen decomposition equation below). To verify it, slow down the video or try to pause it when the direction overlaps the orange dotted line (first principal component) or the second pink dotted line (second Principal componemt). You will also notice that the diagonal elements of the matrix reaches their maximum values (max variance). (Video created by the author using Geogebra6 sw)
This formula is called eigen-decomposition equation. Initial works started in the end of 1800 but only in recent times has become of practical usage in big data analysis thanks to advances in (personal) computing power.
Hi, everybody, my name is Andrea Grianti in Milan, Italy. I wrote this to share my thoughts after many books and papers read on the subject. This is not a textbook but a starting point for further understanding of the topic. As the math/algebra behind is rather tough I divided it into four parts:
PCA Intuition Math/Algebra (easy) Math/Algebra (difficult) Python snippets
1. PCA Intuition
If you have a large data sets with thousands/millions of observations (rows) and hundreds of different variables (columns) one of the first objectives is verifying if it’s possible simplifying and reducing the dataset in order to facilitate the analysis work on a much smaller subset of the original data.
Direct elimination of variables is the obvious way but it has clearly impact on the information content of your data set. Too much or a wrong eliminations and your data set becomes useless, too less and the data set remains large and difficult to analyse.
The term ‘information’ is quite a generic subject and it’s difficult to define it. It depends on the data set. A data set could contain information for me and nothing to others and vice versa.
We could try to define the quantity of information in a data set using concepts like “Information Content” which is a concept bound to the probability that a specific value, among all those possible in the context of a variable of a data set, happens.
Following this concept, as much variety of possible outcomes a variable x has, lower is the probability to predict its value and therefore higher the information content is.
This is our first assumption to keep in mind: higher variance => higher information content even if only the context and data can tell if this assumption is good or not.
As in our data set there are many variables, correlations will likely exist and if they are linear they can be considered a form of “background noise” and redundancy. As for the first assumption only the specific context can tell if linearity exists in the correlations. If the assumption holds we can then say that we should find a way to eliminate the background noise from our data.
Why these two assumptions are important ? Because to reduce the dimensionality of a data set we should evaluate the contribution of each variable to the overall variance(=information) of the dataset in order to choose those with the greatest contribution and discard those with lowest contribution.
The battlefield for this operation is the Covariance Matrix (or the Correlation Matrix which is closely linked by the relation below, but I don’t consider that here as it is not relevant in relation to the understanding the math principles behind)
Relationship between simple covariance and correlation for two variables: sigmas being their standard deviations and rho being the correlation coefficient. More the two variables x and y are uncorrelated (low rho) less covariance we have. At an extreme where correlation is zero, covariance is zero. At the other when correlation is maximum rho=1 and covariance is the product of the standard deviations of each variable
Covariance Matrix of the X original data
When we calculate the Variance/Covariance Matrix of our original data (Cx) we see the variances along the diagonal but we also see the joint covariances in the off diagonal elements which are a (difficult to interpret) measure of the magnitude of the ‘synchronicity’ between one variable and the others.
Because of course we cannot modify our original data X in order to eliminate the correlations between variables (unless we remove completely a variable which is always possible but risky because we could unwillingly remove important information), we can try to find a way to ‘transform’ X into a different data set Y having special (eigen-…) link with X but built in a way that the new variables of Y would have a different Covariance matrix Cy where variances of these variables would be “isolated” from the interefences (correlations) of the other variables (=> covariances = 0).
This is where we would like to land. With a new covariance matrix Cy which is diagonalized. Means that we are looking to solve a problem to find Y, bound to X (we’ll see how), where all the Ys are uncorrelated between them. In this way the covariances in Cy would be zero and the variances isolated. This would clearly allow us to define the total variance of the data set as the sum of diagonal elements (Trace).
This operation is the magic of the eigen decomposition equation. By solving that equation we will find a way to transform the original X data set and by consequence the covariance matrix Cx into a new dataset Y and by consequence a diagonalized covariance matrix Cy using a special transformation matrix B.
The following figure slightly change the letters of the eigen decomposition equation (the one in red) to let it be coherent with letters used in this paper. When solved we will verify our solution (in Python) using matrix form of the same equation (eq.3,4,5) .
Assuming we are able to solve that equation… so what ? If using original data X, I was forced to work with Cx with the problem of being unable to isolate single variables contributions to overall variance based on the presence of correlations, now, solving that equation, I could work with a new Y data set whose covariance matrix is Cy and it’s diagonal so the variance contribution of every variable of Y is clearly defined and not impacted from joint intereferences/correlations. With that we can later decide, based on the values of their variances, where to cut the Y data set to keep a subset of the most important Y variables based on the weight of their contribution with respect to the sum of the values of the variances in Cy.
