chosen
int64 353
41.8M
| rejected
int64 287
41.8M
| chosen_rank
int64 1
2
| rejected_rank
int64 2
3
| top_level_parent
int64 189
41.8M
| split
large_stringclasses 1
value | chosen_prompt
large_stringlengths 236
19.5k
| rejected_prompt
large_stringlengths 209
18k
|
|---|---|---|---|---|---|---|---|
38,296,126 | 38,296,083 | 1 | 3 | 38,295,179 | train | <story><title>Show HN: Tiny LLMs – Browser-based private AI models for a wide array of tasks</title><url>https://tinyllms.vercel.app/dashboard</url></story><parent_chain></parent_chain><comment><author>ilaksh</author><text>Here is an actual LLM that runs in the browser: <a href="https:&#x2F;&#x2F;webllm.mlc.ai&#x2F;#chat-demo" rel="nofollow noreferrer">https:&#x2F;&#x2F;webllm.mlc.ai&#x2F;#chat-demo</a><p>I cannot make this work on Chrome in Linux. I even tried with --enable-dawn-features=allow_unsafe_apis --enable-unsafe-webgpu</text></comment> | <story><title>Show HN: Tiny LLMs – Browser-based private AI models for a wide array of tasks</title><url>https://tinyllms.vercel.app/dashboard</url></story><parent_chain></parent_chain><comment><author>TOMDM</author><text>Which one of these is an LLM?
The youtube summary one maybe?
Doesn&#x27;t seem to run in the browser.</text></comment> |
18,382,717 | 18,382,260 | 1 | 2 | 18,381,605 | train | <story><title>How hustle culture took over advertising</title><url>https://digiday.com/marketing/welcome-hustletown-hustle-culture-took-advertising/</url></story><parent_chain><item><author>village-idiot</author><text>I always thought that hustler was synonymous with confidence artist or a low level scammer. That it’s now considered a positive word is fascinating to me.</text></item><item><author>esotericn</author><text>So, this is the industry that:<p>- puts flashing colored boxes in front of our faces at any possible opportunity<p>- requires us to click through multiple popups to access almost any for-profit website<p>- has normalized the act of wasting human life at a mass scale by injecting multi-minute time-wasting sessions in activities<p>&quot;How hustle culture took over&quot;? It&#x27;s a hustle by default. I&#x27;d call it a swindle.</text></item></parent_chain><comment><author>sin7</author><text>Hustler is used a lot in rap. It&#x27;s always been there. A hustler hustles. They put in work. They are in the grind. It&#x27;s all about turning a nickel into a dime. In other words, they sell a lot of crack. Nothing deep.<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=xZbwTCrnktY" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=xZbwTCrnktY</a> - Cassidy - I&#x27;m a hustler<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=sv6CbP4kVno" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=sv6CbP4kVno</a> - Lil Wayne - Hustler Muzic</text></comment> | <story><title>How hustle culture took over advertising</title><url>https://digiday.com/marketing/welcome-hustletown-hustle-culture-took-advertising/</url></story><parent_chain><item><author>village-idiot</author><text>I always thought that hustler was synonymous with confidence artist or a low level scammer. That it’s now considered a positive word is fascinating to me.</text></item><item><author>esotericn</author><text>So, this is the industry that:<p>- puts flashing colored boxes in front of our faces at any possible opportunity<p>- requires us to click through multiple popups to access almost any for-profit website<p>- has normalized the act of wasting human life at a mass scale by injecting multi-minute time-wasting sessions in activities<p>&quot;How hustle culture took over&quot;? It&#x27;s a hustle by default. I&#x27;d call it a swindle.</text></item></parent_chain><comment><author>Cthulhu_</author><text>Hustler yes, but hustling is a verb, often used in e.g. sports movies &#x2F; events.</text></comment> |
17,829,144 | 17,828,879 | 1 | 3 | 17,828,478 | train | <story><title>Jepsen: Dgraph 1.0.2</title><url>https://jepsen.io/analyses/dgraph-1-0-2</url></story><parent_chain></parent_chain><comment><author>mrjn</author><text>Author of <a href="https:&#x2F;&#x2F;dgraph.io" rel="nofollow">https:&#x2F;&#x2F;dgraph.io</a> here. I&#x27;d say this is one of the best collaborations we&#x27;ve done. Due to Jepsen tests, we were able to proactively identify and fix issues in a complex real time distributed graph DB, that is Dgraph. Dgraph&#x27;s graph sharding, replication and transaction implementation is unique and it took more work to make it all work correctly, while still achieving good performance, and without utilizing any third party system.<p>I&#x27;d like to thank Kyle for his work on Dgraph. We&#x27;ll continue to expand Jepsen tests to try and identify more issues and fix the remaining ones.<p>I&#x27;m around if you have questions! Cheers, Manish</text></comment> | <story><title>Jepsen: Dgraph 1.0.2</title><url>https://jepsen.io/analyses/dgraph-1-0-2</url></story><parent_chain></parent_chain><comment><author>BMarkmann</author><text>I honestly don&#x27;t know that there are any posts I get more excited about on HN than Jepsen analyses. The number of landmines distributed systems create and the ability to suss them out in such detail &#x2F; depth is really incredible. Kudos.</text></comment> |
29,907,890 | 29,905,537 | 1 | 3 | 29,905,288 | train | <story><title>Amish Hackers (2009)</title><url>https://kk.org/thetechnium/amish-hackers-a/</url></story><parent_chain></parent_chain><comment><author>mdip</author><text>The Amish are really misunderstood by people on the outside (myself included). This article does a great job of addressing those misconceptions and I enjoyed reading it after a long afternoon spent talking about the subject with a friend who was a midwife to the Amish community in northern Indiana[0]. My knowledge up to that point was &quot;Those are the folks that I get my turkeys from who ride around in buggies.&quot;<p>The conversation started with an odd statement -- she had 7 children under 18 all home-schooled at least until High School (they were given the choice to attend public school at that point) and is a devout Christian. I was asking her why she allowed them to attend public school in High School given her rigidity on the merits of Home Schooling and her involvement in the Home Schooling movement.<p>She said &quot;They know enough to graduate by the time I&#x27;m done with them, so it&#x27;s an opportunity for them to evangelize to the other kids and enjoy themselves. I even let them attend an occasional party provided it&#x27;s not an <i>Amish party</i>&quot;<p>I chuckled ... wouldn&#x27;t want to be negatively influenced by The Amish! I had no idea what &#x27;rumspringa&#x27; was at that point. She explained and I don&#x27;t think I blinked the entire time. Her son pointed out &quot;The Amish have the best cars and every hard drug around&quot;. Rumspringa[0] is a period of time after age 16 where Amish children are basically allowed to live conventionally in order to make the decision, on their own, if they want to join the community as adults. Many buy (really nice) cars, many -- having experienced freedom for the first time in their lives -- make every mistake a late teenager can make and seem to make it at a level that pales in comparison to my experience as a teenager. The cynical -- and she was one -- believe it is used to <i>encourage</i> children to make every mistake the modern world has to offer so that they are <i>trapped</i> into making the choice to join the Amish community.<p>Because of my experience buying Amish turkeys, which are sold as cage free, hormone free, and otherwise considered superior from a health-standpoint to conventional turkeys[1]. I was surprised to find out that my midwife friend didn&#x27;t buy Amish turkeys despite being <i>extremely</i> into the &quot;all natural&quot; way of thinking[2].<p>The article mentions that the technology choices which are allowed are limited by the parish and are inconsistent among parishes. Gas engines are usually allowed. Electricity is allowed with some restrictions, though AFAIK, being off-grid is required -- if you can generate it from solar&#x2F;gas generator, it&#x27;s OK and gas refrigerators or gas generators powering refrigerators were allowed.<p>While Amish Turkeys may not use conventional methods -- hormone-free&#x2F;pesticide-free feed -- such is not the case with pretty much anything else. Velveeta, apparently, can be found in large quantities in every Amish kitchen in the community she served and is used in so many meals that she turns down Amish meals. Gelatin (non-desert Jello) is another staple of the diets of the people she served. Her take was &quot;Amish food is unhealthy and disgusting&quot;.<p>Outside of that, she explained to me that <i>literally everything is grown conventionally</i>. The parish she serves has no rules against pesticide use&#x2F;GMO&#x2F;conventional fertilizers and the like are <i>common</i> -- basically anything that will keep a plant from dying is <i>common</i>. On the surface, this was a strange thing for me to wrap my brain around given my Amish Turkey experience, however, the explanation she gave was &quot;Velveeta notwithstanding, they rely on their own farming for the majority of their food and use the most effective&#x2F;least expensive options available in order to make sure there&#x27;s enough&quot;. I still can&#x27;t get over the Velveeta. Nasty. Her opinion was that among the conventional farms, Amish farms were <i>the worst</i> and she wouldn&#x27;t consume <i>anything</i> grown on an Amish farm.<p>As a midwife, though, she had a bit of experience with some of the darker side of Amish life (a lot is explained in The Devil&#x27;s Playground). Bed Courtship is among the oddest practices for a community that believes per-marital sex is evil. More than a few of her baby-delivery roles were for late teenage girls who became pregnant before they were married. The practice involves, basically, bringing a girl home to lay in the same bed with as part of a date. Sex is forbidden, and it sometimes includes a barrier that is supposed to separate the two participants, but that practice is rare (as is Bed Courtship, in general). Her opinion was that Bed Courtship, much like &#x27;rumspringa&#x27;, is designed to <i>cause</i> children to make mistakes that result in them feeling forced to stay Amish for life.<p>A small anecdote I share with people choosing to make a switch to a programming&#x2F;IT career was that her husband -- an IT systems analyst -- worked with a gentleman who was Amish up until his late 20s[3]. He was a former-Amish software developer. &quot;If that guy can do it, anyone can.&quot;<p>[0] I strongly recommend the documentary The Devil&#x27;s Playground if you want to learn about &#x27;rumspringa&#x27;. <a href="https:&#x2F;&#x2F;www.imdb.com&#x2F;title&#x2F;tt0293088&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.imdb.com&#x2F;title&#x2F;tt0293088&#x2F;</a><p>[1] I don&#x27;t challenge those labels, but I also can&#x27;t validate them.<p>[2] She refused to buy pasteurized milk, opting for milk from a neighbors farm. Her husband refused to drink it without passing it through their own pasteurizer.<p>[3] Yes, they do shun family members who leave the community.</text></comment> | <story><title>Amish Hackers (2009)</title><url>https://kk.org/thetechnium/amish-hackers-a/</url></story><parent_chain></parent_chain><comment><author>adrianN</author><text>The Amish can evaluate the effects of new technology on their communities by having the rest of the world act as a guinea pig. I think society as a whole could benefit from thinking about the effects of new technology on communities before adopting it, but reasoning about that from first principles seems very hard and putting the genie back into the bottle after a technology has been widely deployed is essentially impossible.</text></comment> |
38,295,363 | 38,295,208 | 1 | 3 | 38,291,015 | train | <story><title>Sweden plans new nuclear reactors by 2035, will share costs</title><url>https://www.reuters.com/business/energy/sweden-plans-new-nuclear-reactors-by-2035-can-take-costs-2023-11-16/</url></story><parent_chain></parent_chain><comment><author>philipkglass</author><text>This article cuts off before revealing much information to non-subscribers.<p>Here&#x27;s the corresponding Reuters article, which I could read in its entirety:<p><a href="https:&#x2F;&#x2F;www.reuters.com&#x2F;business&#x2F;energy&#x2F;sweden-plans-new-nuclear-reactors-by-2035-can-take-costs-2023-11-16&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.reuters.com&#x2F;business&#x2F;energy&#x2F;sweden-plans-new-nuc...</a><p>Key points:<p>- They intend to build the equivalent of two new conventional nuclear reactors by 2035 (presumably meaning something like 2000 megawatts of generating capacity, but it could come from small reactors).<p>- They intend to build the equivalent of 10 new conventional reactors by 2045.<p>The government has not yet announced what sort of funding model will be used to get the new reactors built.</text></comment> | <story><title>Sweden plans new nuclear reactors by 2035, will share costs</title><url>https://www.reuters.com/business/energy/sweden-plans-new-nuclear-reactors-by-2035-can-take-costs-2023-11-16/</url></story><parent_chain></parent_chain><comment><author>fifilura</author><text>The current government is pretty weak and one of their promises before the election was to increase nuclear energy.<p>So this is pretty much something they had to announce to keep voters happy.<p>Whether it will happen in practice is another question.<p>I guess at some point someone will have to make a business plan that shows that it is also a good investment.</text></comment> |
15,982,736 | 15,981,000 | 1 | 3 | 15,979,714 | train | <story><title>Facebook’s Political Unit Enables Propaganda</title><url>https://www.bloomberg.com/news/features/2017-12-21/inside-the-facebook-team-helping-regimes-that-reach-out-and-crack-down</url></story><parent_chain><item><author>decebalus1</author><text>Remember this? <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=15790687" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=15790687</a><p>To some extent it&#x27;s still going on. Protests are still happening (and it&#x27;s really bad) so lately I&#x27;ve been noticing a bunch of - what I think are - shill accounts. Couple of days before I went through some of the posts in news post about the protests and identified several profiles with no personal pictures, no posts, posting in broken Romanian about:<p>- anti-protests<p>- anti-EU and basically nationalistic bullshit<p>- religion and how that will solve everything<p>- government does nothing wrong and protesters are a bunch of thugs<p>- links to well-known fake news websites<p>- inciting violence while appearing to be pro-protest<p>Groups of accounts such as these are brigading some popular news postings and I don&#x27;t really find them inside the anti-protester echo rooms.<p>the accounts look and feel similarly to the spam profiles which were prevalent a few years back having supermodel headshots as profile pictures.<p>I reported about 12 such profiles and the response from Facebook was that I should block these accounts as a resolution, as they aren&#x27;t doing anything wrong per Facebook community guidelines and they are legit accounts.<p>In contrast, &#x27;famous&#x27; people who are famously anti-government (and trolls, that&#x27;s true, like this guy: <a href="https:&#x2F;&#x2F;www.facebook.com&#x2F;macacaur" rel="nofollow">https:&#x2F;&#x2F;www.facebook.com&#x2F;macacaur</a>) are getting their accounts disabled every couple of weeks and require photo ID proof to re-enable.</text></item></parent_chain><comment><author>narrator</author><text>How do you know what goes on outside of your direct perception anyway? It used to be conventional media channels and they would print what was arguably propaganda, but there wouldn&#x27;t be any Russian trolls to poke holes in it.<p>Imagine if the Iraq WMD story broke now. All those Russian trolls spamming Facebook running around saying it was bullshit. That would be total chaos! Why don&#x27;t people trust the mainstream?<p>The reality is that propaganda works. You think you know what fake news is and what&#x27;s reality, but we never did and most people follow the time-honored tradition of letting our chosen authority figures tell us who is conspiring with the bad guys. This frees us from the cognitive pain of considering their arguments and evidence because they are puppets of some nebulous evil forces and therefore safe to ignore.</text></comment> | <story><title>Facebook’s Political Unit Enables Propaganda</title><url>https://www.bloomberg.com/news/features/2017-12-21/inside-the-facebook-team-helping-regimes-that-reach-out-and-crack-down</url></story><parent_chain><item><author>decebalus1</author><text>Remember this? <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=15790687" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=15790687</a><p>To some extent it&#x27;s still going on. Protests are still happening (and it&#x27;s really bad) so lately I&#x27;ve been noticing a bunch of - what I think are - shill accounts. Couple of days before I went through some of the posts in news post about the protests and identified several profiles with no personal pictures, no posts, posting in broken Romanian about:<p>- anti-protests<p>- anti-EU and basically nationalistic bullshit<p>- religion and how that will solve everything<p>- government does nothing wrong and protesters are a bunch of thugs<p>- links to well-known fake news websites<p>- inciting violence while appearing to be pro-protest<p>Groups of accounts such as these are brigading some popular news postings and I don&#x27;t really find them inside the anti-protester echo rooms.<p>the accounts look and feel similarly to the spam profiles which were prevalent a few years back having supermodel headshots as profile pictures.<p>I reported about 12 such profiles and the response from Facebook was that I should block these accounts as a resolution, as they aren&#x27;t doing anything wrong per Facebook community guidelines and they are legit accounts.<p>In contrast, &#x27;famous&#x27; people who are famously anti-government (and trolls, that&#x27;s true, like this guy: <a href="https:&#x2F;&#x2F;www.facebook.com&#x2F;macacaur" rel="nofollow">https:&#x2F;&#x2F;www.facebook.com&#x2F;macacaur</a>) are getting their accounts disabled every couple of weeks and require photo ID proof to re-enable.</text></item></parent_chain><comment><author>trhway</author><text>&gt;
are getting their accounts disabled every couple of weeks and require photo ID proof to re-enable.<p>Sounds like hi tech version if voter ID and serving basically the same purpose.</text></comment> |
41,543,968 | 41,543,429 | 1 | 2 | 41,540,902 | train | <story><title>Terence Tao on O1</title><url>https://mathstodon.xyz/@tao/113132502735585408</url></story><parent_chain><item><author>fnordpiglet</author><text>Rewind your mind to 2019 and imagine reading a post that said<p>“The experience seemed roughly on par with trying to advise a mediocre, but not completely incompetent, graduate student.”<p>With regard to interacting with the equivalent of Alexa. That’s a remarkable difference in 5 years.</text></item></parent_chain><comment><author>JumpCrisscross</author><text>The first profession AI seems on track to decimate is programming. In particular, the brilliant but remote and individual contributor. There is an obvious conflict of interest in this forum.</text></comment> | <story><title>Terence Tao on O1</title><url>https://mathstodon.xyz/@tao/113132502735585408</url></story><parent_chain><item><author>fnordpiglet</author><text>Rewind your mind to 2019 and imagine reading a post that said<p>“The experience seemed roughly on par with trying to advise a mediocre, but not completely incompetent, graduate student.”<p>With regard to interacting with the equivalent of Alexa. That’s a remarkable difference in 5 years.</text></item></parent_chain><comment><author>noch</author><text>The important point is, I feel, that most people are not even at the level of intelligence of a <i>&quot;a mediocre, but not completely incompetent, graduate student.&quot;</i> A mediocre graduate science student, especially of the sort who graduates and doesn&#x27;t quit, is a very impressive individual compared to the rest of us.<p>For &quot;us&quot;, having such a level of intelligence available as an assistant throughout the day is a massive life upgrade, if we can just afford more tokens.</text></comment> |
2,189,397 | 2,189,421 | 1 | 3 | 2,189,166 | train | <story><title>The unfortunate math behind consulting companies</title><url>http://blog.asmartbear.com/consulting-company-accounting.html</url></story><parent_chain></parent_chain><comment><author>pdx</author><text>While I agree with the general "consulting is not the path to riches" tone of the article, I have a small nit.<p>What the author is doing is adding up all the worst case scenarios into one big ugly pile of bad.<p>It reminded me of an mp3 player I designed once. I had to allocate z-height for the LCD. An LCD can have a lot of subcomponents (EL Backlight, FPC cable, rear reflective film, glass, front polarizing film, plastic protective lens, etc. All of these have min/max dimensions on the spec sheet. An inexperienced engineer, such as I was, would just add all the max dimensions to get max z-height of the LCD and design for that. But I also needed it not to rattle around, so I needed my design to also work if all the dimensions came in at their minimum. Woops! It's thickness could apparently vary wildly. I didn't want to burden my product with the complexity it would take to handle so much variance in one component.<p>I then learned that tolerances are never added up "worst case" or you'd never get anything built. Instead, you do a root mean square of them. The chances of every component in your subassembly coming in at it's absolute maximum are minuscule. The root mean square method accounts for this.<p>Anyway, I felt that the author of this piece was doing a little max tolerance stacking.</text></comment> | <story><title>The unfortunate math behind consulting companies</title><url>http://blog.asmartbear.com/consulting-company-accounting.html</url></story><parent_chain></parent_chain><comment><author>raganwald</author><text>I long time ago I was playing bridge with a genius named Arno Hobart. I asked him what he was up to, and he told me he was in the vending machine business. "Oh?" I asked, surprised that such an intelligent man would work with such mundane products.<p>"One of the things I like about this business," he explained, "Is that I make money while I sleep." Yes, he really is smarter than the average bear.</text></comment> |
33,333,615 | 33,333,168 | 1 | 2 | 33,330,864 | train | <story><title>My dad's resume and skills from 1980</title><url>https://github.com/runvnc/dadsresume</url></story><parent_chain><item><author>dkurth</author><text>As Seymour Cray said, &quot;The trouble with programmers is that you can never tell what a programmer is doing until it&#x27;s too late.&quot;<p>It can be months (at a high salary) before you really know whether a hire is likely to work out. I think it makes sense to invest more effort in screening applicants in this case.</text></item><item><author>zackmorris</author><text>If the minimalist resume is appealing, can we also bring back walk-on hiring?<p>In warehouse and construction work, if someone shows up at 7:30 AM on a Monday morning, odds are quite good that the foreman will have something for them to do. Maybe not that day, but maybe tomorrow, or maybe someone on the list above them won&#x27;t show up that week and they&#x27;ll get called. I made rent doing that in my early 20s and they even let me leave early sometimes to work on my internet business because I didn&#x27;t have a family to support and maybe someone else had a bill they needed to pay and wanted my shift.<p>Why again does boutique startup need to interview 500 overqualified people? Hire someone right away and let them quit if they want to and hire someone else. It&#x27;s just business for crying out loud.</text></item></parent_chain><comment><author>indymike</author><text>I&#x27;ve hired hundreds of developers over three decades, and this is completely wrong:<p>&gt; It can be months (at a high salary) before you really know whether a hire is likely to work out.<p>It&#x27;s only that way if you make it take that long. You should know if you have a good programmer 2-3 weeks after the hire. Here a couple things that make making great hires hard:<p>* Making it difficult to learn and understand your system.<p>* Having slow and expensive employee onboarding. I&#x27;ve seen companies spend $3-4K (not including the actual laptop) just getting a laptop to a new employee after IT gets done with it. If it&#x27;s super-expensive to make a hire, the incentive will be to keep people that aren&#x27;t getting the job done.<p>* Not looking at work output for extended periods. In short give new people tickets that can be done in a few days at most so you are able to look at work output in six days instead of measuring at six months.<p>&gt; I think it makes sense to invest more effort in screening applicants in this case<p>There&#x27;s only so much you can really screen before error in your hiring process exceeds 50%. Every step you add to a screening process has an error rate, and some are very subjective and error prone. The more screening you do, the slower you go, and honestly, the worst candidates you have to pick from. Why? Because a good programmer will be on the job market for 1-14 days (I&#x27;m not saying you are bad if it takes you longer to get hired, it&#x27;s just what we&#x27;re seeing in our recruiting software right now).</text></comment> | <story><title>My dad's resume and skills from 1980</title><url>https://github.com/runvnc/dadsresume</url></story><parent_chain><item><author>dkurth</author><text>As Seymour Cray said, &quot;The trouble with programmers is that you can never tell what a programmer is doing until it&#x27;s too late.&quot;<p>It can be months (at a high salary) before you really know whether a hire is likely to work out. I think it makes sense to invest more effort in screening applicants in this case.</text></item><item><author>zackmorris</author><text>If the minimalist resume is appealing, can we also bring back walk-on hiring?<p>In warehouse and construction work, if someone shows up at 7:30 AM on a Monday morning, odds are quite good that the foreman will have something for them to do. Maybe not that day, but maybe tomorrow, or maybe someone on the list above them won&#x27;t show up that week and they&#x27;ll get called. I made rent doing that in my early 20s and they even let me leave early sometimes to work on my internet business because I didn&#x27;t have a family to support and maybe someone else had a bill they needed to pay and wanted my shift.<p>Why again does boutique startup need to interview 500 overqualified people? Hire someone right away and let them quit if they want to and hire someone else. It&#x27;s just business for crying out loud.</text></item></parent_chain><comment><author>CamperBob2</author><text>That&#x27;s an awesome saying -- I&#x27;m surprised I&#x27;ve never heard it. It remains true after a LOT of mutation.<p><i>The trouble with programmers is that you can never tell what a programmer is doing until it&#x27;s too late.</i><p><i>The trouble with programmers is that you can never tell what a program is doing until it&#x27;s too late.</i><p><i>The trouble with programs is that you can never tell what a programmer is doing until it&#x27;s too late.</i><p><i>The trouble with programs is that you can never tell what a program is doing until it&#x27;s too late.</i></text></comment> |
40,367,379 | 40,367,490 | 1 | 2 | 40,366,137 | train | <story><title>How to get 7th graders to smoke</title><url>https://www.experimental-history.com/p/how-to-get-7th-graders-to-smoke</url></story><parent_chain><item><author>patwolf</author><text>This brought back memories of the D.A.R.E. officer coming to my elementary classroom to give us the &quot;don&#x27;t do drugs&quot; speech. At the end of the speech with feigning hesitation he&#x27;d pull out the drug briefcase to the cheers of the kids, like a band holding out their best song for the encore.<p>&quot;This here is methamphetamine, aka crystal meth or speed. If you see a man running down the street with the strength of a bear, there&#x27;s a good chance he&#x27;s high on this. It&#x27;s very bad. You don&#x27;t want to get near this--it&#x27;ll make your heart explode.&quot;<p>It&#x27;s a wonder that didn&#x27;t work.</text></item></parent_chain><comment><author>flatline</author><text>I did not get the briefcase in my DARE education. What I did get is a cop telling me that my parents having a glass of wine with dinner or a cigarette after work was just as bad a them using cocaine, and that you could OD and die on weed. All drugs are equally bad.<p>A few years later when my friends were smoking weed and laughed at my concerns, and I realized I’d been lied to, I pretty much did everything under the sun.<p>An actual education would have let me make better choices about my recreational drug use as a young person.</text></comment> | <story><title>How to get 7th graders to smoke</title><url>https://www.experimental-history.com/p/how-to-get-7th-graders-to-smoke</url></story><parent_chain><item><author>patwolf</author><text>This brought back memories of the D.A.R.E. officer coming to my elementary classroom to give us the &quot;don&#x27;t do drugs&quot; speech. At the end of the speech with feigning hesitation he&#x27;d pull out the drug briefcase to the cheers of the kids, like a band holding out their best song for the encore.<p>&quot;This here is methamphetamine, aka crystal meth or speed. If you see a man running down the street with the strength of a bear, there&#x27;s a good chance he&#x27;s high on this. It&#x27;s very bad. You don&#x27;t want to get near this--it&#x27;ll make your heart explode.&quot;<p>It&#x27;s a wonder that didn&#x27;t work.</text></item></parent_chain><comment><author>delichon</author><text>In high school around 1976 we got a talk by the swim team coach, telling us the horrors of various drugs, going into his wide personal experience with them and why we shouldn&#x27;t follow his example. He told us that if we ever had any questions he was there for us, and gave us all his personal phone number.<p>So me and some buddies got a hold of some peyote and were more than a little afraid of it. We probably would have tossed it but we had a better idea. We called Coach and asked his advice. He gave us detailed preparation instructions. We followed them and got a bit sick, but not altered. Maybe it wasn&#x27;t even peyote.<p>I&#x27;m not sure that this approach was just more nuanced than &quot;just say no&quot; or closer to &quot;just do it&quot;.</text></comment> |
31,677,623 | 31,677,324 | 1 | 2 | 31,676,964 | train | <story><title>Simple Sabotage Field Manual by United States Office of Strategic Services</title><url>https://www.gutenberg.org/ebooks/26184</url></story><parent_chain></parent_chain><comment><author>wolfgang42</author><text>My favorite bit of advice:<p><i>&gt; Anyone can break up a showing of an enemy propaganda film by putting two or three dozen large moths in a paper bag. Take the bag to the movies with you, put it on the floor in an empty section of the theater as you go in and leave it open. The moths will fly out and climb into the projector beam, so that the film will be obscured by fluttering shadows.</i><p>How do you go about surreptitiously obtaining several dozen large moths?</text></comment> | <story><title>Simple Sabotage Field Manual by United States Office of Strategic Services</title><url>https://www.gutenberg.org/ebooks/26184</url></story><parent_chain></parent_chain><comment><author>blamazon</author><text>Check out section &quot;(11) General Interference with Organizations and Production&quot;, if you have worked as part of a corporation it may feel familiar...</text></comment> |
26,166,816 | 26,165,669 | 1 | 2 | 26,159,680 | train | <story><title>What went wrong with the Texas power grid?</title><url>https://www.houstonchronicle.com/business/energy/article/Wholesale-power-prices-spiking-across-Texas-15951684.php</url></story><parent_chain><item><author>roenxi</author><text>It is annoying when this sort of event happens and everyone starts deciding what they believe to be truth within 48 hours of the outage. It makes more sense to be talking about the 2011 outages in terms of what went wrong than the 2021 ones - because we have actual information on 2011 instead of &quot;things went wrong!&quot; now.<p>I want to thank you personally for injecting a PDF into the discussion, but also positively assert that it is not obvious what just happened this week. We don&#x27;t yet know how many of the recommendations were ignored, what happened in the last decade regulator-wise or whether this round of failures are for the same or different reasons.<p>Speculation is much less useful than waiting a few months for the actual investigations. Emergencies are urgent, engineering (and political) decisions and assessments are never emergencies.</text></item><item><author>gregw2</author><text>What went wrong is that the Texas legislature which owns Texas-specific grid process to avoid interference from the Feds didn&#x27;t figure out how to also ensure Texas generator companies got compensated for weatherizing (and ensuring it was done).<p>How many of the 24+6 recommendations from the NERC&#x2F;FERC review of last time this happened in Texas (hint:2011) were taken up by the legislature or those at ERCOT they delegated responsibility to or the power generation providers?<p><a href="https:&#x2F;&#x2F;www.nerc.com&#x2F;pa&#x2F;rrm&#x2F;ea&#x2F;ColdWeatherTrainingMaterials&#x2F;FERC%20NERC%20Findings%20and%20Recommendations.pdf" rel="nofollow">https:&#x2F;&#x2F;www.nerc.com&#x2F;pa&#x2F;rrm&#x2F;ea&#x2F;ColdWeatherTrainingMaterials&#x2F;...</a><p>The eye opener to me from skimming the 2011 recommendations is that there was no explicit rating&#x2F;SLA for a power plant&#x27;s acceptable temperature operating that could be used by planners for assessing the risks of an upcoming weather event by policy planners. It&#x27;d seem pretty basic to be able to ask &quot;How many plants do we lose when temperature drops below X&quot;?
Dunno whether they fixed trying to create such a basic measurement for Texas plants, but it doesn&#x27;t seem like it.<p>If you want to know some of the specifics about what &quot;winterization&quot; means in practice for a power plant including natural gas ones, you can read some of the details in that report. It&#x27;s kinda interesting.</text></item></parent_chain><comment><author>gregw2</author><text>From an engineering standpoint, I agree that you want a true RCA and this event&#x27;s RCA will be a bit different than 2011&#x27;s. And this RCA (and its political consequences) will take time, doesn&#x27;t need to be done right this minute, etc etc.<p>However, this problem is not at its root an engineering one; it is political. While the public attention is on this, we should point attention as close as possible to the most likely cause given the information we have at the time. It&#x27;s bayesian truth and bayesian politics.<p>You&#x27;ll note I didn&#x27;t draw conclusions about which recommendations were ignored but having read through them all, there is no way they were all actually followed and we ended up where we are. I think some basics are fairly obvious. While some winterization perhaps was done, there is some degree of winterization that was never done; power plants in other states far north of us and colder than us are not having the same types of problems. Various professors in various cities in Texas who follow this stuff confirm this in various news outlets I haven&#x27;t cited here. I also wouldn&#x27;t be surprised if there was were issues with gas transport from wells through pipelines to plants that were noted in 2011 but much more severe this time due to even colder temps and unexpected by all. But still...<p>At the end of the day, Texas has optimized for cheap power and has not funded the work of reliable power. This is a political decision at the end of the day due to companies not paying for their externalities of poor service. I&#x27;m open to saying we shouldn&#x27;t blame follower-type politicians who were scared of &quot;raising electrical rates&quot; and we should blame ourselves, but let&#x27;s all acknowledge that some costs that weren&#x27;t borne should have been borne and some oversight that should have occurred didn&#x27;t occur. Is that so hard to concede at this early juncture?</text></comment> | <story><title>What went wrong with the Texas power grid?</title><url>https://www.houstonchronicle.com/business/energy/article/Wholesale-power-prices-spiking-across-Texas-15951684.php</url></story><parent_chain><item><author>roenxi</author><text>It is annoying when this sort of event happens and everyone starts deciding what they believe to be truth within 48 hours of the outage. It makes more sense to be talking about the 2011 outages in terms of what went wrong than the 2021 ones - because we have actual information on 2011 instead of &quot;things went wrong!&quot; now.<p>I want to thank you personally for injecting a PDF into the discussion, but also positively assert that it is not obvious what just happened this week. We don&#x27;t yet know how many of the recommendations were ignored, what happened in the last decade regulator-wise or whether this round of failures are for the same or different reasons.<p>Speculation is much less useful than waiting a few months for the actual investigations. Emergencies are urgent, engineering (and political) decisions and assessments are never emergencies.</text></item><item><author>gregw2</author><text>What went wrong is that the Texas legislature which owns Texas-specific grid process to avoid interference from the Feds didn&#x27;t figure out how to also ensure Texas generator companies got compensated for weatherizing (and ensuring it was done).<p>How many of the 24+6 recommendations from the NERC&#x2F;FERC review of last time this happened in Texas (hint:2011) were taken up by the legislature or those at ERCOT they delegated responsibility to or the power generation providers?<p><a href="https:&#x2F;&#x2F;www.nerc.com&#x2F;pa&#x2F;rrm&#x2F;ea&#x2F;ColdWeatherTrainingMaterials&#x2F;FERC%20NERC%20Findings%20and%20Recommendations.pdf" rel="nofollow">https:&#x2F;&#x2F;www.nerc.com&#x2F;pa&#x2F;rrm&#x2F;ea&#x2F;ColdWeatherTrainingMaterials&#x2F;...</a><p>The eye opener to me from skimming the 2011 recommendations is that there was no explicit rating&#x2F;SLA for a power plant&#x27;s acceptable temperature operating that could be used by planners for assessing the risks of an upcoming weather event by policy planners. It&#x27;d seem pretty basic to be able to ask &quot;How many plants do we lose when temperature drops below X&quot;?
Dunno whether they fixed trying to create such a basic measurement for Texas plants, but it doesn&#x27;t seem like it.<p>If you want to know some of the specifics about what &quot;winterization&quot; means in practice for a power plant including natural gas ones, you can read some of the details in that report. It&#x27;s kinda interesting.</text></item></parent_chain><comment><author>dalbasal</author><text>This is meta&#x2F;tangent, but I&#x27;ve had similar conversations a few times recently. Seems to be in the zeitgeist.<p>Re: <i>everyone starts deciding what they believe to be truth within 48 hours of the outage.</i><p>Maybe, hopefully... the epidemic is humbling us. It&#x27;s been significant and long lasting enough to rub our noses in whatever opinions we so brashly got behind too early. Professors and peasants alike. We&#x27;re more likely to stop and say &quot;I don&#x27;t know.&quot;<p>The classic example of this is governments taking credit&#x2F;blame for economic stuff. Low unemployment, high gdp, etc. Current government decisions are really unlikely to be affecting these, because stuff takes time. Meanwhile, short term data about the economy is both uncertain and fairly useless even if it wasn&#x27;t. The whole thing is so disingenuous, yet I doubt there has ever been an democratic election where this wasn&#x27;t a major factor.</text></comment> |
20,160,185 | 20,157,545 | 1 | 2 | 20,156,158 | train | <story><title>Support for U2F security keys</title><url>https://blog.1password.com/introducing-support-for-u2f-security-keys/</url></story><parent_chain><item><author>tpetry</author><text>Autofill of a password manager is a working countermeasure against phishing too: If autofill does not work there is something wrong and you should look closer...</text></item><item><author>tptacek</author><text>The primary purpose of U2F&#x2F;WebAuthn is to break phishing attacks. Code-based TOTP 2FA, the kind you&#x27;re probably using now, is already adequate to the task of making sure you&#x27;re not credential-stuffed.</text></item><item><author>josefresco</author><text>Got a free YubiKey from Wired. Then I read that mobile is a pain, and that I really need two ... it&#x27;s sat in my bag now for months, unused. I already use 2FA, and it works good enough - I&#x27;m not sold on how this will make my life better, especially on mobile.</text></item></parent_chain><comment><author>ptoomey3</author><text>My experience with password managers is that they work great for me, because I understand every sharp edge and can work around them. My experience when advising family to use them is that they invariably fail them and they get frustrated. Password managers rely on pretty gross heuristics to work. They are effectively trying to automate something built for a human (choosing, remembering, and entering a password). WebAuthn gives us a real API built by and for machines. This will make the flow much less error-prone and more secure.</text></comment> | <story><title>Support for U2F security keys</title><url>https://blog.1password.com/introducing-support-for-u2f-security-keys/</url></story><parent_chain><item><author>tpetry</author><text>Autofill of a password manager is a working countermeasure against phishing too: If autofill does not work there is something wrong and you should look closer...</text></item><item><author>tptacek</author><text>The primary purpose of U2F&#x2F;WebAuthn is to break phishing attacks. Code-based TOTP 2FA, the kind you&#x27;re probably using now, is already adequate to the task of making sure you&#x27;re not credential-stuffed.</text></item><item><author>josefresco</author><text>Got a free YubiKey from Wired. Then I read that mobile is a pain, and that I really need two ... it&#x27;s sat in my bag now for months, unused. I already use 2FA, and it works good enough - I&#x27;m not sold on how this will make my life better, especially on mobile.</text></item></parent_chain><comment><author>tedunangst</author><text>Until the next exploit that can steal passwords from autofill...</text></comment> |
17,057,993 | 17,057,471 | 1 | 2 | 17,050,994 | train | <story><title>The Sound of Madness</title><url>https://harpers.org/archive/2018/06/the-sound-of-madness/</url></story><parent_chain><item><author>danieltillett</author><text>As someone who normally doesn&#x27;t have the self-talk module running, I was amazed to learn that most people have it running all the time. I only use the module when writing, otherwise my consciousness is normally non-verbal.<p>I suspect it is related, but when I read I don&#x27;t hear the words unless I deliberately choose to read the text in my head, otherwise my process is written word -&gt; concept.</text></item><item><author>mirimir</author><text>Good article.<p>But it doesn&#x27;t mention the self-talk aka &quot;stream of consciousness&quot; that virtually everyone hears. And a key insight is that it&#x27;s not <i>our</i> consciousness. It&#x27;s basically just recordings that get played back on queue. A central aspect of meditation is becoming aware of that voice as something distinct from self.<p>So anyway, it&#x27;s best to think &quot;Thanks for sharing :)&quot; and then to do something intentional.</text></item></parent_chain><comment><author>TeMPOraL</author><text>It seems to be similar with mental pictures.<p><a href="https:&#x2F;&#x2F;www.lesswrong.com&#x2F;posts&#x2F;baTWMegR42PAsH9qJ&#x2F;generalizing-from-one-example" rel="nofollow">https:&#x2F;&#x2F;www.lesswrong.com&#x2F;posts&#x2F;baTWMegR42PAsH9qJ&#x2F;generalizi...</a><p>&quot;The debate [over mental imagery] was resolved by Francis Galton (...). Galton gave people some very detailed surveys, and found that some people did have mental imagery and others didn&#x27;t. The ones who did had simply assumed everyone did, and the ones who didn&#x27;t had simply assumed everyone didn&#x27;t, to the point of coming up with absurd justifications for why they were lying or misunderstanding the question. There was a wide spectrum of imaging ability, from about five percent of people with perfect eidetic imagery to three percent of people completely unable to form mental images.&quot;<p>Personally, I&#x27;m someone with small-talk module running almost constantly, verbalizing most of my thinking and reading. On the other hand, I have very low capability of forming mental images.<p>I wonder how many other things in brain are like that, widely differing between people.</text></comment> | <story><title>The Sound of Madness</title><url>https://harpers.org/archive/2018/06/the-sound-of-madness/</url></story><parent_chain><item><author>danieltillett</author><text>As someone who normally doesn&#x27;t have the self-talk module running, I was amazed to learn that most people have it running all the time. I only use the module when writing, otherwise my consciousness is normally non-verbal.<p>I suspect it is related, but when I read I don&#x27;t hear the words unless I deliberately choose to read the text in my head, otherwise my process is written word -&gt; concept.</text></item><item><author>mirimir</author><text>Good article.<p>But it doesn&#x27;t mention the self-talk aka &quot;stream of consciousness&quot; that virtually everyone hears. And a key insight is that it&#x27;s not <i>our</i> consciousness. It&#x27;s basically just recordings that get played back on queue. A central aspect of meditation is becoming aware of that voice as something distinct from self.<p>So anyway, it&#x27;s best to think &quot;Thanks for sharing :)&quot; and then to do something intentional.</text></item></parent_chain><comment><author>topmonk</author><text>You might be interested in this (start on page 6): <a href="https:&#x2F;&#x2F;www.scribd.com&#x2F;document&#x2F;326903960&#x2F;Feynman-Thinking" rel="nofollow">https:&#x2F;&#x2F;www.scribd.com&#x2F;document&#x2F;326903960&#x2F;Feynman-Thinking</a><p>According to Feyman&#x27;s experiment, above, it turns out there are at least two kinds of people, those that can read while counting in their head, and those that can speak.<p>I always wondered why empirically derived differences in people, such as the above, was never used much in other studies. It seems that it would be a lot more interesting data point than the typical self-evaluation questionnaire that most psychology studies seem to use.</text></comment> |