What’s the price to pay when we work with Y instead of X? Because variables of the new Y data set (we will see this) are a linear combination of the original X data , their meaining is undefined. So to give a contextual meaning to the y variables it’s required a certain acumen to make them ‘talking’ from a practical stand point. Of course the matter is a lot complex and there are many details we skip here, but to frame the problem and solution this is it.
In any case to go from X to Y (and from Cx to Cy) we need to understand the logic of the transformation (through a still to be defined B matrix) we can do on our original data. So we need to talk (study) about projections and linear transformations as they are strictly connected.
2. Math/Algebra behind PCA (easier)
Projection concept: in short what we draw in charts depends on the system of coordinates we use to represent the data. Think if you like to perspective in art: it’s a projection of real life measures transformed according to specific rules.
In 3 dimensions for example we use the cartesian system which is represented by a matrix (E) of orthonormal vectors (orthogonal vectors of length=1).
When we have for example a data set X (measures are in 3 dimensions: column vectors x1,x2,x3), to calculate the variance of variable x1 we apply the usual formula for variance. But what we actually do is to project the coordinates of each data point of X on the 3 directions in E through a dot product.
So in general we have X･E=X. This projection operation is “transparent” with cartesian axis and we don’t even realize that as it is “automatic”.
But if we decide to move away from Cartesian system we can build a similar set of orthonormal vectors that define a new basis B made of b1, b2, b3 (instead of E made of e1, e2, e3) like this:
And if we dot multiply (for example) a generic row vector of our data set X containing the coordinates of a single data point:
generic row of data, (1,3) means 1 row, 3 columns
with B (=3 rows,3 columns) we obtain a new row vector (=1 row, 3 columns):
A generic row of Y corresponding to a generic row of X
where each element of this vector has new “measures”/”coordinates” given by:
the three equations here are the explosion of the yi equation in the previous figure. So dimensionally the shape is (1,3)
In algebra terms and in general in our case we change basis simply with:
the first is the matrix form. In our case Y is made of 3 columns vectors (y1,y2,y3 with m rows) and B is made of 3 column vectors (b1,b2,b3 with 3 rows). So Y is (m,n) = X (m,n) . B (n,n). The 3 equations, when the solution to B will be found, will represent the Principal Components.
where X is the original data matrix (m rows by n columns) in basis E , B is a new orthonormal basis (n by n),Y (m by n) is the resulting new data matrix with the measures in the new basis B.
Note that y1,y2,y3 are new variables, whose meaning is to be semantically defined as they are s linear combination of all the X rows weighted by the corresponding components of the b1,b2,b3 vectors respectively. That is:
y1 is made of every X row weighted by b1=[b11,b12,b13]
y2 is made of every X row weighted by b2=[b21,b22,b23]
y3 is made of every X row weighted by b3=[b31,b32,b33].
So with a solution for B we could work with the y’s new variables. But what B should we use to reach of our objective to find Y with a Cy diagonal matrix?
B can be built progressively starting to find the first unit vector (direction) that maximizes the variance of X projected along that direction. Once we find a max value of variance (which is the first eigenvalue lambda1) which corresponds to a specific unit vector b1 (which is the eigenvector b1) we have found the first principal component PC1: y1=X.b1
To evaluate y1 in terms of how much variance explains, at this stage you can divide lambda1 by the trace of the covariance matrix Cx.
Then with the same principle we can find the second direction b2 (second eigenvector) as the one that maximize the variance (second eigenvalue) between all the possible projections of X along a second direction of unitary length and orthogonal to b1. When found this is the second principal component: PC2: y2=X.b2
Then the third direction b3 maximize the variance of X along a third direction defined again by a unit vector that must also be orthogonal to both b2 and b1. When found this is the third principal component: PC3: y3=X.b3
At the end of these iterations we would have built a special B=[b1*,b2*,b3*] (eigenvector matrix), a new dataset Y =[y1=X.b1, y2=X.b2, y3=X.b3] made of 3 ‘principal components’ ranked by strength of variance, a special vector of eigenvalues lambdas=(lambda1, lambda2, lambda3) where each lambda is the variance of each y1,y2,y3.
Of course the iteration can go on up to n dimension. At the end we have all the elements to reduce Y according to the corresponding weight of every varianceover the total variance sum(lambda1,lambda2,lambda3). Hopefully with few variables of Y we can “explain”” a large percentage of total variance and we have reached our objective to reduce a large dimensional dataset in X to few dimensions in Y.
3. Math/Algebra behind PCA (difficult)
The above was the verbose part to explain a manual iterative procedure to solve the eigen decomposition equation. Now we take a look at the whole mathematical procedure from the start to the solution of the eigen decomposition equation.