9,511,446 | 9,511,218 | 1 | 3 | 9,510,614 | train | <story><title>Django Community Survey</title><url>https://www.djangoproject.com/weblog/2015/may/07/community-survey/</url><text></text></story><parent_chain><item><author>iyn</author><text>Can you elaborate on the solutions in frontend?<p>Currently I&#x27;m working on a project and I want to use React &quot;the right way&quot;. While it&#x27;s not so hard to just integrate React into Django project, it&#x27;s not so easy to manage the chaos when things start to grow. At the moment I&#x27;m trying to integrate React, webpack, gulp and node (server side rendering for react) into Django workflow and make it work &quot;natural&quot;, but right now it just feels hacky and not the way I want it to work.<p>Would be great if you could share your workflows &amp; ideas.</text></item><item><author>simonpantzare</author><text>As a long time Django user (first project on 1.1) I want to thank for a great framework and your wonderful release approach!<p>I often try out new new web dev platforms and frameworks but tend to rewrite projects that get too serious or large in Django, or when I&#x27;m starting to miss too many features.<p>If I know that I am building something that others are going to work on, or if I know that the project will need to be maintained for a considerable amount of time, I tend to go with Django and a few familiar libraries (Celery, Django REST framework, django-extensions) right off the bat, as I feel comfortable that we won&#x27;t run into issues where some important dependency becomes unmaintained or introduces backwards-incompatible changes. Much thanks to Django&#x27;s predictable and well thought out releases.<p>Further, when I pick Django, I know that there is good support in monitoring and crash report systems such as Sentry, New Relic, and others. So that&#x27;s one less thing to worry about.<p>The only area where I have started to use other solutions is frontend dev, as I find that the momentum in the Node community is so much higher in this area. I only wish that the Node community wasn&#x27;t as fragmented and had a more long-term view on releases.</text></item></parent_chain><comment><author>simonpantzare</author><text>I put nginx in front and a location &#x2F;static&#x2F; block with a try_files directive that first tries to serve static files from Django (those collected with python manage.py collectstatic) and falls back to serve files that a gulp task generated.<p>To serve different assets in dev and prod I let gulp generate a manifest and have a custom template tag in Django that expands to different paths depending on settings.DEBUG:<p><pre><code> {% my_static &#x27;css&#x2F;all.css&#x27; %}
</code></pre>
expands to either &#x2F;static&#x2F;css&#x2F;all.css or something like &#x2F;static&#x2F;css&#x2F;all.min.abcd1234.css.<p>If nginx is serving a minified, hashed path it adds a Cache-Control: max-age=31556926 header to the response.</text></comment> | <story><title>Django Community Survey</title><url>https://www.djangoproject.com/weblog/2015/may/07/community-survey/</url><text></text></story><parent_chain><item><author>iyn</author><text>Can you elaborate on the solutions in frontend?<p>Currently I&#x27;m working on a project and I want to use React &quot;the right way&quot;. While it&#x27;s not so hard to just integrate React into Django project, it&#x27;s not so easy to manage the chaos when things start to grow. At the moment I&#x27;m trying to integrate React, webpack, gulp and node (server side rendering for react) into Django workflow and make it work &quot;natural&quot;, but right now it just feels hacky and not the way I want it to work.<p>Would be great if you could share your workflows &amp; ideas.</text></item><item><author>simonpantzare</author><text>As a long time Django user (first project on 1.1) I want to thank for a great framework and your wonderful release approach!<p>I often try out new new web dev platforms and frameworks but tend to rewrite projects that get too serious or large in Django, or when I&#x27;m starting to miss too many features.<p>If I know that I am building something that others are going to work on, or if I know that the project will need to be maintained for a considerable amount of time, I tend to go with Django and a few familiar libraries (Celery, Django REST framework, django-extensions) right off the bat, as I feel comfortable that we won&#x27;t run into issues where some important dependency becomes unmaintained or introduces backwards-incompatible changes. Much thanks to Django&#x27;s predictable and well thought out releases.<p>Further, when I pick Django, I know that there is good support in monitoring and crash report systems such as Sentry, New Relic, and others. So that&#x27;s one less thing to worry about.<p>The only area where I have started to use other solutions is frontend dev, as I find that the momentum in the Node community is so much higher in this area. I only wish that the Node community wasn&#x27;t as fragmented and had a more long-term view on releases.</text></item></parent_chain><comment><author>anewhnaccount</author><text>I&#x27;ve not done it myself (I&#x27;ve looked but it was a small project so I just ignored server side rendering and used django-compressor for convienience&#x27;s sake) but you should definitely check <a href="https:&#x2F;&#x2F;github.com&#x2F;markfinger&#x2F;python-react" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;markfinger&#x2F;python-react</a> if you haven&#x27;t already</text></comment> |
14,508,191 | 14,508,228 | 1 | 2 | 14,507,917 | train | <story><title>Uber executive, who obtained medical records of customer rape victim, is fired</title><url>https://www.recode.net/2017/6/7/15754316/uber-executive-india-assault-rape-medical-records</url></story><parent_chain></parent_chain><comment><author>kevinburke</author><text>To summarize: An Uber rider in India alleged she was raped by an Uber driver. The driver was already awaiting trial for four other criminal offenses, and was ultimately convicted of the crime. An Uber executive (Eric Alexander) flew to India and (legally or illegally) obtained her medical records, then showed these to multiple executives at the company.<p>These records were then used by Alexander, Travis Kalanick, and Emil Michael to dispute (internally) the woman&#x27;s account of the story. They claimed that a local competitor was trying to sabotage them by planting this story, which again, involves an Uber customer being raped by a driver facing multiple (other) criminal accusations. Needless to say they lack the training necessary to decipher a medical record, a rape may or may not have left evidence present in that record, and it&#x27;s a gross violation of customer privacy, business ethics, and (probably) the law.<p>This is repulsive conduct and it&#x27;s unbelievable any of these people are still attached to the company.</text></comment> | <story><title>Uber executive, who obtained medical records of customer rape victim, is fired</title><url>https://www.recode.net/2017/6/7/15754316/uber-executive-india-assault-rape-medical-records</url></story><parent_chain></parent_chain><comment><author>kafkaesq</author><text><i>The executive in question, Eric Alexander, the president of business in the Asia Pacific, then showed the medical records to Uber CEO Travis Kalanick and SVP Emil Michael.</i><p>At least the article makes it clear who needs to be fired next -- for both not immediately firing Alexander for gross misconduct, and (just as egregiously) not adamantly refusing to even consider looking at a rape victim&#x27;s private medicate records in the first place.<p>Nevermind the legal and criminal implications of that latter misstep; this is a matter of pure and simple human dignity.</text></comment> |
40,432,964 | 40,433,307 | 1 | 2 | 40,432,446 | train | <story><title>iTerm2 and AI Hype Overload</title><url>https://xeiaso.net/notes/2024/ai-hype/</url></story><parent_chain><item><author>eriri</author><text>I looked into the issue tracker and gosh, its getting so toxic there.<p><a href="https:&#x2F;&#x2F;gitlab.com&#x2F;gnachman&#x2F;iterm2&#x2F;-&#x2F;issues&#x2F;11470" rel="nofollow">https:&#x2F;&#x2F;gitlab.com&#x2F;gnachman&#x2F;iterm2&#x2F;-&#x2F;issues&#x2F;11470</a></text></item></parent_chain><comment><author>hiatus</author><text>With gems like this:<p>&gt; +1 to the people who&#x27;d like to donate 50$ for a version which does not send my input anywhere. Okay, previously I never donated the project. But use the iTerm2 for ~10yrs and would like to continue.<p>So, this person has been using iTerm2 for 10 years without paying, and would only consider donating if this feature is removed?</text></comment> | <story><title>iTerm2 and AI Hype Overload</title><url>https://xeiaso.net/notes/2024/ai-hype/</url></story><parent_chain><item><author>eriri</author><text>I looked into the issue tracker and gosh, its getting so toxic there.<p><a href="https:&#x2F;&#x2F;gitlab.com&#x2F;gnachman&#x2F;iterm2&#x2F;-&#x2F;issues&#x2F;11470" rel="nofollow">https:&#x2F;&#x2F;gitlab.com&#x2F;gnachman&#x2F;iterm2&#x2F;-&#x2F;issues&#x2F;11470</a></text></item></parent_chain><comment><author>mrozbarry</author><text>I think this issue is actually two issues<p><pre><code> 1. A terminal shouldn&#x27;t be able to ask some resource on the internet what to type and auto-execute it.
2. AI fear&#x2F;fatigue&#x2F;???
</code></pre>
I think point 1 is reasonable to an extent, but it should be taken in context. iTerm2 is a free app, and as far as I can tell, not even remotely required on any mac platform, since there is technically a default dumb terminal, which can be customized. I think the context issue is from the video demos I&#x27;ve seen, nothing directly types into your terminal, it&#x27;s up to the user to review&#x2F;copy&#x2F;paste the generated code snippet. The underlying tech has been in iTerm for a while, from the best I can see. Auto-fill also enables things like the 1password integration, and anyone can open a chatgpt client and copy&#x2F;paste shell code from there in the same way the iTerm2 integration works.<p>I understand point 2, I have never cared for any AI hype, it has near-zero interest for me, and doesn&#x27;t affect my work. Almost every editor has some capacity to ask the internet for data and paste it in, from AI or otherwise, and no one is really sounding a major alarm bell around that. You could argue there is a big push for these integrations to train models, but even that requires a key.</text></comment> |
27,962,403 | 27,962,499 | 1 | 2 | 27,962,195 | train | <story><title>The Economic Consequences of Major Tax Cuts for the Rich [pdf]</title><url>http://eprints.lse.ac.uk/107919/1/Hope_economic_consequences_of_major_tax_cuts_published.pdf</url></story><parent_chain><item><author>mhoad</author><text>“We find that major reforms reducing taxes on the rich lead to higher income inequality as measured by the top 1% share of pre-tax national income. The effect remains stable in the medium term. In contrast, such reforms do not have any significant effect on economic growth and unemployment.”</text></item></parent_chain><comment><author>mushufasa</author><text>This has been the consensus within academic econometrics for ~15-20 years, FYI, going back to the Laffer Curve argument. Econometrics != economics; you will still find economics people, especially at think tanks, who will hawk the &#x27;tax cuts solve all problems&#x27; theory (typically associated with think tanks funded by wealth people).</text></comment> | <story><title>The Economic Consequences of Major Tax Cuts for the Rich [pdf]</title><url>http://eprints.lse.ac.uk/107919/1/Hope_economic_consequences_of_major_tax_cuts_published.pdf</url></story><parent_chain><item><author>mhoad</author><text>“We find that major reforms reducing taxes on the rich lead to higher income inequality as measured by the top 1% share of pre-tax national income. The effect remains stable in the medium term. In contrast, such reforms do not have any significant effect on economic growth and unemployment.”</text></item></parent_chain><comment><author>ramesh31</author><text>&gt;We find that major reforms reducing taxes on the rich lead to higher income inequality as measured by the top 1% share of pre-tax national income.<p>This isn&#x27;t surprising; a tautology almost. But it doesn&#x27;t really give us any answers. The real question is: <i>is that a bad thing, and if so how bad is it?</i>. Wealth inequality has been a constant of the human condition since the first farmers learned to aggregate more grain than others. But what, if any, deleterious effect does this actually have on society? And does the concentration of wealth lead to innovations only possible with such an ability to deploy capital in a highly targeted manner?</text></comment> |
20,780,865 | 20,781,083 | 1 | 3 | 20,778,926 | train | <story><title>Show HN: SponsorBlock – Skip sponsorship segments of YouTube videos</title><url>http://sponsor.ajay.app</url></story><parent_chain></parent_chain><comment><author>natpalmer1776</author><text>I don&#x27;t agree with this at all.<p>Sponsored segments are typically produced by, or use the same production values as, the creator who is providing the content you intended to consume. The sponsored segments are an accurate reflection of the content you&#x27;re consuming and typically do not detract from the overall experience.<p>Like the normal web ads we see, they are intended to pay the bills of the individual(s) or company that is providing this content to us for free, so they may continue doing so.<p>By blocking the sponsorships, you&#x27;re denying them the right to make profit from their work in a manner that doesn&#x27;t execute unexpected code, intrusively collect your information, or otherwise violate mutual trust.</text></comment> | <story><title>Show HN: SponsorBlock – Skip sponsorship segments of YouTube videos</title><url>http://sponsor.ajay.app</url></story><parent_chain></parent_chain><comment><author>jasode</author><text>Rather than use a browser plugin, I&#x27;d prefer that Youtubers use the built-in metadata tool to mark their video as <i>&quot;paid promotion&quot;</i>: <a href="https:&#x2F;&#x2F;support.google.com&#x2F;youtube&#x2F;answer&#x2F;154235" rel="nofollow">https:&#x2F;&#x2F;support.google.com&#x2F;youtube&#x2F;answer&#x2F;154235</a><p>A lot of Youtubers aren&#x27;t marking their videos with that flag. (Many may be unaware of it.)<p>A screenshot showing what a &quot;<i>Includes paid promotion&quot;</i> title card looks like: <a href="https:&#x2F;&#x2F;wersm.com&#x2F;youtube-creators-includes-paid-promotion-disclaimer-videos&#x2F;" rel="nofollow">https:&#x2F;&#x2F;wersm.com&#x2F;youtube-creators-includes-paid-promotion-d...</a><p>An example video with the <i>&quot;Includes paid promotion&quot;</i> displayed at the beginning: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=o4ijHIRaGXc" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=o4ijHIRaGXc</a><p>As an aside, I don&#x27;t mind creators sponsored with relevant ads. They often show products that interest me. (This doesn&#x27;t include generic ads like Blue Apron and Nord VPN that seem to be attached to any topic.)</text></comment> |
38,615,892 | 38,614,038 | 1 | 3 | 38,612,585 | train | <story><title>NASA says SpaceX’s next Starship flight could test refueling tech</title><url>https://arstechnica.com/space/2023/12/nasa-wants-to-see-gas-stations-in-space-but-so-far-its-tanks-are-empty/</url></story><parent_chain><item><author>thebigman433</author><text>This is pretty interesting, but Id be surprised if it actually happens on the next flight. Depending on the cost of the tanks&#x2F;fittings&#x2F;machines, it could be a fairly expensive loss if Starship doesnt reach orbit, which seems fairly likely still. Their last launch was impressive, but Id be wary to put a payload worth more than a few pennies on the next launch</text></item></parent_chain><comment><author>spikels</author><text>Faulty assumptions and logic.<p>The “payload” is an experiment that is at most two tanks, a valve and monitoring equipment. However it is pretty clear they will use the existing LOX main and header tanks and valves already present in every Starship. This will meet the requirements of NASA’s 2020 $53 million challenge to “transfer 10 metric tons of cryogenic propellant, specifically liquid oxygen, between tanks on a Starship vehicle”[1] as the smaller header tank holds roughly 20 tons.<p>If you are already attempting to get into orbit it makes sense to include this experiment in case you succeed because you save the time and cost of an entire launch and might get a $53 million award while advancing a key requirement for HLS, a $2.9 billion dollar contract and the current core mission.<p>This experiment has been planned for years, Starship was designed to do it and it makes sense to do it now. SpaceX clearly knows what they are doing. Unfortunately HN is terrible on this subject.<p>[1] <a href="https:&#x2F;&#x2F;www.nasa.gov&#x2F;technology&#x2F;2020-nasa-tipping-point-selections&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.nasa.gov&#x2F;technology&#x2F;2020-nasa-tipping-point-sele...</a></text></comment> | <story><title>NASA says SpaceX’s next Starship flight could test refueling tech</title><url>https://arstechnica.com/space/2023/12/nasa-wants-to-see-gas-stations-in-space-but-so-far-its-tanks-are-empty/</url></story><parent_chain><item><author>thebigman433</author><text>This is pretty interesting, but Id be surprised if it actually happens on the next flight. Depending on the cost of the tanks&#x2F;fittings&#x2F;machines, it could be a fairly expensive loss if Starship doesnt reach orbit, which seems fairly likely still. Their last launch was impressive, but Id be wary to put a payload worth more than a few pennies on the next launch</text></item></parent_chain><comment><author>BryanLegend</author><text>It&#x27;s fantastic that the Super Heavy booster fully completed it&#x27;s mission! Much bigger deal than the ship itself failing.<p>Before that last flight it seemed like engine reliability wouldn&#x27;t get solved. But now they have a clear path to orbit.<p>I doubt the heat shields will hold up on the first several re-entry attempts but there&#x27;s been recent sightings of a new heat shield that has tiles about half the size. Maybe smaller tiles will have less lever action and will have 4x the fasteners.</text></comment> |
2,171,770 | 2,171,845 | 1 | 2 | 2,171,585 | train | <story><title>The 4-Hour Dentist</title><url>http://bygonebureau.com/2011/02/02/4-hour-dentist/</url></story><parent_chain></parent_chain><comment><author>ryanwaggoner</author><text>Huzzah! A straw man attack!<p>Am I seriously the only person on HN who has gotten useful, effective stuff from both of Tim's books? Does everyone just hate him because he's such an effective promoter that they assume he must be selling snake oil?<p>4HWW led me to quit my job and start freelancing, spend three months traveling through Asia, build some online passive income streams, and generally have a healthier respect towards my time and how I spend it. The 4 Hour Body has led me to completely change my diet over the last 6-8 weeks and I feel incredible, better than I have in a long time.<p>I know I'm not the only one, but am I the only one on HN?</text></comment> | <story><title>The 4-Hour Dentist</title><url>http://bygonebureau.com/2011/02/02/4-hour-dentist/</url></story><parent_chain></parent_chain><comment><author>joshkaufman</author><text>Much of what Tim Ferriss writes is genuinely useful, particularly when it comes to experimentation, testing, and questioning assumptions.<p>Unfortunately, these messages are (1) packaged in the form of breathlessly-sold, instant-results snake oil; (2) explicitly modeled on the past 20 years of <i>Men's Health</i> and <i>Cosmopolitan</i> magazine covers; (3) contain obnoxious levels of social signaling and influence-via-association.<p>The reality is the above three factors are what have contributed the most to his renown. People want to instantly become wealthy, famous, and desirable with zero effort. They want to believe that someone (anyone) can provide these things, and they're willing to pay money on the off-chance they're able to deliver. They want to be associated with someone who travels around the world, hangs out with famous people all day, and seemingly succeeds in everything.<p>Ferriss promises people exactly what they want, so they buy. Simple as that.<p>The operative question for all of us on HN: what Ferriss does clearly works. Do you choose to emulate him? Why or why not?</text></comment> |
32,706,466 | 32,704,674 | 1 | 2 | 32,692,012 | train | <story><title>Roaring bitmaps: what they are and how they work</title><url>https://vikramoberoi.com/a-primer-on-roaring-bitmaps-what-they-are-and-how-they-work/</url></story><parent_chain><item><author>celeritascelery</author><text>I see this common theme among very fast practical algorithms (like timsort or pdqsort) where there is not some secret math or algorithm, rather they are just a bunch of special cases based on heuristics. Often they involve specific knowledge of the hardware instead of treating software as abstract. To me this is the big difference between “computer science” and “computer engineering”.</text></item></parent_chain><comment><author>jandrewrogers</author><text>Much of this is just threshold effects for algorithm performance, not special cases per se.<p>You don&#x27;t even need to have specific knowledge of the hardware as long as you can identify cross-over points where one algorithm starts significantly outperforming others. An old HPC trick is to write software that thoroughly measures several algorithm strategies on your specific hardware environment and then code-gens an algorithm that has an optimal set of strategies and strategy-switching thresholds. The meta-algorithm is mostly focused on cheaply detecting these thresholds at runtime.</text></comment> | <story><title>Roaring bitmaps: what they are and how they work</title><url>https://vikramoberoi.com/a-primer-on-roaring-bitmaps-what-they-are-and-how-they-work/</url></story><parent_chain><item><author>celeritascelery</author><text>I see this common theme among very fast practical algorithms (like timsort or pdqsort) where there is not some secret math or algorithm, rather they are just a bunch of special cases based on heuristics. Often they involve specific knowledge of the hardware instead of treating software as abstract. To me this is the big difference between “computer science” and “computer engineering”.</text></item></parent_chain><comment><author>voberoi</author><text>I had the same takeaway when I read these papers. It felt like a game of whack-a-mole.<p>There&#x27;s a massive performance benefit to doing this at the cost of implementation complexity. I haven&#x27;t studied the implementations or tried my hand one, but I get the impression that these are tough to implement correctly in a way that takes full advantage of the hardware.<p>(In that sense, it&#x27;s awesome that the researchers also did the legwork to implement and maintain a library!)</text></comment> |
11,450,687 | 11,450,046 | 1 | 2 | 11,449,767 | train | <story><title>Why Erlang Matters</title><url>https://sameroom.io/blog/why-erlang-matters</url></story><parent_chain></parent_chain><comment><author>atemerev</author><text>AFAIK, Erlang is still (as of 2016) the only distributed actor model implementation with preemptive scheduler.<p>All other major implementations (including Akka) have cooperative scheduling, i.e. forbidding blocking code in actors. Erlang allows it. This is huge.<p>And actor supervision is the best way to write reliable systems. I have wrote some code in Akka without much effort and testing (streaming market data aggregation), and it is still running with a few years uptime.</text></comment> | <story><title>Why Erlang Matters</title><url>https://sameroom.io/blog/why-erlang-matters</url></story><parent_chain></parent_chain><comment><author>jandrese</author><text>I&#x27;m a very green Erlang noob, but given what I have seen from it I find articles like this kind of strange. Sure concurrent programming is difficult and we need to think hard about how to make programs run quickly in a multiprocessor environment, but the fundamental architecture of Erlang seems to be in conflict with big data and high speed computing. It seems like a language that can scale much better, but has such enormous constant time penalties that the scaling can&#x27;t overcome the hurdle until you&#x27;re talking about thousands of processors.<p>Every single state change requires a function call, every function call involves a pattern matching algorithm that has to account for all of your program arguments (effectively almost the entire state of your thread!) and isn&#x27;t even strongly typed. And then there is synchronization and data sharing between threads, which seems a bit handwavy and effectively requires a database running in your program that the threads can poll.<p>If your problem set is lots and lots of small independent tasks that don&#x27;t have to finish overly quickly, then Erlang is fantastic. Stuff like switching voice circuits for example. But I&#x27;m trying to imagine manipulating a 1TB dataset with thousands of worker threads if you have to make a copy on the stack for every change every worker makes.<p>I know people do some big data stuff with Erlang, so these have to be solved problems somehow, but I can&#x27;t help but to suspect that they have to compromise some of the ideals espoused in this article to make it work.</text></comment> |
18,552,230 | 18,552,382 | 1 | 2 | 18,549,361 | train | <story><title>You Only Need 50% of Job “Requirements”</title><url>https://talent.works/blog/2018/11/27/the-science-of-the-job-search-part-vii-you-only-need-50-of-job-requirements/</url></story><parent_chain><item><author>rossdavidh</author><text>I think many people write their job description&#x27;s requirements, as if it were the opening stage of a negotiation. You list everything you want as if it were a requirement, and you will only get about half of that. But the problem is, it isn&#x27;t a negotiation, it&#x27;s a minimum spec (especially if you put the word &quot;required&quot; next to that line), and so many quite good candidates won&#x27;t apply.<p>However, typically, I see 20-30 requirements, many of which (e.g. &quot;passionate about software development&quot;, &quot;good communicator&quot;) aren&#x27;t specific enough to tell the candidate whether they are a good match or not (do poor communicators know they are poor communicators?). Of the rest, really only 2 or 3 are actual requirements, and the candidate has to guess which those are.</text></item></parent_chain><comment><author>SilasX</author><text>&gt;I think many people write their job description&#x27;s requirements, as if it were the opening stage of a negotiation. You list everything you want as if it were a requirement, and you will only get about half of that. But the problem is, it isn&#x27;t a negotiation, it&#x27;s a minimum spec (especially if you put the word &quot;required&quot; next to that line), and so many quite good candidates won&#x27;t apply.<p>They have the option to distinguish &quot;minimum&#x2F;required&quot; vs &quot;nice to have&quot;, though.</text></comment> | <story><title>You Only Need 50% of Job “Requirements”</title><url>https://talent.works/blog/2018/11/27/the-science-of-the-job-search-part-vii-you-only-need-50-of-job-requirements/</url></story><parent_chain><item><author>rossdavidh</author><text>I think many people write their job description&#x27;s requirements, as if it were the opening stage of a negotiation. You list everything you want as if it were a requirement, and you will only get about half of that. But the problem is, it isn&#x27;t a negotiation, it&#x27;s a minimum spec (especially if you put the word &quot;required&quot; next to that line), and so many quite good candidates won&#x27;t apply.<p>However, typically, I see 20-30 requirements, many of which (e.g. &quot;passionate about software development&quot;, &quot;good communicator&quot;) aren&#x27;t specific enough to tell the candidate whether they are a good match or not (do poor communicators know they are poor communicators?). Of the rest, really only 2 or 3 are actual requirements, and the candidate has to guess which those are.</text></item></parent_chain><comment><author>mbesto</author><text>&gt; I think many people write their job description&#x27;s requirements, as if it were the opening stage of a negotiation.<p>As someone who has written and hired technical talent - this is absolutely true.<p>&gt; so many quite good candidates won&#x27;t apply.<p>I agree with your sentiment but generally disagree in practice. The reality is I get 100&#x27;s of applications that are nowhere near qualified for positions I&#x27;ve listed. We&#x27;re talking like recent grads applying for a VP of E-commerce (~10 years exp on the spec).</text></comment> |
12,964,038 | 12,959,808 | 1 | 2 | 12,959,313 | train | <story><title>HP Unveils Mini Workstation</title><url>http://www8.hp.com/us/en/hp-news/press-release.html?id=2364757</url></story><parent_chain><item><author>ChuckMcM</author><text>See also the Intel NUC thing: <a href="http:&#x2F;&#x2F;www.intel.com&#x2F;content&#x2F;www&#x2F;us&#x2F;en&#x2F;nuc&#x2F;nuc-kit-nuc6i7kyk-features-configurations.html" rel="nofollow">http:&#x2F;&#x2F;www.intel.com&#x2F;content&#x2F;www&#x2F;us&#x2F;en&#x2F;nuc&#x2F;nuc-kit-nuc6i7kyk...</a><p>What really astonishes me though is how you can have these really nice machines and Apple doesn&#x27;t bother to update the Mac mini. Seriously, an Apple version of this would be a nice desktop.</text></item></parent_chain><comment><author>orangecat</author><text><i>What really astonishes me though is how you can have these really nice machines and Apple doesn&#x27;t bother to update the Mac mini.</i><p>It would almost have been better if they didn&#x27;t update it. The 2012 mini was fantastic; you could take it to a quad i7 and 16GB for under $1000 thanks to user-accessible RAM. For most purposes that&#x27;s much better than the current $1000 mini.</text></comment> | <story><title>HP Unveils Mini Workstation</title><url>http://www8.hp.com/us/en/hp-news/press-release.html?id=2364757</url></story><parent_chain><item><author>ChuckMcM</author><text>See also the Intel NUC thing: <a href="http:&#x2F;&#x2F;www.intel.com&#x2F;content&#x2F;www&#x2F;us&#x2F;en&#x2F;nuc&#x2F;nuc-kit-nuc6i7kyk-features-configurations.html" rel="nofollow">http:&#x2F;&#x2F;www.intel.com&#x2F;content&#x2F;www&#x2F;us&#x2F;en&#x2F;nuc&#x2F;nuc-kit-nuc6i7kyk...</a><p>What really astonishes me though is how you can have these really nice machines and Apple doesn&#x27;t bother to update the Mac mini. Seriously, an Apple version of this would be a nice desktop.</text></item></parent_chain><comment><author>daxorid</author><text>I have the Skullcanyon NUC, and this is not comparable.<p>I&#x27;ve been waiting for someone to offer me ECC in a small form factor. This is the first time I&#x27;ve ever seen it.</text></comment> |
19,203,111 | 19,202,472 | 1 | 2 | 19,199,647 | train | <story><title>Code Shelter: A Maintainer Community for Abandoned FOSS Projects</title><url>https://www.codeshelter.co/</url></story><parent_chain></parent_chain><comment><author>entelechy</author><text>This is amazing!<p>As one of the creators of a distributed package manager for C++ and friends [1] we made a funny discovery:<p>Many C libraries that a big chunk of the ecosystem depends on, have not been updated for many years. Some of those can only be downloaded from sourceforge or ftp server.<p>Even worse, some libraries are copy and pasted from project to project and have no actual home.<p>We uploaded them to github and started maintaining them.<p>If you know any abandoned C&#x2F;C++ projects or C&#x2F;C++ projects you need a hand in maintaining, we are happy to help.<p>[1] <a href="https:&#x2F;&#x2F;github.com&#x2F;loopperfect&#x2F;buckaroo" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;loopperfect&#x2F;buckaroo</a></text></comment> | <story><title>Code Shelter: A Maintainer Community for Abandoned FOSS Projects</title><url>https://www.codeshelter.co/</url></story><parent_chain></parent_chain><comment><author>edzillion</author><text>You should set up an opencollective.com account (like Patreon for FOSS projects) - I would support this project and I am sure there are others that would too.</text></comment> |
8,898,625 | 8,898,650 | 1 | 2 | 8,898,299 | train | <story><title>Rules for Developing Safety Critical Code [pdf]</title><url>http://pixelscommander.com/wp-content/uploads/2014/12/P10.pdf</url></story><parent_chain><item><author>tgb</author><text>The most impressive thing I&#x27;ve read about NASA&#x27;s software development isn&#x27;t mentioned here. It was the system where bugs and defects were (at least in principle) not treated as the failing of an individual but rather as a failing of the entire development process. So when a problem is found, the question isn&#x27;t &quot;Who did this and how can we punish them?&quot; it&#x27;s &quot;What can we as a team do so that this never happens again?&quot;<p>I don&#x27;t know how this works in practice, but it sounds like a great system even when dealing with code in a lower-risk setting.</text></item></parent_chain><comment><author>thearn4</author><text>As a software person at NASA, one caveat to note is that this is in part due to the intricacies of the federal civil servant system. Frankly, it is very hard to punish or fire government employees or some prime contractors, so most often (even in non-software contexts) accountability is applied to a process rather than to an individual because it&#x27;s a path of less resistance for management. Even when an individual approach would be more appropriate.</text></comment> | <story><title>Rules for Developing Safety Critical Code [pdf]</title><url>http://pixelscommander.com/wp-content/uploads/2014/12/P10.pdf</url></story><parent_chain><item><author>tgb</author><text>The most impressive thing I&#x27;ve read about NASA&#x27;s software development isn&#x27;t mentioned here. It was the system where bugs and defects were (at least in principle) not treated as the failing of an individual but rather as a failing of the entire development process. So when a problem is found, the question isn&#x27;t &quot;Who did this and how can we punish them?&quot; it&#x27;s &quot;What can we as a team do so that this never happens again?&quot;<p>I don&#x27;t know how this works in practice, but it sounds like a great system even when dealing with code in a lower-risk setting.</text></item></parent_chain><comment><author>UK-AL</author><text>This is the standard in manufacturing techniques. A failure in quality control is considered failure of the system for allowing it happen in the first place, not the person making the mistake.<p>See:
<a href="http://en.wikipedia.org/wiki/Poka-yoke" rel="nofollow">http:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Poka-yoke</a><p>If a bug got through, and you don&#x27;t test. Then obviously its because you don&#x27;t test. Not because the programmer isn&#x27;t superhuman, and doesn&#x27;t make mistakes.<p>Everybody makes mistakes, design the system to handle it.</text></comment> |
22,124,113 | 22,124,057 | 1 | 3 | 22,119,962 | train | <story><title>250M Microsoft customer service and support records exposed on the web</title><url>https://www.comparitech.com/blog/information-security/microsoft-customer-service-data-leak/</url></story><parent_chain><item><author>deepspace</author><text>I truly believe the era of privacy is over.<p>There are zero consequences to anyone important when a data breach happens; therefore there is no incentive for companies to protect their user data and the number of breaches will continue to grow for the foreseeable future.<p>By 2025 at the latest, there will effectively be no such thing as privacy anymore. All personal data belonging to everyone will have been exfiltrated and will be available for sale.<p>You will be able to purchase the full medical history, all financial transactions, all addresses, phone numbers, location history (often with photo&#x2F;video evidence), all account numbers, ID numbers (government and otherwise), biometric data, browser&#x2F;search history, every email, sms or chat message they ever sent or received and any other information you can think of for the vast majority of people on earth.</text></item></parent_chain><comment><author>josephg</author><text>I&#x27;ve been long arguing that there should be severe legal consequences for companies who leak data. Right now there&#x27;s almost no legal repercussions for this sort of thing. I expect better from Microsoft, but I really don&#x27;t expect any better from the thousands of tiny startups out there. Unless the people involved suffer any serious consequences for leaking their customer&#x27;s data, they won&#x27;t bother spending the time and money to do a good job and these breaches will continue to be commonplace.<p>As much as I love the free market, we&#x27;re completely failing to protect consumers. I think we need the government to step in and align incentives. I don&#x27;t know if we need engineers to be personally liable in the case of data breaches, but I&#x27;m serious enough about this that I wouldn&#x27;t take it off the table. Medicine has malpractice suits. Engineers have a professional duty of care. Builders have building codes. We need an equivalent for software engineering.<p>Its not the wild west anymore when we didn&#x27;t know how to do this right. For almost all modern software, best practices are out there and well known. The way you secure a password database hasn&#x27;t changed much in the last decade. Apparently people just don&#x27;t care enough to learn and apply those techniques. Bootcamps don&#x27;t even bother to teach any security practices. Given how much the world relies on our industry&#x27;s ability and knowledge, that needs to change in a hurry.</text></comment> | <story><title>250M Microsoft customer service and support records exposed on the web</title><url>https://www.comparitech.com/blog/information-security/microsoft-customer-service-data-leak/</url></story><parent_chain><item><author>deepspace</author><text>I truly believe the era of privacy is over.<p>There are zero consequences to anyone important when a data breach happens; therefore there is no incentive for companies to protect their user data and the number of breaches will continue to grow for the foreseeable future.<p>By 2025 at the latest, there will effectively be no such thing as privacy anymore. All personal data belonging to everyone will have been exfiltrated and will be available for sale.<p>You will be able to purchase the full medical history, all financial transactions, all addresses, phone numbers, location history (often with photo&#x2F;video evidence), all account numbers, ID numbers (government and otherwise), biometric data, browser&#x2F;search history, every email, sms or chat message they ever sent or received and any other information you can think of for the vast majority of people on earth.</text></item></parent_chain><comment><author>mathdev</author><text>I hope not. Much too late, I recently realised that I probably would not survive my WhatsApp history getting out.</text></comment> |
24,977,709 | 24,975,668 | 1 | 3 | 24,958,157 | train | <story><title>Is it better to plant trees or let forests regrow naturally?</title><url>https://www.wired.com/story/is-it-better-to-plant-trees-or-let-forests-regrow-naturally</url></story><parent_chain><item><author>chris_va</author><text>(disclaimer, I am part of a Climate&amp;Energy R&amp;D lab)<p>The goal is not really to produce a set of big trees, but to maximize the amount of carbon you can suck out of the atmosphere per year, heavily weighted to the near future. As a result, producing an old growth forest is not the correct metric (if you are optimizing purely for climate).<p>A stable forest (from purely a climate perspective) is also not the right optimization. Ideally, you&#x27;d lock away the carbon outside of the biosphere, instead of letting it get recycled.</text></item><item><author>JoeAltmaier</author><text>A healthy old-growth forest is maybe 100 trees per acre.<p>A &#x27;naturally regrowing&#x27; forest may be 400 trees per acre. They&#x27;re scrub trees, none of the ever getting very large, unhealthy and scraggly and broken by every storm.<p>It takes maybe a century of trees dying, giving one of their cohort a chance to get ahead of the others, then shading them and &#x27;thinning the herd&#x27; until you have a stable population. Then, they&#x27;re still not very healthy because they &#x27;grew from scrub&#x27;.<p>They have to then die of natural causes one at a time and be replaced by a single healthy one. Maybe another 50 years. Finally you have a stable healthy forest.<p>Or, you can just plant 100 trees per acre in a grid. And mow the volunteers until you have what you want. 20 years maybe.</text></item></parent_chain><comment><author>stinos</author><text>Doesn&#x27;t hurt explicitly mentioning the biodiversity crisis is, just like the climate one, also a pressing issue. And when it comes to solutions revolving around forests (or a collection of trees if you wish), it&#x27;s the area where both clash: optimizing for one will be less ideal for the other. Personally I feel like in such cases it would be wise to strive for a solution which is sliglty in favor of biodiversity, just because for the climate there are a number of other solutions which have a larger potential effect.</text></comment> | <story><title>Is it better to plant trees or let forests regrow naturally?</title><url>https://www.wired.com/story/is-it-better-to-plant-trees-or-let-forests-regrow-naturally</url></story><parent_chain><item><author>chris_va</author><text>(disclaimer, I am part of a Climate&amp;Energy R&amp;D lab)<p>The goal is not really to produce a set of big trees, but to maximize the amount of carbon you can suck out of the atmosphere per year, heavily weighted to the near future. As a result, producing an old growth forest is not the correct metric (if you are optimizing purely for climate).<p>A stable forest (from purely a climate perspective) is also not the right optimization. Ideally, you&#x27;d lock away the carbon outside of the biosphere, instead of letting it get recycled.</text></item><item><author>JoeAltmaier</author><text>A healthy old-growth forest is maybe 100 trees per acre.<p>A &#x27;naturally regrowing&#x27; forest may be 400 trees per acre. They&#x27;re scrub trees, none of the ever getting very large, unhealthy and scraggly and broken by every storm.<p>It takes maybe a century of trees dying, giving one of their cohort a chance to get ahead of the others, then shading them and &#x27;thinning the herd&#x27; until you have a stable population. Then, they&#x27;re still not very healthy because they &#x27;grew from scrub&#x27;.<p>They have to then die of natural causes one at a time and be replaced by a single healthy one. Maybe another 50 years. Finally you have a stable healthy forest.<p>Or, you can just plant 100 trees per acre in a grid. And mow the volunteers until you have what you want. 20 years maybe.</text></item></parent_chain><comment><author>Joakal</author><text>Have you considered something different for climate; mass production of climate change resistant (ie heat&#x2F;acid) photosynthetic plankton and releasing to ocean?<p>How does it compare to trees?</text></comment> |
39,922,603 | 39,922,724 | 1 | 3 | 39,917,509 | train | <story><title>Improvements to static analysis in GCC 14</title><url>https://developers.redhat.com/articles/2024/04/03/improvements-static-analysis-gcc-14-compiler</url></story><parent_chain><item><author>quincepie</author><text>To me fanalyzer is one of GCC killer features over clang. It makes programming C much easier by explaining errors. The error messages also began to feel similar to Rust in terms of being developer friendly.</text></item></parent_chain><comment><author>mr_00ff00</author><text>I know Rust (esp on HN) is very hyped for its memory safety and nice abstractions, but I really wonder how much Rust owes its popularity to its error messages.<p>I would say the #1 reason I stop learning a technology is because of frustrating or unclear errors.<p>EDIT: Getting a bit of topic, but I meant more because I love C and would love it more with rust level error messages.</text></comment> | <story><title>Improvements to static analysis in GCC 14</title><url>https://developers.redhat.com/articles/2024/04/03/improvements-static-analysis-gcc-14-compiler</url></story><parent_chain><item><author>quincepie</author><text>To me fanalyzer is one of GCC killer features over clang. It makes programming C much easier by explaining errors. The error messages also began to feel similar to Rust in terms of being developer friendly.</text></item></parent_chain><comment><author>chc4</author><text>I have had the exact opposite experience: clang constantly gives me much better error messages than GCC, implementations of some warnings or errors catch more cases, and clang-tidy is able to do much better static analysis.</text></comment> |
5,300,227 | 5,300,363 | 1 | 2 | 5,299,534 | train | <story><title>Name.com hijacks non-existent subdomains and redirects to their servers</title><url>http://www.destructuring.net/2013/02/28/name-com-is-doing-some-really-sketchy-stuff/</url><text></text></story><parent_chain><item><author>machrider</author><text>My workaround for this was to add a TXT record for *.mydomain.com that just returns a string like "Unused". This seems to stop them from hijacking any subdomains, and it's not an A record so undefined subdomain names do not resolve, just like if you had not defined them in the first place.<p>(Workaround shouldn't be necessary of course, but this kind of bullshit is par for the course with cheap hosting companies.)</text></item></parent_chain><comment><author>ams6110</author><text>Just fyi, mydomain.com is a real domain, example.com is better to use for illustrative purposes.</text></comment> | <story><title>Name.com hijacks non-existent subdomains and redirects to their servers</title><url>http://www.destructuring.net/2013/02/28/name-com-is-doing-some-really-sketchy-stuff/</url><text></text></story><parent_chain><item><author>machrider</author><text>My workaround for this was to add a TXT record for *.mydomain.com that just returns a string like "Unused". This seems to stop them from hijacking any subdomains, and it's not an A record so undefined subdomain names do not resolve, just like if you had not defined them in the first place.<p>(Workaround shouldn't be necessary of course, but this kind of bullshit is par for the course with cheap hosting companies.)</text></item></parent_chain><comment><author>soult</author><text>While it is probably not relevant for must users, there is a subtle difference between "There exists no A record for x.example.com" and "There is no x.example.com".</text></comment> |