We call X the original data set (m rows x n columns) where each variable in columns have been already “centered” around their respective means, Cx is the covariance matrix of X, b1 is an unknown vector of n elements, y1 is a projection of X along an unknown vector b1: y1=X.b1
Variance of y1:
To maximize Var(y1) and find the corresponding first direction we use the Laplace equation with the Lagrange multiplier. The symbol is the gradient, f is the function to maximize which is in our case the variance of X.b1, g is the function where we set the constraint that the length of the b vector must be 1:
Fig.1
Partial derivatives of f over b1 (the unknown vector of n components) is:
Fig.2 If you want to find out why you can try with a simple b1 vector (b11,b12) a simmetric Cx like ([4,2][2,1]) and again b1. You will have a second degree polinomial and the two partial derivatives (for b11 and b12) make up a matrix which is 2 times the Cx multiplied by b1
Partial derivatives of g over b1 is:
Fig.3
The eigen decomposition equation becomes:
Fig.4
To find some b1 <>0 we must find the determinant of term in parenthesis and set it equal to zero. Only in this case the system of n equations in n unknown components of b1 has non trivial solutions:
Fig.5
The solution to this system of n equations in n unknowns will likely give (I keep it simple because there could be exceptions) n different lambdas. These lambdas represent the n different variances of the projected points of X along still to be determined directions.
Now we should assign the max of lambdas (let’s call it lambda1) found into the equation (in Fig.1) and find b1 by solving the system given by:
Fig.6
After we have found b1, we take lambda2 and we solve again the system in the equation above to find b2 and so on until for all the lambdas we have all the b vectors.
The calculations as you can imagine are very long for just 3 variables, that’s why we need Python and numerical algorithms to make the dirty work.
In the end we will (exceptions apart) have: the eigenvector B matrix as a sequence of b1, b2, b3 … bn column vectors; the eigenvalue vector L(ambda) representing the variances of every y of Y (being the Principal Components) or the magnitude of the eigenvectors.
4. Python snippets:
There are many examples and libraries around but here I want to use Python just to show with just numpy code snippets that you can quickly try your sample small dataset and understand what’s going on by looking at the results. I skipped the print statements as you can work on console and check yourself the content of the variables. Even charting is omitted but you can do that as you like with matplotlib or seaborn. Here the point is not building an app, but showing the understanding of the logic.
import numpy as np
import pandas as pd pd.options.display.max_columns = 200
pd.options.display.width=200
pd.options.display.max_rows=200 def fcenter(X):
data_mean=np.mean(X,axis=0) #calc mean by column
X=X-data_mean #centered data
return X def fcov(X):
covx=(X.T.dot(X))/X.shape[0] #calc cov matrix of X
#alternative to: covx=np.cov(X,rowvar=False,ddof=0)
return covx def feigen_decomp(Cx):
eigval,eigvec=np.linalg.eig(Cx) #solve eigen equation
return eigval,eigvec X=np.array([[1,1,6],[4,2,9],[2,-2,3],[-3,3,1],[-5,1,7]]) #some data
Xc=fcenter(X) #centered data
#----------------------
#this shows that variance does not change when centering data
vX=np.var(X,axis=0) #variance of columns of X
vXc=np.var(Xc,axis=0) #variance of columns of Xc
#----------------------
Cx=fcov(Xc) #Cx=covariance of Xc
L,B=feigen_decomp(Cx) #L=eigenvalues vector, B=eigenvectors matrix
Lw=L/L.sum() #weight in % of every PC (not cumulative)
Y=Xc.dot(B) #Y=Principal Components matrix (in columns = y scores)
Cy=fcov(Y) #shows that diagonal of covariance of Y coincides con L
#----------------------
#Loadings analysis
Loadings=np.sqrt(L)*B #see comments
Loadings_sq=Loadings**2 #see comments
Loadings_sq.sum(axis=0) #see comments
Loadings_sq.sum(axis=1) #see comments
One note on Loadings: Loadings tells how much of the square root of each eigenvalues in L go into the components of each eigenvector in B. In other words: high values in the Loadings matrix in positive or negative imply high relevance of corresponding original x variable (which are in rows) in relation to Principal Components (in columns). The sign tells about a positive or negative correlation between that x with that PC (grow or decrease together, or when one grows the other decrease and vice versa). Near zero values imply low releavance of that x with respect to that PC.
Second note: Loadings_sq matrix is the square of Loadings matrix and represent vertically how much of the variance of each PC is explained by each variable x of X: in fact the sum of each column is equal to L, and horizontally how much of the variance of each x is explained by each PC: infact the sum by row is equal to variance of x.
Loadings are important for trying to define a name of the PCs in relation to the names of the most relevant x that contribute to the score of the PC.
5. Conclusions
I hope you appreciate this long paper it and I hope to prepare a practical use case in one of the next papers using just python and focusing on the displaying results of PCA application and meaning using seaborn library. | https://towardsdatascience.com/principal-component-analysis-3c39fbf5cb9d | ['Andrea Grianti'] | 2020-10-09 17:00:08.929000+00:00 | ['Machine Learning', 'Python', 'Analytics', 'Unsupervised Learning', 'Mathematics'] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.