21,642,105 | 21,639,997 | 1 | 3 | 21,638,290 | train | <story><title>Psilocybin for major depression granted Breakthrough Therapy by FDA</title><url>https://newatlas.com/science/psilocybin-major-depression-mdd-usona-breakthrough-therapy-fda/</url></story><parent_chain><item><author>WhompingWindows</author><text>Super placebo is a scientifically incorrect description. These drugs suppress activity of the default mode network of the brain, which is a sort of &quot;conductor&quot; of the brain, which chooses which &quot;instruments&quot; to prioritize and how to allocate the train of thought. All of the different parts of your mind are able to sing out more creatively and independently, forging new paths that they are not accustomed to taking -- this is why it&#x27;s great for mental illnesses like depression, anxiety, alcoholism, smoking -- it allows your mind an incredible flexibility.<p>In my personal experiences with psychedelics, these drugs remove mental heuristics, allowing the mind to think unimpeded by usual thought patterns, simply accepting raw data as it comes in. This was striking in sensory information, if you closed your eyes or just stared at something long enough, you observed incredibly different visual input, geometric patterns, greatly strengthened colors (a strawberry was INCREDIBLY beautiful, I almost felt bad eating such a beautiful entity). There was also an incredible inter-disciplinary thought process, I felt the musical, mathematical, computational, logical, natural, all these parts of my mind mixed together wonderfully.<p>LSD was much longer lasting, around 12 hours, whereas psilocybin was around 6 hours. Psilocybin enhanced my connection with nature, the flying bugs, a handful of berries, and the plants all around me caused an effusion of love and connection. Whereas with LSD, I felt an extremely powerful visual stimulation, the changing fall leaves were extremely vibrant and beautiful, it was akin to 12 hour, mentally-clear cannabis high. I didn&#x27;t experience any &quot;flashbacks&quot; per se, there was a positive glow in my mood and thoughts for around 5 days, and the day after LSD I still had lingering effects...I also wonder if I will evermore look at the surface of ponds and lakes differently, they have an incredible dynamism and vitality that I appreciated even more on first trip.<p>For anyone considering psychedelics, do it thoughtfully and methodically. Read the book mentioned in the parent, practice meditation, explore your mind, and really ensure your mindset (&quot;set&quot;) is positive and open. Further, plan to occupy a peaceful, quiet place, hopefully lacking too many strangers&#x2F;social interaction (&quot;setting&quot;). If set and setting are good, you&#x27;re in for a wonderful time.</text></item><item><author>brenden2</author><text>I&#x27;ve been reading a book called &quot;How to Change Your Mind&quot;[1] which contains a collection of history, anecdotes, trip reports, and some of the science behind these types of drugs. The book mostly discusses psilocybin and LSD, but it also touches on some of the other related drugs. If you&#x27;re curious and want to learn more, it&#x27;s worth a read.<p>The one thing I&#x27;ll say is that it seems like these drugs affect the brain in a way that&#x27;s more akin to a super-placebo, rather than being therapeutic on their own. In other words, you would need to use the drugs in combination with therapy to obtain good results.<p>[1]: <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;How_to_Change_Your_Mind" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;How_to_Change_Your_Mind</a></text></item></parent_chain><comment><author>corporate_shi11</author><text>I took a strong psychedelic once, and a high dose at that. I tripped for over 12 hours, and by the end of the trip I was seriously considering changing my life in significant ways. For days after the trip I was considering making these changes. My thought patterns and assumptions had made a complete jump from one train of thought to another.<p>This experience made me realize how much of our outlook on things is dominated by whatever mode of thinking we happen to be in, and that mode is more shaped by custom and habit than anything approaching logic and reason.</text></comment> | <story><title>Psilocybin for major depression granted Breakthrough Therapy by FDA</title><url>https://newatlas.com/science/psilocybin-major-depression-mdd-usona-breakthrough-therapy-fda/</url></story><parent_chain><item><author>WhompingWindows</author><text>Super placebo is a scientifically incorrect description. These drugs suppress activity of the default mode network of the brain, which is a sort of &quot;conductor&quot; of the brain, which chooses which &quot;instruments&quot; to prioritize and how to allocate the train of thought. All of the different parts of your mind are able to sing out more creatively and independently, forging new paths that they are not accustomed to taking -- this is why it&#x27;s great for mental illnesses like depression, anxiety, alcoholism, smoking -- it allows your mind an incredible flexibility.<p>In my personal experiences with psychedelics, these drugs remove mental heuristics, allowing the mind to think unimpeded by usual thought patterns, simply accepting raw data as it comes in. This was striking in sensory information, if you closed your eyes or just stared at something long enough, you observed incredibly different visual input, geometric patterns, greatly strengthened colors (a strawberry was INCREDIBLY beautiful, I almost felt bad eating such a beautiful entity). There was also an incredible inter-disciplinary thought process, I felt the musical, mathematical, computational, logical, natural, all these parts of my mind mixed together wonderfully.<p>LSD was much longer lasting, around 12 hours, whereas psilocybin was around 6 hours. Psilocybin enhanced my connection with nature, the flying bugs, a handful of berries, and the plants all around me caused an effusion of love and connection. Whereas with LSD, I felt an extremely powerful visual stimulation, the changing fall leaves were extremely vibrant and beautiful, it was akin to 12 hour, mentally-clear cannabis high. I didn&#x27;t experience any &quot;flashbacks&quot; per se, there was a positive glow in my mood and thoughts for around 5 days, and the day after LSD I still had lingering effects...I also wonder if I will evermore look at the surface of ponds and lakes differently, they have an incredible dynamism and vitality that I appreciated even more on first trip.<p>For anyone considering psychedelics, do it thoughtfully and methodically. Read the book mentioned in the parent, practice meditation, explore your mind, and really ensure your mindset (&quot;set&quot;) is positive and open. Further, plan to occupy a peaceful, quiet place, hopefully lacking too many strangers&#x2F;social interaction (&quot;setting&quot;). If set and setting are good, you&#x27;re in for a wonderful time.</text></item><item><author>brenden2</author><text>I&#x27;ve been reading a book called &quot;How to Change Your Mind&quot;[1] which contains a collection of history, anecdotes, trip reports, and some of the science behind these types of drugs. The book mostly discusses psilocybin and LSD, but it also touches on some of the other related drugs. If you&#x27;re curious and want to learn more, it&#x27;s worth a read.<p>The one thing I&#x27;ll say is that it seems like these drugs affect the brain in a way that&#x27;s more akin to a super-placebo, rather than being therapeutic on their own. In other words, you would need to use the drugs in combination with therapy to obtain good results.<p>[1]: <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;How_to_Change_Your_Mind" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;How_to_Change_Your_Mind</a></text></item></parent_chain><comment><author>surfsvammel</author><text>To me, the most stunning hallucination, during LSD, was the total confusion of my senses. A specific flower was in bloom at the time, and my mind confused it’s smell with the colour purple. It was the strangest experience. Smelling colour... There is no way to describe it in words.<p>Even though this episode was decades ago, I still, kind of, experience the scent of that flower as the colour purple.</text></comment> |
35,367,092 | 35,366,919 | 1 | 2 | 35,365,440 | train | <story><title>Teen-age fentanyl deaths in a Texas county</title><url>https://www.newyorker.com/news/letter-from-the-southwest/the-horrifying-epidemic-of-teen-age-fentanyl-deaths-in-a-texas-county</url></story><parent_chain><item><author>ChancyChance</author><text>Notice how Barnett has zero empathy when he thought it was just an Austin thing, then did an about face after the subsequent deaths in Kyle.<p>However I still don’t understand the point of lacing drugs with fentanyl. As a drug dealer you make more money with living clients, right?</text></item></parent_chain><comment><author>yetanotherloss</author><text>For the most part it&#x27;s not intentionally lacing drugs, it&#x27;s poor quality control and cross contamination.<p>No one is intentionally mixing fentanyl and MDMA or cocaine but with poor lab work it&#x27;s easy to get a bit of the wrong whitish powder and the people selling it down stream have no more idea than their buyers what&#x27;s really in it.<p>Affluent recreational drug connoisseurs can use basic chemistry and test kits to be highly confident what they&#x27;re taking but even they get bit sometimes if reckless and using a new source.</text></comment> | <story><title>Teen-age fentanyl deaths in a Texas county</title><url>https://www.newyorker.com/news/letter-from-the-southwest/the-horrifying-epidemic-of-teen-age-fentanyl-deaths-in-a-texas-county</url></story><parent_chain><item><author>ChancyChance</author><text>Notice how Barnett has zero empathy when he thought it was just an Austin thing, then did an about face after the subsequent deaths in Kyle.<p>However I still don’t understand the point of lacing drugs with fentanyl. As a drug dealer you make more money with living clients, right?</text></item></parent_chain><comment><author>akiselev</author><text>Fentanyl costs on the order of $1,000 per kilogram to produce and that’s enough to cut hundreds of thousands of doses of another drug while making them significantly more addictive. It makes other drugs so profitable that the loss in customers is well worth it. There’s a future user born every minute, afterall.</text></comment> |
8,160,859 | 8,160,388 | 1 | 2 | 8,160,149 | train | <story><title>Low-cost 3D printers and crowdfunding suicide</title><url>http://3dprototypesandmodels.com.au/blog-2/</url></story><parent_chain></parent_chain><comment><author>ChuckMcM</author><text>This was an interesting article. Primarily about the inexperience in people building printers (and some evil people) who through Kickstarter or Indiegogo are making promises that it is mathematically impossible for them to keep. People have been doing this since forever but the Internet makes it easier than ever.<p>That said, the 3D printer &quot;market&quot;, if you will, is remarkable. It is being created in exactly the same way that the microcomputer market was created, and from that we can potentially predict its path and some opportunities there.<p>For those of you too young to remember, IBM was not a participant in the microcomputer market, which was called microcomputers (as opposed to minicomputers from the likes of DEC or Data General). Between 1972 (introduction of the 8008 microprocessor from Intel) and 1981 (introduction of the IBM PC from IBM), the market consisted of lots of documentation (datasheets) from manufacturers, re-using 100 pin edge connectors for wiring busses, and lots of source code published in &quot;magazines&quot; which were no more than photocopied pages stapled together. And then, as with 3D printers today, the nerdiest of the nerds all had to have one but they didn&#x27;t know why, and the rest of the world felt that owning your own computer was a ridiculous idea. Why pay all that money so that you could play &#x27;hunt the wumpus&#x27; in a painfully slow loading stripped down version of the BASIC programming language?<p>Stepping back though, the excitement, the nerd fests, the people going nuts over these things, like a spinning top in a maze randomly bouncing off the walls, eventually found a solution, or a use, for this technology that non-nerds could relate too. Initially that was spreadsheets, then small databases, and data collection and analysis that the &#x27;big&#x27; computers had been doing since the start. There is a lot of capital trying to figure out which company is the next Microsoft.<p>When I was at Google, a relative of one of the Makerbot founders worked at Google as well, and sought out volunteers to help build their first &#x27;Cupcake&#x27; printer kit and give feedback. It was an amazing little box which, when the stars and planets aligned, would actually print reliably for the 3 hours or so needed to make a print of any decent size. It was a marvel, and sometimes it worked reliably.<p>I ended up buying a Makerbot Replicator 1 at a weak moment at Makerfair when they were 15% off list (they were clearing out old stock to make room for the Replicator 2) I printed a few test prints and then it sat on the shelf for nearly two years. I recently fired it up again because my daughter wanted me to print her a measuring spoon for chai.<p>Diving back into it I realized that there are a lot of people &quot;building&quot; 3D printers but not nearly so many &quot;engineering&quot; 3D printers. Circuit design for example on the RAMPS 1.4 board has little to no self protection circuitry and so it is prone to being damaged by things like moving the steppers when the power isn&#x27;t applied. Extruders of all types with varying levels of material feed, feedback, and consistency. Machines that require amazing amounts of calibration to print correctly assembled with fasteners without any locking capability at all. It is so much like early S-100 microcomputers it is practically a deja-vu experience.<p>3D printing is a &#x27;thing&#x27;, and its going to be a huge thing. While I get that there may not be anything today you want to print, I can also imagine a day where all your currently disposable plastic goods (cups, utensils, clips, etc) would be printed on demand and disposed in a hopper that would recycle them as raw material for the next round. Because of the amount of investment going on, I don&#x27;t worry about novices who go bankrupt trying to build a printer after a successful crowd funding campaign. I feel feel badly for them sure, but I don&#x27;t think it will derail this particular technology bloom.</text></comment> | <story><title>Low-cost 3D printers and crowdfunding suicide</title><url>http://3dprototypesandmodels.com.au/blog-2/</url></story><parent_chain></parent_chain><comment><author>tmuir</author><text>Another big problem that makes it hard to raise this kind of money is that none of these FDM printers have anything that differentiates them from every other printer. There are only a handful of different gcode interpreter&#x2F;motor driver boards. Almost all of those boards use the same Marlin firmware. They all use Repetier Host has the PC software. They all have roughly the same printing resolution. They all use the same hotends. They all require too much tinkering, maintenance, and learning curve to ever be a mass market product. On top of that, ask any hobbyist who owns one how often they use it after the initial wow factor wears off. Most people run out of stuff to print pretty quickly.<p>Besides Makerbot, who has the advantages of 5 years of experience and VC money, none of the low end FDM printers have any sort of moat to make new customers want their printer over the newest kickstarter project. And Makerbot isn&#x27;t even competing on price anymore.<p>Without some major innovation in ease of use, print resolution, print time, print material, or reliability, I don&#x27;t see how a startup FDM printer can raise the kind of money outlined in the article.</text></comment> |
21,878,137 | 21,875,895 | 1 | 2 | 21,871,645 | train | <story><title>Java JIT vs. Java AOT vs. Go for Small, Short-Lived Processes</title><url>http://macias.info/entry/201912201300_graal_aot.md</url></story><parent_chain><item><author>mdasen</author><text>GC is complicated because what works well for one thing might not work well for other things. GC can be both about clever things and about hard trade-offs. That said, I&#x27;ll try to talk about GC without having a religious war erupt. Keep in mind, something here might be wrong.<p>Go&#x27;s GC is tuned for low latency. People hate pause times. They hate it even more than they hate slowness. Go makes a trade-off for short pause times, but in doing so they do sacrifice throughput. That&#x27;s great for a web server. With a web server, you care a lot about pauses offering a bad experience. A 500ms pause is going to give a customer a bad experience. That&#x27;s less good for batch processing. Let&#x27;s say you&#x27;re running a backfill job that you expect to take 10-15 hours. You don&#x27;t really care if it pauses for 1 minute to GC every 30 minutes. No one will even know. However, you will know whether it took 10 hours or 15 hours.<p>Go&#x27;s GC is meant to be simple for users. I think the only option is &quot;how much memory should I be targeting to use?&quot; That makes it dead simple for users. I think the Go authors are right that too many tuning knobs can be a bad thing. People can spend months doing nonsense work. Worst is that knobs are hard to verify that anything is really different. If you&#x27;re running a web server, was the traffic really the same? What about other things happening on the machine? Wouldn&#x27;t you rather be writing software instead?<p>Go&#x27;s GC is a non-copying GC. One thing this means is that Go&#x27;s (heap) memory allocation ends up being very different because Go&#x27;s memory is going to become fragmented. So Go needs to keep a map of free memory and allocations are a bit more expensive. Java (with GCs like G1) can just bump allocate which is insanely cheap. This is because Java is allocating everything contiguously so it just needs to move a pointer. How does that work once something becomes freed? Java&#x27;s G1 (the default in the latest LTS Java) will copy everything that&#x27;s still alive to a new portion of memory and the old portion is then just empty. You kind of see this in Go&#x27;s culture. Web frameworks obsess about not making heap allocations. Libraries often have you pass in pointers to be filled in rather than returning something.<p>Go misses out on the generational hypothesis. The generational hypothesis is one of the more durable observations we have about programming - that most things that are allocated die really quickly. C# and Java both use generational collectors by default and they&#x27;ve done way better than what came before. C# and Java don&#x27;t have as-low pause times as Go, but part of that is that they&#x27;re targeting other things like throughput or heap overhead more.<p>Go doesn&#x27;t need GC as much. Go can allocate more on the stack than Java can and, well, Go programmers are sometimes a bit obsessed with stack allocations even when it makes for more complicated code. Having structs means creating something where you can just have contiguous memory rather than allocating separate things for the fields in your object. Go&#x27;s authors have observed that a lot of their objects that die-young are stack allocated and so while the generational hypothesis holds, it&#x27;s a bit different. Go has put a good amount of effort into escape analysis to get more stuff stack allocated.<p>Java has two new algorithms ZGC and Shenandoah which are available in the latest Java. They&#x27;re pretty impressive and usually get down to sub-millisecond pause times and even 99th percentile pauses of 1-2ms.<p>Go&#x27;s new GC was constrained by the fact that &quot;Go also desperately needed short term success in 2015&quot; (Rick Hudson from Google&#x27;s Go team) and the fact that they wanted their foreign function interface to be simple - if you don&#x27;t move objects in memory, you don&#x27;t have to worry about dealing with the indirection you&#x27;d need between C&#x2F;C++ expecting them to be in one place and Go moving them around. Google&#x27;s going to have a lot of code they want to use without replacing it with Go code and so C&#x2F;C++ interop is going to be huge in terms of their internal goals (and in terms of what the team targeted regardless of whether it&#x27;s useful to you). And I think once they had shown off sub-millisecond pause times, they were really hesitant to do something that might introduce things like 5ms pause times. I think they might have also said that Google had an obsession with the long-tail at the time. Especially at Google, there&#x27;s going to be a very long tail and if that&#x27;s what people are all talking about and caring about, you end up wanting to target that.<p>Go has tried other algorithms. They had a request-oriented-collector and that worked well, but it slowed down certain applications that Go programmers care about - namely, the compiler. They tried a non-copying generational GC, but didn&#x27;t have a lot of success there.<p>Ultimately, Go wanted fast success and to solve the #1 complaint people had: extreme pause times (pause times that would make JVM developers feel sorry for Go programmers). Going with a copying GC might have offered better performance, but would have meant a lot more work. And Go gets away with some things because more stuff gets stack allocated and Go programmers try to avoid heap allocations which would be more expensive given Go&#x27;s choices (programming for the GC algorithm).<p>I don&#x27;t think that JVM languages have a programming style that lends themselves to an over-reliance on churning through short-term garbage. Well, Clojure probably since I think it does go for the functional&#x2F;immutable allocate-a-lot style. Maybe Kotlin and Scala if you&#x27;re creating lots of immutable stuff just to re-allocate&#x2F;copy when you want to change one field. That doesn&#x27;t really apply to most Java programs. And I have covered the way that Go potentially leads to more stack allocations. However, I don&#x27;t think most people know how their Go programs work any more than their JVM programs and this really just seems to be a &quot;I want to dislike Java&quot; kind of thing rather than something about memory.<p>Java programs tend to start slow because of the JVM and JIT compilation. Java has been focused on throughput more than Go has (at the expense of responsiveness). That is changing with the two latest Java GC algorithms (and even G1 which is really good). Java is also working on modularizing itself so that you won&#x27;t bring along as much of what you don&#x27;t need (Jigsaw) and AOT compilation. Saying that Java is slow just isn&#x27;t really true, but it might feel true - things like startup times and pause times can inform our opinions a lot. There&#x27;s absolutely no question that Java is a lot faster than Python, but Python can feel faster for simple programs that aren&#x27;t doing a lot (or are just small amounts of Python doing most of the heavy work in C).<p>I mean, are you including Android in &quot;all Java apps&quot;?<p>Java, C#, and Go are all really wonderful languages&#x2F;platforms (including things hosted on them like Kotlin). They&#x27;re all around the same performance, but they do have some differences. I think Go should re-visit their GC decisions in the future, especially as ZGC and Shenandoah take shape, but their GC works pretty well. But there are certainly trade-offs being made (and it isn&#x27;t around language features that make the platform productive for programmers). I think GC is very interesting, but ultimately Java, C#, and Go all have very good GC that offers a good experience.</text></item><item><author>smabie</author><text>The style of programming for the dominant JVM languages (Java, Scala, Kotlin) involves an over reliance on churning through a lot of short-term garbage. I think this partially due to how annoying the platform is to actually use. Incredibly complex and chalk full of programmer pitfalls (like type-erasure). In fact, the majority of devs have no clue at how the JVM works, treating it as just “magic.”<p>I would imagine that Go’s GC is worse than the tunable JVM ones, but go isn’t powerful enough that one would ever be tempted to program in an abstraction heavy style. While I would argue it takes a lot more work to program in Go ca, day Scala, I think the constraints imposed lead to better software. I for one have never seen a JVM desktop or business app that worked super well. The JVM manages to be a highly optimized efficient platform in which almost exclusively slow, laggy, and memory hungry apps are produced. With that being said, it’s a highly productive platform (if you’re using something besides Java), especially for backend business apps.</text></item><item><author>shantly</author><text>It’s been the case ever since I can remember that “Java’s basically as efficient as native, it’s super fast” but all the actual java software I encounter is a slowish, bloated memory hog. I don’t know why, but that’s how it is.</text></item><item><author>robfig</author><text>It&#x27;s hard to square these articles with the reality I see on the ground: our baseline memory usage for common types of Java service is 1 GB, vs 50 MB for Go. We do have a few mammoth servers at the top end in both languages though (e.g. 75 GB heaps)<p>The deploy JARs have 100+ MB of class files, so perhaps it&#x27;s a function of all the dependencies that you &quot;need&quot; for an enterprise Java program and not something more fundamental.<p>These blog posts also present AOT as if its just another option you can toggle, while my impression is that it&#x27;s incompatible with common Java libraries and requires swapping &#x2F; maintaining a parallel compiler in the build toolchain, configuring separate rules to build it AOT, etc. I don&#x27;t have actual experience with it though so I could be missing something.</text></item></parent_chain><comment><author>loopz</author><text>Nice writeup!<p>If a programmer really cares for performant batch-processing (not sure it&#x27;s a direction we&#x27;re going for anymore but), you&#x27;ll just reuse your objects and dataspace. Understanding Go in order to minimize GC is pretty trivial and the standard library includes packages and tooling exactly for such purposes. Most likely, there are only a few hotspots needing to be tended to like this.<p>So this sounds like Golang is optimized for what people love about computing: fast response time. Also providing programmers with basic performance &quot;out of the box&quot;, which is another good tradeoff for me.<p>The tradeoff works best when you make simple designs, not huge behemoths. Spending more time reiterating clever designs, rather than jamming the keyboard until done.</text></comment> | <story><title>Java JIT vs. Java AOT vs. Go for Small, Short-Lived Processes</title><url>http://macias.info/entry/201912201300_graal_aot.md</url></story><parent_chain><item><author>mdasen</author><text>GC is complicated because what works well for one thing might not work well for other things. GC can be both about clever things and about hard trade-offs. That said, I&#x27;ll try to talk about GC without having a religious war erupt. Keep in mind, something here might be wrong.<p>Go&#x27;s GC is tuned for low latency. People hate pause times. They hate it even more than they hate slowness. Go makes a trade-off for short pause times, but in doing so they do sacrifice throughput. That&#x27;s great for a web server. With a web server, you care a lot about pauses offering a bad experience. A 500ms pause is going to give a customer a bad experience. That&#x27;s less good for batch processing. Let&#x27;s say you&#x27;re running a backfill job that you expect to take 10-15 hours. You don&#x27;t really care if it pauses for 1 minute to GC every 30 minutes. No one will even know. However, you will know whether it took 10 hours or 15 hours.<p>Go&#x27;s GC is meant to be simple for users. I think the only option is &quot;how much memory should I be targeting to use?&quot; That makes it dead simple for users. I think the Go authors are right that too many tuning knobs can be a bad thing. People can spend months doing nonsense work. Worst is that knobs are hard to verify that anything is really different. If you&#x27;re running a web server, was the traffic really the same? What about other things happening on the machine? Wouldn&#x27;t you rather be writing software instead?<p>Go&#x27;s GC is a non-copying GC. One thing this means is that Go&#x27;s (heap) memory allocation ends up being very different because Go&#x27;s memory is going to become fragmented. So Go needs to keep a map of free memory and allocations are a bit more expensive. Java (with GCs like G1) can just bump allocate which is insanely cheap. This is because Java is allocating everything contiguously so it just needs to move a pointer. How does that work once something becomes freed? Java&#x27;s G1 (the default in the latest LTS Java) will copy everything that&#x27;s still alive to a new portion of memory and the old portion is then just empty. You kind of see this in Go&#x27;s culture. Web frameworks obsess about not making heap allocations. Libraries often have you pass in pointers to be filled in rather than returning something.<p>Go misses out on the generational hypothesis. The generational hypothesis is one of the more durable observations we have about programming - that most things that are allocated die really quickly. C# and Java both use generational collectors by default and they&#x27;ve done way better than what came before. C# and Java don&#x27;t have as-low pause times as Go, but part of that is that they&#x27;re targeting other things like throughput or heap overhead more.<p>Go doesn&#x27;t need GC as much. Go can allocate more on the stack than Java can and, well, Go programmers are sometimes a bit obsessed with stack allocations even when it makes for more complicated code. Having structs means creating something where you can just have contiguous memory rather than allocating separate things for the fields in your object. Go&#x27;s authors have observed that a lot of their objects that die-young are stack allocated and so while the generational hypothesis holds, it&#x27;s a bit different. Go has put a good amount of effort into escape analysis to get more stuff stack allocated.<p>Java has two new algorithms ZGC and Shenandoah which are available in the latest Java. They&#x27;re pretty impressive and usually get down to sub-millisecond pause times and even 99th percentile pauses of 1-2ms.<p>Go&#x27;s new GC was constrained by the fact that &quot;Go also desperately needed short term success in 2015&quot; (Rick Hudson from Google&#x27;s Go team) and the fact that they wanted their foreign function interface to be simple - if you don&#x27;t move objects in memory, you don&#x27;t have to worry about dealing with the indirection you&#x27;d need between C&#x2F;C++ expecting them to be in one place and Go moving them around. Google&#x27;s going to have a lot of code they want to use without replacing it with Go code and so C&#x2F;C++ interop is going to be huge in terms of their internal goals (and in terms of what the team targeted regardless of whether it&#x27;s useful to you). And I think once they had shown off sub-millisecond pause times, they were really hesitant to do something that might introduce things like 5ms pause times. I think they might have also said that Google had an obsession with the long-tail at the time. Especially at Google, there&#x27;s going to be a very long tail and if that&#x27;s what people are all talking about and caring about, you end up wanting to target that.<p>Go has tried other algorithms. They had a request-oriented-collector and that worked well, but it slowed down certain applications that Go programmers care about - namely, the compiler. They tried a non-copying generational GC, but didn&#x27;t have a lot of success there.<p>Ultimately, Go wanted fast success and to solve the #1 complaint people had: extreme pause times (pause times that would make JVM developers feel sorry for Go programmers). Going with a copying GC might have offered better performance, but would have meant a lot more work. And Go gets away with some things because more stuff gets stack allocated and Go programmers try to avoid heap allocations which would be more expensive given Go&#x27;s choices (programming for the GC algorithm).<p>I don&#x27;t think that JVM languages have a programming style that lends themselves to an over-reliance on churning through short-term garbage. Well, Clojure probably since I think it does go for the functional&#x2F;immutable allocate-a-lot style. Maybe Kotlin and Scala if you&#x27;re creating lots of immutable stuff just to re-allocate&#x2F;copy when you want to change one field. That doesn&#x27;t really apply to most Java programs. And I have covered the way that Go potentially leads to more stack allocations. However, I don&#x27;t think most people know how their Go programs work any more than their JVM programs and this really just seems to be a &quot;I want to dislike Java&quot; kind of thing rather than something about memory.<p>Java programs tend to start slow because of the JVM and JIT compilation. Java has been focused on throughput more than Go has (at the expense of responsiveness). That is changing with the two latest Java GC algorithms (and even G1 which is really good). Java is also working on modularizing itself so that you won&#x27;t bring along as much of what you don&#x27;t need (Jigsaw) and AOT compilation. Saying that Java is slow just isn&#x27;t really true, but it might feel true - things like startup times and pause times can inform our opinions a lot. There&#x27;s absolutely no question that Java is a lot faster than Python, but Python can feel faster for simple programs that aren&#x27;t doing a lot (or are just small amounts of Python doing most of the heavy work in C).<p>I mean, are you including Android in &quot;all Java apps&quot;?<p>Java, C#, and Go are all really wonderful languages&#x2F;platforms (including things hosted on them like Kotlin). They&#x27;re all around the same performance, but they do have some differences. I think Go should re-visit their GC decisions in the future, especially as ZGC and Shenandoah take shape, but their GC works pretty well. But there are certainly trade-offs being made (and it isn&#x27;t around language features that make the platform productive for programmers). I think GC is very interesting, but ultimately Java, C#, and Go all have very good GC that offers a good experience.</text></item><item><author>smabie</author><text>The style of programming for the dominant JVM languages (Java, Scala, Kotlin) involves an over reliance on churning through a lot of short-term garbage. I think this partially due to how annoying the platform is to actually use. Incredibly complex and chalk full of programmer pitfalls (like type-erasure). In fact, the majority of devs have no clue at how the JVM works, treating it as just “magic.”<p>I would imagine that Go’s GC is worse than the tunable JVM ones, but go isn’t powerful enough that one would ever be tempted to program in an abstraction heavy style. While I would argue it takes a lot more work to program in Go ca, day Scala, I think the constraints imposed lead to better software. I for one have never seen a JVM desktop or business app that worked super well. The JVM manages to be a highly optimized efficient platform in which almost exclusively slow, laggy, and memory hungry apps are produced. With that being said, it’s a highly productive platform (if you’re using something besides Java), especially for backend business apps.</text></item><item><author>shantly</author><text>It’s been the case ever since I can remember that “Java’s basically as efficient as native, it’s super fast” but all the actual java software I encounter is a slowish, bloated memory hog. I don’t know why, but that’s how it is.</text></item><item><author>robfig</author><text>It&#x27;s hard to square these articles with the reality I see on the ground: our baseline memory usage for common types of Java service is 1 GB, vs 50 MB for Go. We do have a few mammoth servers at the top end in both languages though (e.g. 75 GB heaps)<p>The deploy JARs have 100+ MB of class files, so perhaps it&#x27;s a function of all the dependencies that you &quot;need&quot; for an enterprise Java program and not something more fundamental.<p>These blog posts also present AOT as if its just another option you can toggle, while my impression is that it&#x27;s incompatible with common Java libraries and requires swapping &#x2F; maintaining a parallel compiler in the build toolchain, configuring separate rules to build it AOT, etc. I don&#x27;t have actual experience with it though so I could be missing something.</text></item></parent_chain><comment><author>ptx</author><text>Java is slower than Python for simple short-running programs. When Python is finished Java is still struggling through start-up and class-loading.</text></comment> |
11,519,328 | 11,519,230 | 1 | 2 | 11,518,795 | train | <story><title>Instagram Is Ruining Vacation</title><url>https://backchannel.com/instagram-is-ruining-vacation-701086a67440#.axm8wkawc</url></story><parent_chain><item><author>dkopi</author><text>&quot;But as I stood there in Cambodia, swatted by selfie sticks, bruised by elbows, perfumed by the body odor of my fellow photogs, I realized the irony of being at a temple in which no one was really present. Was Instagram ruining travel?&quot;
...
&quot;But here’s to hoping that before clicking “share,” there’s a moment to enjoy the view and a sip of instant coffee. Without getting elbowed. Or poked with a selfie stick. Or tagging that photo #Instabruise.&quot;</text></item><item><author>onewaystreet</author><text>&gt; I find these types of criticisms of photography just part of the hipster &quot;backpacker mentality&quot;<p>You should actually read the article because the author isn&#x27;t one of those people.</text></item><item><author>dkopi</author><text>Photography is an incredible way to experience a new location.
It isn&#x27;t ruining vacations, its redefining them.<p>In Laos, I met a young munk named Khao at the top of Phou si in Luang Prabang. He kindly posed for a photo, and then we had a conversation. He gave me his email, and I sent him the photo later on.<p>The &quot;perfect photo&quot; isn&#x27;t the goal. Memories and being able to share our awesome travel experiences with friends and family is.<p>I find these types of criticisms of photography just part of the hipster &quot;backpacker mentality&quot;: I&#x27;m better than the mainstream. I don&#x27;t visit the same tourist destinations like everyone else. I eat the local food and not at touristic restaurants. I see a sunset and I don&#x27;t upload it to Instagram.</text></item></parent_chain><comment><author>l33tbro</author><text>Still don&#x27;t really see what is &quot;hipster&quot; about these quotes from the article. They&#x27;re pretty fair criticisms of what mainstream tourism is like today.<p>While it&#x27;s a bit much to say that Instagram is ruining vacations, it&#x27;s also a touch naive to say that the product is facilitating &quot;memories and being able to share our awesome travel experiences with friends and family is&quot;. Social media is, for the most parts, a tool to elevate your perceived social status. So saying &quot;It isn&#x27;t ruining vacations, its redefining them&quot; in the context of Instagram just reads like a Zuckerberg pull-quote.</text></comment> | <story><title>Instagram Is Ruining Vacation</title><url>https://backchannel.com/instagram-is-ruining-vacation-701086a67440#.axm8wkawc</url></story><parent_chain><item><author>dkopi</author><text>&quot;But as I stood there in Cambodia, swatted by selfie sticks, bruised by elbows, perfumed by the body odor of my fellow photogs, I realized the irony of being at a temple in which no one was really present. Was Instagram ruining travel?&quot;
...
&quot;But here’s to hoping that before clicking “share,” there’s a moment to enjoy the view and a sip of instant coffee. Without getting elbowed. Or poked with a selfie stick. Or tagging that photo #Instabruise.&quot;</text></item><item><author>onewaystreet</author><text>&gt; I find these types of criticisms of photography just part of the hipster &quot;backpacker mentality&quot;<p>You should actually read the article because the author isn&#x27;t one of those people.</text></item><item><author>dkopi</author><text>Photography is an incredible way to experience a new location.
It isn&#x27;t ruining vacations, its redefining them.<p>In Laos, I met a young munk named Khao at the top of Phou si in Luang Prabang. He kindly posed for a photo, and then we had a conversation. He gave me his email, and I sent him the photo later on.<p>The &quot;perfect photo&quot; isn&#x27;t the goal. Memories and being able to share our awesome travel experiences with friends and family is.<p>I find these types of criticisms of photography just part of the hipster &quot;backpacker mentality&quot;: I&#x27;m better than the mainstream. I don&#x27;t visit the same tourist destinations like everyone else. I eat the local food and not at touristic restaurants. I see a sunset and I don&#x27;t upload it to Instagram.</text></item></parent_chain><comment><author>onewaystreet</author><text>&quot;So, whether at Angkor Wat, Times Square, or my breakfast table, I will continue to pay homage to the melt. But here’s to hoping that before clicking “share,” there’s a moment to enjoy the view and a sip of instant coffee. Without getting elbowed. Or poked with a selfie stick. Or tagging that photo #Instabruise.&quot;</text></comment> |
6,521,118 | 6,521,073 | 1 | 2 | 6,520,702 | train | <story><title>How much can an extra hour's sleep change you?</title><url>http://www.bbc.co.uk/news/magazine-24444634</url><text></text></story><parent_chain></parent_chain><comment><author>joonix</author><text>Doesn&#x27;t matter for me. I hardly get much deep sleep due to anatomical issues essentially blocking my airways. I can&#x27;t afford the surgery (even though I have insurance) nor will my government contribute to it (even though it would increase my productivity multiple times over, meanwhile they throw trillions at universities offering useless degrees).<p>But yes, we have a culture that doesn&#x27;t value sleep. I know of people who schedule emails to be sent in the middle of the night just so they appear to be working all the time. Construction can begin at 7AM in NYC, waking up hundreds of people in surrounding buildings. Gas leaf blowers emit one of the most stressful drone sounds out in the suburbs, causing stress to dozens of neighbors, so one man can clear a pavement of dead leaves quicker.<p>And just this morning over Manhattan, a helicopter hovered at 6AM for at least an <i>hour</i>. An hour, sitting there, hovering, for God knows what reason, awaking possibly <i>thousands</i> of people early, causing possibly <i>millions</i> in lost productivity today.</text></comment> | <story><title>How much can an extra hour's sleep change you?</title><url>http://www.bbc.co.uk/news/magazine-24444634</url><text></text></story><parent_chain></parent_chain><comment><author>Peroni</author><text>I will. As soon as my son becomes a teenager I have every intention of remaining asleep beyond 7AM. Until then, I&#x27;m at his unlimited, energetic mercy.</text></comment> |
16,615,892 | 16,615,859 | 1 | 3 | 16,615,275 | train | <story><title>An attempt to upgrade to Webpack 4</title><url>https://gist.github.com/gricard/e8057f7de1029f9036a990af95c62ba8</url></story><parent_chain></parent_chain><comment><author>andrew_</author><text>Hello. Webpack team member here. The webpack@4 release was, well, interesting. I can&#x27;t add much that hasn&#x27;t already been said - but I can say that we&#x27;re working to make sure there will never be another webpack release that suffers from the same kind of documentation shortcomings. There&#x27;s a metric poopton of good stuff in the release, and we&#x27;re still working to get all of it written down. That said, look for future major releases to be far more proactive than reactive with regard to documentation, announcements, and migration details.</text></comment> | <story><title>An attempt to upgrade to Webpack 4</title><url>https://gist.github.com/gricard/e8057f7de1029f9036a990af95c62ba8</url></story><parent_chain></parent_chain><comment><author>lmcardle</author><text>As someone who has worked with many of the various tools out there, I have a lot of respect for the way the EmberJS team handles upgrades. I personally wish more teams learned from their success. Their mantra is “stability without stagnation.” While it is not perfect, the team appears to try hard to move the framework forward without breaking backwards compatibility. When it comes to a major release, I.e. the just released 3.0, the team released no new features, but rather just removed the features that had been depreciated in earlier releases.<p>Like or hate Ember as a framework, the team has developed some very good release processes and practices a lot of us could learn from.</text></comment> |
33,301,001 | 33,300,529 | 1 | 3 | 33,299,222 | train | <story><title>Ask HN: Stripe holding funds for 120 days for no reason</title><text>Hi Guys--<p>I started a Stripe account (even incorporated through them) for a basic graphic design and web design service business.<p>I process a few charges and even though I didn&#x27;t get a single chargeback or dispute, Stripe decided to deactivate my account and said they would refund all the charges that were processed.<p>Which would have been fine with me. They said they would refund on Oct 17, but that date came and past. So I kept emailing.<p>Now they&#x27;re saying they&#x27;re holding all the funds for 120 days because of &quot;elevated risk&quot;.<p>Which is insane because they have already withdrawn all the funds, meaning their risk would be zero if they refunded everyone.<p>I am beyond hurt and confused as I did need this money for my daughter. These decisions have real impacts on real families.<p>What do you do in this scenario? I have tried contacting support at Stripe but seems to be of no help.</text></story><parent_chain></parent_chain><comment><author>johnhaddock</author><text>I’m the person leading the project to make sure that people have fewer bad experiences with Stripe. I’m not sure what is driving the uptick in posts on the subject to HN in particular (obviously we pay attention to broader online discussion in addition to monitoring our support systems, though we know it creates an incentive for people to publicize their situation).<p>On the topic of Stripe and these kinds of incidents more broadly, there’s a lot to say, but here are a few pieces of context that are probably relevant:<p>- We are a giant distributed bounty system for people to find interesting and scalable ways to defraud us.<p>- We’ve seen significant upticks in certain kinds of fraud over the past couple of months. When businesses default, Stripe takes on the loss. It’s worth noting that certain kinds of fraud, like card testing, can also have significant collateral costs for legitimate Stripe businesses, and our systems and processes are not only to protect Stripe itself.<p>- We are far from oblivious to the harm that mistakes in our systems can cause. (I interact with a lot of these cases personally.) One of my highest priorities is creating better appeals flows for when we’re wrong.<p>- We’ve shipped 7 substantial improvements just in the last 10 days that should meaningfully reduce the occurrence of false positives.<p>- Publicly-described facts of specific cases don’t always match the actual facts. Stripe is sometimes just wrong. (We made some mistakes that I feel bad about in one recent case and we ended up bringing the company’s founders to an all hands last week to make sure we learned as much as possible.) But users do also sometimes publicly misrepresent what’s going on. We’re also restricted by privacy rules to not share specifics in those cases.<p>- Stripe works with millions of businesses and we see all kinds of “rare” failure modes fairly frequently. (Disputes between staff at a business, business impersonation, businesses that start legitimate and go bad, and so on.)<p>- I’m working on a post to share some of our broader philosophy + policy changes that I hope to publish before the end of this year. In that, I’m also hoping we can share some relevant metrics. If HNers have any suggestions for things that might be useful to see covered (though obviously certain things can’t be publicly disclosed), feel free to suggest them.<p>Ultimately, we work hard to be worthy of the trust of businesses across the internet, and my personal mandate (supported by many others, from our cofounders down) is to find effective new ways of making mistakes less likely. “Uniformly good support at scale, in a highly adversarial environment, with very financially-motivated actors” is not easy, but I’m pretty confident that we can make a lot of progress.<p>It goes without saying we&#x27;re working on a review of OP situation. I’m happy to take general questions as well. You can also always reach me directly at [email protected].</text></comment> | <story><title>Ask HN: Stripe holding funds for 120 days for no reason</title><text>Hi Guys--<p>I started a Stripe account (even incorporated through them) for a basic graphic design and web design service business.<p>I process a few charges and even though I didn&#x27;t get a single chargeback or dispute, Stripe decided to deactivate my account and said they would refund all the charges that were processed.<p>Which would have been fine with me. They said they would refund on Oct 17, but that date came and past. So I kept emailing.<p>Now they&#x27;re saying they&#x27;re holding all the funds for 120 days because of &quot;elevated risk&quot;.<p>Which is insane because they have already withdrawn all the funds, meaning their risk would be zero if they refunded everyone.<p>I am beyond hurt and confused as I did need this money for my daughter. These decisions have real impacts on real families.<p>What do you do in this scenario? I have tried contacting support at Stripe but seems to be of no help.</text></story><parent_chain></parent_chain><comment><author>unsupp0rted</author><text>Welcome to HN Stripe Support. How may we direct your call to Edwin @ Stripe?</text></comment> |
37,286,158 | 37,283,895 | 1 | 2 | 37,283,375 | train | <story><title>Some of the error messages produced by Apple's MPW C compiler (2006)</title><url>https://www.cs.cmu.edu/~jasonh/personal/humor/compile.html</url></story><parent_chain><item><author>jonhohle</author><text>I miss this kind of playfulness in computing.<p>When I was at Amazon my manager told me that several years earlier he was responsible for updating the 404 page so he scanned a picture of a cat his daughter drew and made that the body of the page. In 2009 when I started, that was still the image, but at some point someone must have noticed and replaced it with a stock photo of a dog. The asset was still called kayli-kitty.jpg, though. It’s since been changed again to rotating pictures and references to the original are gone.</text></item></parent_chain><comment><author>varun_ch</author><text>This is really cool! The filename on certain Amazon 404 pages (eg. <a href="https:&#x2F;&#x2F;www.amazon.co.jp&#x2F;404" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.amazon.co.jp&#x2F;404</a>) is still kailey-kitty.gif (but the image has been replaced with a standard icon).<p>I also found this comment from him on a blog: <a href="https:&#x2F;&#x2F;www.davebellous.com&#x2F;2006&#x2F;09&#x2F;25&#x2F;what-the&#x2F;#comment-29013" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.davebellous.com&#x2F;2006&#x2F;09&#x2F;25&#x2F;what-the&#x2F;#comment-290...</a></text></comment> | <story><title>Some of the error messages produced by Apple's MPW C compiler (2006)</title><url>https://www.cs.cmu.edu/~jasonh/personal/humor/compile.html</url></story><parent_chain><item><author>jonhohle</author><text>I miss this kind of playfulness in computing.<p>When I was at Amazon my manager told me that several years earlier he was responsible for updating the 404 page so he scanned a picture of a cat his daughter drew and made that the body of the page. In 2009 when I started, that was still the image, but at some point someone must have noticed and replaced it with a stock photo of a dog. The asset was still called kayli-kitty.jpg, though. It’s since been changed again to rotating pictures and references to the original are gone.</text></item></parent_chain><comment><author>hgs3</author><text>&gt; The compiler is 324k in size<p>Playfulness isn&#x27;t the only thing we&#x27;ve lost. Software bloat has reached comedic levels.</text></comment> |
40,285,752 | 40,285,791 | 1 | 2 | 40,285,476 | train | <story><title>Microsoft closes several large Bethesda affiliated game studios</title><url>https://www.ign.com/articles/microsoft-closes-redfall-developer-arkane-austin-hifi-rush-developer-tango-gameworks-and-more-in-devastating-cuts-at-bethesda</url></story><parent_chain></parent_chain><comment><author>DannyPage</author><text>Hi-Fi Rush was a fantastic game, won major awards, showed off the promise of XBox Game Pass… and for their efforts their studio has been shut down. There have multiple instances of this in just the past few weeks; are publishers really going to just bet on their prime AAA titles (Call of Duty, Halo, GTA6, etc) and nothing else? And those games either take a lot of rotating studios and a long development cycle to release. What’s going to fill the gap?</text></comment> | <story><title>Microsoft closes several large Bethesda affiliated game studios</title><url>https://www.ign.com/articles/microsoft-closes-redfall-developer-arkane-austin-hifi-rush-developer-tango-gameworks-and-more-in-devastating-cuts-at-bethesda</url></story><parent_chain></parent_chain><comment><author>fidotron</author><text>What is extraordinary about the Microsoft games unit and Xbox is how immune their senior staff are to the repercussions of their bad decisions. They&#x27;re certainly not taking responsibility for the failed gamepass experiment, trashing the Xbox brand, or the acquisitions they now regret since successfully closing Activision.<p>Failing upwards has never been so conspicuously obvious as it is in modern corporate America thanks to the pervasive use of social media.</text></comment> |
17,505,588 | 17,505,617 | 1 | 2 | 17,505,357 | train | <story><title>Djbsort: A new software library for sorting arrays of integers</title><url>https://sorting.cr.yp.to/</url></story><parent_chain></parent_chain><comment><author>yoklov</author><text>The problem with AVX2 accelerated code (and much of AVX) is that unless you have a lot of it to run you end up with a substantial speed hit that comes from the cost of switching to a different power bin (which often takes 1 or 2ms!) and then running at a lower clock speed.<p>This often still ends up being an improvement over scalar code (at the cost of higher power usage), but for occasional workloads that don&#x27;t need to do multiple milliseconds of AVX instructions you tend to have better results from 4-wide vectors, which don&#x27;t have this cost.</text></comment> | <story><title>Djbsort: A new software library for sorting arrays of integers</title><url>https://sorting.cr.yp.to/</url></story><parent_chain></parent_chain><comment><author>lixtra</author><text>The main feature of the algorithm is to sort fast in constant time (for fixed n) for cryptographic purposes [0].<p>[0] <a href="https:&#x2F;&#x2F;ntruprime.cr.yp.to&#x2F;ntruprime-20170816.pdf" rel="nofollow">https:&#x2F;&#x2F;ntruprime.cr.yp.to&#x2F;ntruprime-20170816.pdf</a> P. 48</text></comment> |
25,318,548 | 25,318,632 | 1 | 2 | 25,315,180 | train | <story><title>Apple’s “Extended Dynamic Range” Brings HDR to Non-HDR Displays</title><url>https://prolost.com/blog/edr</url></story><parent_chain><item><author>boromi</author><text>It seems to me that Microsoft really makes no effort to improve subtle aspects of Windows and hardware integration. In fact, their OS is in such shambles and is a disoriented mess with respect to UI consistency. They introduced Vista, Metro, and now Fluent. Yet there&#x27;s almost no coherence and the UI is now a mish-mash of XP, Metro, and Fluent era elements. By the time they announce their next UI, you can bet you&#x27;ll now see yet another ingredient added to the jumbled soup.<p>It really is beyond belief that an organization with so many employees can fail to adhere to a uniform vision and standard and focus on correcting details.<p>I&#x27;m a life long Windows and Android user. But honestly, seeing articles like this and how smooth the UI on macOS and uniformly they apply new updates and UI changes makes me extremely jealous and resentful that Microsoft is so bad at something so basic.<p>Features are great, but users at their start point interact with UI first. They need to fix that before anything else.<p>Now they want to give you the option to run Android apps on Windows through emulation. This just going to create a bigger jumbled mess.</text></item></parent_chain><comment><author>ChuckNorris89</author><text><i>&gt;It seems to me that Microsoft really makes no effort to improve subtle aspects of Windows and hardware integration.</i><p>Latest Windows 10 iteration is by far the snappiest OS I&#x27;ve used in a long time since it uses GPU acceleration for the desktop window manager. You can check this in task manager. The icing on the cake, if you gave a laptop with 2 GPUS(Optimus) is when you can run a demanding 3D app like a game in windowed mode in parallel with other stuff like watching videos on youtube and you can see in task manager how windows uses the external GPU to render the game and the integrated GPU to accelerate your web browser, all running butterly smooth.<p><i>&gt;In fact, their OS is in such shambles and is a disoriented mess with respect to UI consistency.</i><p>True, but that&#x27;s what you get with 30 years worth of built in backwards compatibility. I can run a copy of Unreal Tournament 1999 that was just copied off an old PC with no sweat right after ripping and tearing in Doom Eternal. Can you run 20 year old software on current Apple without emulation? Apple can afford to innovate in revolutionary ways when it dumps older baggage whenever it feels like it and start from a fresh drawing board without looking back, see intel to apple silicon transition. In 2 years x86 apps will be considered legacy&#x2F;obsolete on Mac hardware. Microsoft can&#x27;t really do this with windows so yeah, it&#x27;s a mess of new GUI elements for the simple stuff and windows 2000 era GUI elements for the deep <i>pro</i> settings. The advantage is that if you&#x27;re an old time Windows user you can easily find your way using the &quot;old&quot; settings and if you&#x27;re new to windows you can do most configs through the &quot;new&quot; GUI without touching the scary looking &quot;old&quot; settings.</text></comment> | <story><title>Apple’s “Extended Dynamic Range” Brings HDR to Non-HDR Displays</title><url>https://prolost.com/blog/edr</url></story><parent_chain><item><author>boromi</author><text>It seems to me that Microsoft really makes no effort to improve subtle aspects of Windows and hardware integration. In fact, their OS is in such shambles and is a disoriented mess with respect to UI consistency. They introduced Vista, Metro, and now Fluent. Yet there&#x27;s almost no coherence and the UI is now a mish-mash of XP, Metro, and Fluent era elements. By the time they announce their next UI, you can bet you&#x27;ll now see yet another ingredient added to the jumbled soup.<p>It really is beyond belief that an organization with so many employees can fail to adhere to a uniform vision and standard and focus on correcting details.<p>I&#x27;m a life long Windows and Android user. But honestly, seeing articles like this and how smooth the UI on macOS and uniformly they apply new updates and UI changes makes me extremely jealous and resentful that Microsoft is so bad at something so basic.<p>Features are great, but users at their start point interact with UI first. They need to fix that before anything else.<p>Now they want to give you the option to run Android apps on Windows through emulation. This just going to create a bigger jumbled mess.</text></item></parent_chain><comment><author>TwoBit</author><text>The long-standing problem with Microsoft&#x27;s internal culture is that it rewards people and teams for &#x27;innovating&#x27; where innovation in many cases is just reinvention.<p>I worked at Facebook for years and it now has a similar problem. Developers are evaluated every six months on their &#x27;impact&#x27;, which results in many dropping boring work and joining teams that are doing new things, even if they aren&#x27;t needed.</text></comment> |
19,038,965 | 19,038,898 | 1 | 2 | 19,038,327 | train | <story><title>Why isn't the internet more fun and weird?</title><url>https://jarredsumner.com/codeblog/</url></story><parent_chain><item><author>beezischillin</author><text>A lot of online communities today are unbearable tho as everything&#x27;s become so politicised. It started really wildly happening around 2013-2014 where no community was to remain a zone without some minority of users politicising it for attention. It was probably something that also happened before that but it somehow became really prevalent, at least. And often that took the fun out of it. I kind of think fondly of those times. Now of course this is just anecdotal, but that&#x27;s what I&#x27;ve experienced in most communities I used to frequent and perhaps lots of other people here did too.<p>In fact, it&#x27;s something I really appreciate here on HN, the tone of discussions is rather pleasant and on topic and it&#x27;s very rare to see people intentionally driving it into the ground, although in contrast to something like a Facebook group it&#x27;s a lot less personal.</text></item><item><author>cal5k</author><text>Jarred was probably young when the internet was first taking shape, and every generation shares the feeling that things were better in their youth (aka the Golden Age Fallacy).<p>The internet was a lot smaller and inhabited by curious nerds - to find similar fun and weirdness, just find a community today that shares those properties. Packet radio, infosec, crypto, gaming, music production, etc. - there&#x27;s plenty of weird and fun to be found if you look for it.</text></item></parent_chain><comment><author>tenpies</author><text>I distinctly remember a discussion about Something Awful&#x27;s slogan (&quot;the internet makes you stupid&quot;) and how it was actually a keen observation about the internet. Pre-internet you would have people with really bad ideas. These people were so obvious to any normal person that bad thinkers generally shut up because sharing their bad ideas meant being socially ostracized or at least getting shunned.<p>But then you add the internet, and all these radically bad thinkers find each other and their ideas almost seem normal amongst their type. They not only normalize bad thinking, but they also push for even more radically bad thinking in an effort to out-do each other. End result, you end up with a vociferous contingent of town idiots who don&#x27;t realize they are town idiots because they only listen to fellow town idiots. Add advertising companies who function on a metrics-first approach, and those town idiots dictate how companies act.</text></comment> | <story><title>Why isn't the internet more fun and weird?</title><url>https://jarredsumner.com/codeblog/</url></story><parent_chain><item><author>beezischillin</author><text>A lot of online communities today are unbearable tho as everything&#x27;s become so politicised. It started really wildly happening around 2013-2014 where no community was to remain a zone without some minority of users politicising it for attention. It was probably something that also happened before that but it somehow became really prevalent, at least. And often that took the fun out of it. I kind of think fondly of those times. Now of course this is just anecdotal, but that&#x27;s what I&#x27;ve experienced in most communities I used to frequent and perhaps lots of other people here did too.<p>In fact, it&#x27;s something I really appreciate here on HN, the tone of discussions is rather pleasant and on topic and it&#x27;s very rare to see people intentionally driving it into the ground, although in contrast to something like a Facebook group it&#x27;s a lot less personal.</text></item><item><author>cal5k</author><text>Jarred was probably young when the internet was first taking shape, and every generation shares the feeling that things were better in their youth (aka the Golden Age Fallacy).<p>The internet was a lot smaller and inhabited by curious nerds - to find similar fun and weirdness, just find a community today that shares those properties. Packet radio, infosec, crypto, gaming, music production, etc. - there&#x27;s plenty of weird and fun to be found if you look for it.</text></item></parent_chain><comment><author>rootusrootus</author><text>As someone who has been interacting online since FidoNet was a thing, I have to say ... this is a problem as old as humans. There were <i>plenty</i> of politicized discussions back then, <i>plenty</i> of contentious people, <i>plenty</i> of trolls. In some ways the vibe was different, but really, it wasn&#x27;t better. Or worse, to be honest.</text></comment> |
4,448,859 | 4,448,323 | 1 | 3 | 4,448,193 | train | <story><title>Blit: a multitasking, windowed Unix GUI from 1982</title><url>http://techchannel.att.com/play-video.cfm/2012/8/27/AT&T-Archives-BLIT-UNIX-GUI</url></story><parent_chain></parent_chain><comment><author>wickedchicken</author><text><i>God</i>, here we go again. Yet another example of companies stealing from Apple. This is clearly similar enough to the Macintosh (1984) that I'm surprised Apple didn't sue the hell out of AT&#38;T. They certainly sued the hell out of Microsoft[1], so AT&#38;T must have been lucky.<p>Around the same time, Myron Krueger had the gall to demonstrate pinch-to-zoom[2], nearly 23 years before Apple patented it. So many people piggy-backing off of Cupertino's innovation :/<p>[1] <a href="http://en.wikipedia.org/wiki/Apple_Computer,_Inc._v._Microsoft_Corporation" rel="nofollow">http://en.wikipedia.org/wiki/Apple_Computer,_Inc._v._Microso...</a><p>[2] <a href="http://www.youtube.com/watch?v=dmmxVA5xhuo" rel="nofollow">http://www.youtube.com/watch?v=dmmxVA5xhuo</a> (skip to 4:32)</text></comment> | <story><title>Blit: a multitasking, windowed Unix GUI from 1982</title><url>http://techchannel.att.com/play-video.cfm/2012/8/27/AT&T-Archives-BLIT-UNIX-GUI</url></story><parent_chain></parent_chain><comment><author>majke</author><text>Rob Pike again. Over the years he's done pretty much _everything_.<p><a href="https://en.wikipedia.org/wiki/Rob_Pike" rel="nofollow">https://en.wikipedia.org/wiki/Rob_Pike</a></text></comment> |
20,625,944 | 20,626,167 | 1 | 3 | 20,625,192 | train | <story><title>Millions of Books Are Secretly in the Public Domain</title><url>https://www.vice.com/en_us/article/kz4e3e/millions-of-books-are-secretly-in-the-public-domain-you-can-download-them-free</url></story><parent_chain></parent_chain><comment><author>crazygringo</author><text>Previous discussion from 4 days ago:<p><a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=20591071" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=20591071</a><p>(Although this one is a proper news article, and that was a blog post.)</text></comment> | <story><title>Millions of Books Are Secretly in the Public Domain</title><url>https://www.vice.com/en_us/article/kz4e3e/millions-of-books-are-secretly-in-the-public-domain-you-can-download-them-free</url></story><parent_chain></parent_chain><comment><author>blihp</author><text>It is sad that Copyright, working closer to how it was originally intended, gets spun as &#x27;a quirk&#x27;. Copyright is <i>supposed</i> to expire after a period of time rather than exist in perpetuity for the sole benefit of Copyright holders.</text></comment> |
37,414,910 | 37,414,583 | 1 | 2 | 37,409,524 | train | <story><title>Exploring EXIF</title><url>https://hturan.com/writing/exploring-exif</url></story><parent_chain></parent_chain><comment><author>rhettbull</author><text>I&#x27;m the author of the osxphotos[0] tool mentioned in the article. For photos in an Apple Photos library, osxphotos gives you access to a rich set of metadata beyond what&#x27;s in the actual EXIF&#x2F;IPTC&#x2F;XMP of the image. Apple performs object classification and other AI techniques on your images but generally doesn&#x27;t expose this to the user. For example, photos are categorized as to object in them (dog, cat, breed of dog, etc.), rich reverse geolocation info (neighborhood, landmarks, etc.) and an interesting set of scores such as &quot;overall aesthetic&quot;, &quot;pleasant camera tilt&quot;, &quot;harmonious colors&quot;, etc. These can be queried using osxphotos, either from the command line, or in your own python code. (Ref API docs[1])<p>For example, to find your &quot;best&quot; photos based on overall aesthetic score and add them to the album &quot;Best Photos&quot; you could run:<p>osxphotos query --query-eval &quot;photo.score.overall &gt; 0.8&quot; --add-to-album &quot;Best Photos&quot;<p>To find good photos with trees in them you could try something like:<p>osxphotos query --query-eval &quot;photo.score.overall &gt; 0.5&quot; --label Tree --add-to-album &quot;Good Tree Photos&quot;<p>There&#x27;s quite a bit of other interesting data in Photos that you can explore with osxphotos. Run `osxphotos inspect` and it will show you all the metadata for whichever photo is currently selected in the Photos app.<p>[0] <a href="https:&#x2F;&#x2F;github.com&#x2F;RhetTbull&#x2F;osxphotos">https:&#x2F;&#x2F;github.com&#x2F;RhetTbull&#x2F;osxphotos</a>
[1] <a href="https:&#x2F;&#x2F;rhettbull.github.io&#x2F;osxphotos&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;rhettbull.github.io&#x2F;osxphotos&#x2F;</a></text></comment> | <story><title>Exploring EXIF</title><url>https://hturan.com/writing/exploring-exif</url></story><parent_chain></parent_chain><comment><author>Figs</author><text>&gt; the EXIF data marker (FFE1)<p>This isn&#x27;t quite right. That&#x27;s a JPEG application specific marker (APP1). It indicates that some additional data is embedded in the JPEG file at that point which should be interpreted in an application specific manner (e.g. skipped if the program doesn&#x27;t know how to handle it). That could be Exif tags but could also be other things (like XMP). An APP1 segment can be distinguished as Exif data based on the presence of the &quot;Exif&quot; string immediately after the segment size. (The tags themselves are stored in essentially a TIFF file embedded after the Exif identifier, if you&#x27;re curious.)</text></comment> |
33,059,170 | 33,058,441 | 1 | 2 | 33,045,539 | train | <story><title>Impoverished Emotional Lives</title><url>https://theconvivialsociety.substack.com/p/impoverished-emotional-lives</url></story><parent_chain></parent_chain><comment><author>CliffStoll</author><text>Slightly paraphrasing Donald Knuth:<p><pre><code> The Internet is a great way to stay on top of things.
But I want to get to the bottom of things.
</code></pre>
Understanding, compassion, and emotional depth require reflection, concentration, and time ... a quiet walk in a garden does more for me than an hour of social media.</text></comment> | <story><title>Impoverished Emotional Lives</title><url>https://theconvivialsociety.substack.com/p/impoverished-emotional-lives</url></story><parent_chain></parent_chain><comment><author>PuppyTailWags</author><text>I think the essay does not share my understanding of emotion. Emotions are a complex but transient physiological+psychological state in conversation with and reacting to events and thinking. The essay describes months of mourning as if people don&#x27;t still grieve for months; the performative aspect of mourning constantly for months has gone out of fashion. People still grieve for months and years after the death of a loved one and the thought that somehow the internet has caused a shortening of grieving is a claim that would require much more than a substack essay for me to believe.<p>All emotions generally last at most 10 or 15 minutes. An emotional state can refresh due to rumination, current circumstances, etc. Losing my dog causes me to weep over multiple days, but it&#x27;s not like I&#x27;m crying constantly. Instead, I cry when I would normally walk him. I cry when I would normally feed him. I cry when I go to bed by myself, without the sounds of my friend settling in too. On the same day I feel happy when I hit a new high at my gym. I feel hungry when I skip breakfast. I feel annoyed when a car booming music passes me by. It&#x27;s not like I&#x27;m sad all the time even though I am grieving.<p>There is a conversation to be had about how social media affects the rhythms of the multiple transient emotional states humans go through regularly as part of normal behavior. But I don&#x27;t think they <i>shorten</i> emotions specifically, because emotions are already pretty short phenomena. I would speculate they instead add in additional stimuli that affect what emotional states humans enter and how often they enter them.</text></comment> |
41,008,949 | 41,007,867 | 1 | 2 | 40,978,820 | train | <story><title>10-acre underground home and gardens in Fresno (2023) [video]</title><url>https://www.youtube.com/watch?v=mUKRPoQKynk</url></story><parent_chain><item><author>illuminant</author><text>It astounds me that these things do not catch on.</text></item><item><author>about3fitty</author><text>I have been both to this site and to Coober Pedy, South Australia. Pretty neat bit of architectural convergent evolution for extremely high temperature environments.</text></item></parent_chain><comment><author>Aurornis</author><text>&gt; It astounds me that these things do not catch on.<p>Basements are very common throughout the country. Once you&#x27;ve built a basement, you might as well put a house on the next floor up.<p>Modern insulation technology is very effective. It&#x27;s much cheaper to put a lot of insulation in the above-ground portion of a house than to try to build the equivalent area entirely below the grade.<p>Building entirely below the surface without putting anything on top would be massively expensive compared to the same square footage in a traditional home, even if you accounted for equivalent insulation and cooling costs. It&#x27;s not even close.</text></comment> | <story><title>10-acre underground home and gardens in Fresno (2023) [video]</title><url>https://www.youtube.com/watch?v=mUKRPoQKynk</url></story><parent_chain><item><author>illuminant</author><text>It astounds me that these things do not catch on.</text></item><item><author>about3fitty</author><text>I have been both to this site and to Coober Pedy, South Australia. Pretty neat bit of architectural convergent evolution for extremely high temperature environments.</text></item></parent_chain><comment><author>joshuanapoli</author><text>I think that underground or earth bermed homes are often failures. There are different water and ventilation concerns for these. An above ground home is pretty accessible for repairs, but these can be nearly impossible if the home is underground.</text></comment> |
37,222,794 | 37,222,746 | 1 | 2 | 37,222,191 | train | <story><title>Microsoft is bringing Python to Excel</title><url>https://www.theverge.com/2023/8/22/23841167/microsoft-excel-python-integration-support</url></story><parent_chain><item><author>cableshaft</author><text>I wish it wasn&#x27;t solely powered only by Microsoft Cloud and could support python running locally, but regardless I think this will still be huge and single-handedly modernizes Excel by a large margin.<p>This alone could eliminate the need for websites that just want this sort of data. I can think of a past project at a previous job I did building an analytics website used only by a handful of people internally that could have been just as well served with something like this, had it existed at the time.</text></item></parent_chain><comment><author>taylodl</author><text>&gt; This alone could eliminate the need for websites that just want this sort of data<p>I catch flak for this all the time! Our users just want to be able to use Excel to free-form analyze their data. IT, of which I&#x27;m a part of, insists on building web apps to do that for them. They&#x27;re <i>never</i> happy because all they wanted was Excel.<p>I keep explaining all we have to do is build a data mart built using SQL Server, and use our ETL tools to keep it populated with the data they need to analyze. Don&#x27;t let them anywhere near the source data. They keep looking at me like I have ten heads. Instead they want to launch multi-million dollar projects to bring Tableau to our users - and they don&#x27;t want it! It&#x27;s asinine!</text></comment> | <story><title>Microsoft is bringing Python to Excel</title><url>https://www.theverge.com/2023/8/22/23841167/microsoft-excel-python-integration-support</url></story><parent_chain><item><author>cableshaft</author><text>I wish it wasn&#x27;t solely powered only by Microsoft Cloud and could support python running locally, but regardless I think this will still be huge and single-handedly modernizes Excel by a large margin.<p>This alone could eliminate the need for websites that just want this sort of data. I can think of a past project at a previous job I did building an analytics website used only by a handful of people internally that could have been just as well served with something like this, had it existed at the time.</text></item></parent_chain><comment><author>inglor</author><text>Hi I work on Excel, can you write a summary of why that&#x27;s important to you and send it to [email protected] ? I&#x27;ll get it in front of a PM.<p>Alternatively PMs read all the feedback sent via the feedback UI on Excel web and desktop so you can use that instead if you prefer.</text></comment> |
37,497,706 | 37,497,697 | 1 | 3 | 37,497,199 | train | <story><title>Apple finally put USB-C in new iPhone, but limited to 23-year-old USB 2.0 speeds</title><url>https://www.pcgamer.com/apple-finally-put-usb-c-in-the-new-iphone-but-its-inexplicably-limited-to-23-year-old-usb-20-speeds/</url></story><parent_chain><item><author>ryaneager</author><text>Honestly, who cares? Does anyone even transfers data via USB from their phone anymore? It’s been 4-5 years since I’ve done that. It’s all in synced via iCloud now.</text></item></parent_chain><comment><author>ink_13</author><text>There&#x27;s no shortage of people shooting 4K video on iPhone these days. Being able to get it onto a laptop to edit in a reasonable amount of time matters to them.</text></comment> | <story><title>Apple finally put USB-C in new iPhone, but limited to 23-year-old USB 2.0 speeds</title><url>https://www.pcgamer.com/apple-finally-put-usb-c-in-the-new-iphone-but-its-inexplicably-limited-to-23-year-old-usb-20-speeds/</url></story><parent_chain><item><author>ryaneager</author><text>Honestly, who cares? Does anyone even transfers data via USB from their phone anymore? It’s been 4-5 years since I’ve done that. It’s all in synced via iCloud now.</text></item></parent_chain><comment><author>moelf</author><text>In their own launch presentation, Apple was taunting how the USB-C port now enables &quot;professional&quot; film makers to shot 4K60FPS directly to external storage</text></comment> |
36,538,109 | 36,538,123 | 1 | 2 | 36,537,598 | train | <story><title>Lawsuit claims OpenAI stole 'massive amounts of personal data'</title><url>https://www.businessinsider.com/openai-chatgpt-generative-ai-stole-personal-data-lawsuit-children-medical-2023-6</url></story><parent_chain></parent_chain><comment><author>changoplatanero</author><text>The lawsuit seems to define the &quot;private&quot; data that was stolen as all the data that is publicly availably on the internet.</text></comment> | <story><title>Lawsuit claims OpenAI stole 'massive amounts of personal data'</title><url>https://www.businessinsider.com/openai-chatgpt-generative-ai-stole-personal-data-lawsuit-children-medical-2023-6</url></story><parent_chain></parent_chain><comment><author>tedunangst</author><text>&gt; The data accessed included &quot;private information and private conversations, medical data, information about children — essentially every piece of data exchanged on the internet it could take — without notice to the owners or users of such data, much less with anyone&#x27;s permission,&quot; per the lawsuit.<p>Why aren&#x27;t they suing the hosts that allowed the data to be stolen? Where&#x27;s the notice of security breach?</text></comment> |
21,693,775 | 21,693,580 | 1 | 2 | 21,692,165 | train | <story><title>My Show HN Project got acquired after 6 months</title><url>https://medium.com/@joshahowarth/how-my-show-hn-project-got-acquired-after-6-months-954f81ca6a18</url></story><parent_chain><item><author>netsharc</author><text>Let&#x27;s say SF.. I&#x27;ll say half a year&#x27;s wage im SF is 100 - 125K. What&#x27;s X? 4 to 10? So I&#x27;m guessing he got $500K to a million.</text></item><item><author>throwawaymath</author><text><i>&gt; As for the acquisition number… that’s why you’ve read this far, right?!</i><p>Yes.<p><i>&gt; We settled on an amount equivalent to how much I’d earn as a US-based engineer working 6 months, but then multiplied by X for the traction&#x2F;success factor already achieved. And I also still have a stake in the project going forward.</i><p>Working <i>where</i> in the US? Software engineers can make wildly different amounts of money depending on where in the US they work. Even within SFBA and NYC, a subset of companies pay significantly more than others as well.<p>So why not just say the actual acquisition number?</text></item></parent_chain><comment><author>Akababa</author><text>While we&#x27;re guessing, I&#x27;ll throw my hat in the ring at 250k. Given that it failed to monetize, I&#x27;d be surprised if it&#x27;s significantly higher.<p>I know of a project that did have decent monetization and was impressive technically that only went for mid 6 figures. However the owner wanted to sell quickly so that might have depressed the price a little bit.</text></comment> | <story><title>My Show HN Project got acquired after 6 months</title><url>https://medium.com/@joshahowarth/how-my-show-hn-project-got-acquired-after-6-months-954f81ca6a18</url></story><parent_chain><item><author>netsharc</author><text>Let&#x27;s say SF.. I&#x27;ll say half a year&#x27;s wage im SF is 100 - 125K. What&#x27;s X? 4 to 10? So I&#x27;m guessing he got $500K to a million.</text></item><item><author>throwawaymath</author><text><i>&gt; As for the acquisition number… that’s why you’ve read this far, right?!</i><p>Yes.<p><i>&gt; We settled on an amount equivalent to how much I’d earn as a US-based engineer working 6 months, but then multiplied by X for the traction&#x2F;success factor already achieved. And I also still have a stake in the project going forward.</i><p>Working <i>where</i> in the US? Software engineers can make wildly different amounts of money depending on where in the US they work. Even within SFBA and NYC, a subset of companies pay significantly more than others as well.<p>So why not just say the actual acquisition number?</text></item></parent_chain><comment><author>throwawaymath</author><text>I&#x27;m pretty skeptical the numbers are anywhere in that neighborhood. But if they are, that&#x27;s a stunning success.</text></comment> |
18,091,909 | 18,089,932 | 1 | 2 | 18,086,042 | train | <story><title>The Cartel Next Door</title><url>https://www.texasmonthly.com/articles/the-cartel-next-door/</url></story><parent_chain></parent_chain><comment><author>wazoox</author><text>Notice that in the US, more than 50% of murders go unresolved (for comparison, in France 9% of murders are unresolved).<p><a href="https:&#x2F;&#x2F;www.washingtonpost.com&#x2F;graphics&#x2F;2018&#x2F;investigations&#x2F;unsolved-homicide-database&#x2F;?noredirect=on&amp;utm_term=.4dfdbdb7e98a" rel="nofollow">https:&#x2F;&#x2F;www.washingtonpost.com&#x2F;graphics&#x2F;2018&#x2F;investigations&#x2F;...</a></text></comment> | <story><title>The Cartel Next Door</title><url>https://www.texasmonthly.com/articles/the-cartel-next-door/</url></story><parent_chain></parent_chain><comment><author>walterbell</author><text><i>&gt; Instead, he explained the concept of plomo o plata. Literally translated as “lead or silver,” it’s a tactic employed by all manner of criminal organizations. It implies a choice, though the options are stark: accept a bribe or be killed.</i><p>There&#x27;s a couple of github repos of alleged iOS vulnerabilities with the same name.<p><a href="https:&#x2F;&#x2F;github.com&#x2F;MTJailed&#x2F;plataoplomo" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;MTJailed&#x2F;plataoplomo</a><p><a href="https:&#x2F;&#x2F;github.com&#x2F;userlandkernel&#x2F;plataoplomo" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;userlandkernel&#x2F;plataoplomo</a></text></comment> |
34,332,850 | 34,332,139 | 1 | 2 | 34,328,069 | train | <story><title>Theory-building and why employee churn is lethal to software companies</title><url>https://www.baldurbjarnason.com/2022/theory-building/</url></story><parent_chain><item><author>twawaaay</author><text>I specialise in joining teams with problems (inability to ship, reliability issues, performance issues, etc.)<p>At least thrice I joined teams where all original developers and the manager was long gone. The situation for some reason looked very similar in all cases. The team is complaining about supposedly bad design that is preventing them from getting work done. But the system is also large and complex and they are unable to fix&#x2F;rewrite it.<p>Lo and behold, in two out of three cases I found that the original developers actually did pretty good job designing the system but in all cases any one of new developers understands only small part of the design and is unable to contribute changes across the entire system, there is a core of the system that nobody understands even though everybody uses and the team has made changes that are against original ideas of the system quickly creating more problems and exploding complexity.<p>In all cases developers are only able to comfortably do some kinds of changes to the system that are routinely done (like adding new fields to forms, messages, etc.) that do not change how the system functions. Changing the basic structure of the system is a taboo.<p>In essence, developers treat &quot;legacy&quot; code as if it was some kind of framework or language runtime -- to be used but never to be studied to have its internals understood. (I will never understand this approach to development)<p>Making &quot;non-standard&quot; changes (yes, one team used this exact term!) pretty much always ends up with something resembling long archeological expedition with ancient traps, diseases and wild animals.</text></item></parent_chain><comment><author>belfalas</author><text><i>&gt; In essence, developers treat &quot;legacy&quot; code as if it was some kind of framework or language runtime -- to be used but never to be studied to have its internals understood. (I will never understand this approach to development)</i><p>I was baffled by this as well for 15+ years until just last year I had a key insight: most developers don’t like to read. In fact, most people don’t like to read. Most people are intrinsically lazy and since growing your skills involves hard work, they don’t do it unless pushed to do so, either by life circumstance or necessity.<p>Time and time again people have asked me: “how do I get really good at $tech or software engineering?” I recommend a few books or send them my standard list of must-reads. In 999&#x2F;1000 cases the person approaching me gives me that defeated look that I know means: “Oh. I was hoping you’d say something not involving working hard.” Most people are polite enough not to say it to my face.<p>When folks have problems with HTTP I ask them if they’ve read RFC 2616, and if not, I offer to bring them a nicely printed and bound copy. It typically goes unread though some use it as a reference. Only once did someone follow my recommendation and read it cover to cover.<p>Folks seems to lack a kind of basic economic perspective. Spend a week, month, something along those lines growing your skills (or even just 30 minutes reading the docs on that thing you just spent 2 hours declaring idiotic). Yes, this means probably giving up your favorite Netflix binge, video game, or drinking on weeknights in favor of study. But once these skills are internalized they pay back for a lifetime with good returns. Yet for some strange reason people prefer to keep thinking short term and so they never learn what they’re doing. It used to make me sad but I’m making peace with it.</text></comment> | <story><title>Theory-building and why employee churn is lethal to software companies</title><url>https://www.baldurbjarnason.com/2022/theory-building/</url></story><parent_chain><item><author>twawaaay</author><text>I specialise in joining teams with problems (inability to ship, reliability issues, performance issues, etc.)<p>At least thrice I joined teams where all original developers and the manager was long gone. The situation for some reason looked very similar in all cases. The team is complaining about supposedly bad design that is preventing them from getting work done. But the system is also large and complex and they are unable to fix&#x2F;rewrite it.<p>Lo and behold, in two out of three cases I found that the original developers actually did pretty good job designing the system but in all cases any one of new developers understands only small part of the design and is unable to contribute changes across the entire system, there is a core of the system that nobody understands even though everybody uses and the team has made changes that are against original ideas of the system quickly creating more problems and exploding complexity.<p>In all cases developers are only able to comfortably do some kinds of changes to the system that are routinely done (like adding new fields to forms, messages, etc.) that do not change how the system functions. Changing the basic structure of the system is a taboo.<p>In essence, developers treat &quot;legacy&quot; code as if it was some kind of framework or language runtime -- to be used but never to be studied to have its internals understood. (I will never understand this approach to development)<p>Making &quot;non-standard&quot; changes (yes, one team used this exact term!) pretty much always ends up with something resembling long archeological expedition with ancient traps, diseases and wild animals.</text></item></parent_chain><comment><author>jacobsenscott</author><text>&gt; as much as possible, do not allow custom patterns&#x2F;structures<p>This is key. Most engineers want to build frameworks instead of applications. But leave the frameworks to the framework builders unless you were hired to build a framework.<p>I can jump into a 10 year old rails code base and be productive right away if they stuck to the framework defaults. But invariably someone read some blog posts about how &quot;rails doesn&#x27;t scale&quot; and built some rats nest of service objects&#x2F;decorators&#x2F;portsAndAdapters wtfs and it takes some serious code spelunking to figure out what&#x27;s going on (hint - the end result is always creating or updating a database record, or making a single api call).<p>I would argue microservices also encourage bitrot. You can leave some service running on some ancient libraries for years it doesn&#x27;t need changes. But if you have a monolith where you constantly keep the dependencies up to date - everything stays up to date.</text></comment> |
10,817,534 | 10,817,411 | 1 | 2 | 10,811,570 | train | <story><title>Do the math on your stock options</title><url>http://jvns.ca/blog/2015/12/30/do-the-math-on-your-stock-options/</url></story><parent_chain><item><author>JumpCrisscross</author><text>Always ask for:<p>1. TRANSFERABILITY. If you are given options to buy privately-held common stock in lieu of compensation, you must demand transferability. Rights of first refusal (ROFRs) are fine. &quot;Board approval&quot; is not. &quot;Board approval&quot; means &quot;you may not sell your shares until we go public, except to us, if and when we feel like it, and at a price we get to unilaterally decide&quot;.<p>2. CASHLESS EXERCISABILITY. Always ask for cashless exercisability. In the public market, if you ask your broker to &quot;cashlessly exercise&quot; in-the-money options, here is what happens. First, your broker lends you the money to exercise the options. Then, the broker sells some of the resulting stock. Finally, the broker pays herself back, plus a pre-disclosed fee, and returns the rest to you.<p>This also works with private stock. (Unless you forgot Rule No. 1; if your shares aren&#x27;t transferable you&#x27;ve already been screwed.) But you may not require a broker. Some companies allow for direct cashless exercise. Suppose you hold options for 100 shares struck at $100 per share. You need $10,000 to exercise. The company figures its stock&#x27;s &quot;fair market value&quot; is $110. You check with outside sources, <i>e.g.</i> a private-stock broker, and conclude this is not much less than what you could sell your stock for in the market. Your stock is worth $11,000 to the company; it is $1,000 in the money. After cashlessly exercising with the company you would get $1,000 of stock, <i>i.e.</i> 9 shares. (Fractions of shares are usually lopped off in these calculations.)<p>Plus: you got shares without putting up capital. Minus: you lost the upside (and downside) on ninety-one shares. That said, 9 shares is better than 0 because you had no capital to exercise upon termination.</text></item></parent_chain><comment><author>inthewoods</author><text>I&#x27;ve never seen the ability to customize these kind of things because they generally are, in my experience, part of the core options agreement for the entire company and would require BoD approval. Thus, you&#x27;re asking a lot - works if your a key employee but probably not for the average employee. Would love to hear if people have been able to get these terms.</text></comment> | <story><title>Do the math on your stock options</title><url>http://jvns.ca/blog/2015/12/30/do-the-math-on-your-stock-options/</url></story><parent_chain><item><author>JumpCrisscross</author><text>Always ask for:<p>1. TRANSFERABILITY. If you are given options to buy privately-held common stock in lieu of compensation, you must demand transferability. Rights of first refusal (ROFRs) are fine. &quot;Board approval&quot; is not. &quot;Board approval&quot; means &quot;you may not sell your shares until we go public, except to us, if and when we feel like it, and at a price we get to unilaterally decide&quot;.<p>2. CASHLESS EXERCISABILITY. Always ask for cashless exercisability. In the public market, if you ask your broker to &quot;cashlessly exercise&quot; in-the-money options, here is what happens. First, your broker lends you the money to exercise the options. Then, the broker sells some of the resulting stock. Finally, the broker pays herself back, plus a pre-disclosed fee, and returns the rest to you.<p>This also works with private stock. (Unless you forgot Rule No. 1; if your shares aren&#x27;t transferable you&#x27;ve already been screwed.) But you may not require a broker. Some companies allow for direct cashless exercise. Suppose you hold options for 100 shares struck at $100 per share. You need $10,000 to exercise. The company figures its stock&#x27;s &quot;fair market value&quot; is $110. You check with outside sources, <i>e.g.</i> a private-stock broker, and conclude this is not much less than what you could sell your stock for in the market. Your stock is worth $11,000 to the company; it is $1,000 in the money. After cashlessly exercising with the company you would get $1,000 of stock, <i>i.e.</i> 9 shares. (Fractions of shares are usually lopped off in these calculations.)<p>Plus: you got shares without putting up capital. Minus: you lost the upside (and downside) on ninety-one shares. That said, 9 shares is better than 0 because you had no capital to exercise upon termination.</text></item></parent_chain><comment><author>ryanackley</author><text>Restrictions on transferability is standard for a private company. Kudos if you are able to negotiate this but I think a small company would be crazy to agree to this. It opens the door to a whole list of potential burdens for the company.</text></comment> |
10,132,009 | 10,131,864 | 1 | 2 | 10,130,758 | train | <story><title>Your Job as a Founder Is to Create Believers</title><url>http://josephwalla.com/your-job-as-a-founder-is-to-create-believers</url></story><parent_chain><item><author>riebschlager</author><text>I worked for a stretch at a startup where I was asked to &quot;believe&quot;. The company&#x27;s core values were made into wall artwork and recited aloud at staff meetings. It was all eerily similar to my Catholic upbringing. Including the guilt. &quot;Oh, you want to leave work at 5? Yeah, that&#x27;s fine. I just thought you <i>really</i> believed.&quot;<p>You have to be a little insane to start a company, I totally get that. But you&#x27;re truly doing yourself a disservice if you&#x27;re attempting to amass a legion of followers.<p>A good developer might not give two shits about your carefully crafted mission statement, but she&#x27;ll do some kick ass work. Isn&#x27;t that good enough any more? Why is &quot;startup culture&quot; so obsessed with maintaining the appearance of success rather than focusing on creating something amazing?</text></item></parent_chain><comment><author>joncalhoun</author><text>A few potential downsides to an employee not believing:<p>1. If they are offered an identical position at a company they believe in, they will leave. You won&#x27;t be able to do anything to keep them around, so you have less ability to retain valuable employees. Hiring and training new employees is expensive.<p>2. They won&#x27;t push back when bad decisions are made. Would you waste your time arguing about a bad decision if you didn&#x27;t give a shit about the company? You want people who will work for the best interest of the company. Granted there comes a time when you need people to accept a decision, but you want their to be a discussion around those decisions.<p>3. They will negotiate for more salary than equity&#x2F;options, giving them less incentive to help make the company succeed.<p>4. They are less likely to convince their friends to join the company. Stripe built an incredibly talented engineering team through mostly referrals[1]. You lose that if you are hiring people who won&#x27;t convince their friends to join.<p>I&#x27;m sure I am missing many others, and all startups may not need a team so dedicated that they would give up their salary, but having a team who believes in your company does have real benefits.<p>[1]: <a href="http:&#x2F;&#x2F;firstround.com&#x2F;review&#x2F;How-Stripe-built-one-of-Silicon-Valleys-best-engineering-teams&#x2F;" rel="nofollow">http:&#x2F;&#x2F;firstround.com&#x2F;review&#x2F;How-Stripe-built-one-of-Silicon...</a></text></comment> | <story><title>Your Job as a Founder Is to Create Believers</title><url>http://josephwalla.com/your-job-as-a-founder-is-to-create-believers</url></story><parent_chain><item><author>riebschlager</author><text>I worked for a stretch at a startup where I was asked to &quot;believe&quot;. The company&#x27;s core values were made into wall artwork and recited aloud at staff meetings. It was all eerily similar to my Catholic upbringing. Including the guilt. &quot;Oh, you want to leave work at 5? Yeah, that&#x27;s fine. I just thought you <i>really</i> believed.&quot;<p>You have to be a little insane to start a company, I totally get that. But you&#x27;re truly doing yourself a disservice if you&#x27;re attempting to amass a legion of followers.<p>A good developer might not give two shits about your carefully crafted mission statement, but she&#x27;ll do some kick ass work. Isn&#x27;t that good enough any more? Why is &quot;startup culture&quot; so obsessed with maintaining the appearance of success rather than focusing on creating something amazing?</text></item></parent_chain><comment><author>superuser2</author><text>Indeed. I work for a large tech company that values excitedness about the company over skill. We reject perfectly good engineers because they don&#x27;t display sufficient excitement or (gasp!) want to be sold on the company in interviews, rather than grovel for an opportunity to get near us.<p>I identify as a mercenary more than a missionary, and believe in professionalism independent of passion, but you don&#x27;t want to be caught saying that out loud here.</text></comment> |
40,364,350 | 40,363,937 | 1 | 2 | 40,361,387 | train | <story><title>The most talented person in the world</title><url>https://matt.sh/the-most-talented-person</url></story><parent_chain><item><author>frabjoused</author><text>Has anyone built a search engine that uses LLMs to pre-grade every page with metrics such as:<p>- Commercial bias (content compared to the source, which it learns about)<p>- Insincere motives<p>- Bloat (how many words it takes to say how little to penalize SEO bloat)<p>I would assume that using LLMs, we can get a pretty good idea of what is SEO bloat and who the bad actors are by this point, and just penalize those results.</text></item></parent_chain><comment><author>pona-a</author><text>Did exactly that, hacked together a small pipeline in Nushell with simonw&#x2F;llm. Even with GPT-4 turbo and given direct guidelines on common spam heuristics, it seems to perform worse than a Bayesian BoW. Endless trails of questions with no relation to the title it describes as informative, presence of affiliate links often gets forgotten in a mass of tokens in their tracking parameters, relevance is consistently near 0.8 even if there&#x27;s relationship between title and content, and as for insincerity, our favorite BS generator cannot for the life of it correctly recognize its own creations.<p>Your ideas for metrics are good, but LLMs seem to be quite terrible at any of these. A simple set of heuristics and maybe a tiny language model for named entity detection and &quot;vibe checking&quot; would serve you much better.<p>Also, a lot of the worst offenders seem to use the same Q&amp;A +- conclusion structure, which Viktor from marginalia.nu wrote a simple heuristic for, which I recall he said did wonders for pruning it. Solving SEO spam is easy when you aren&#x27;t the one being optimized against. What&#x27;s left is scaling and information retrieval.</text></comment> | <story><title>The most talented person in the world</title><url>https://matt.sh/the-most-talented-person</url></story><parent_chain><item><author>frabjoused</author><text>Has anyone built a search engine that uses LLMs to pre-grade every page with metrics such as:<p>- Commercial bias (content compared to the source, which it learns about)<p>- Insincere motives<p>- Bloat (how many words it takes to say how little to penalize SEO bloat)<p>I would assume that using LLMs, we can get a pretty good idea of what is SEO bloat and who the bad actors are by this point, and just penalize those results.</text></item></parent_chain><comment><author>Uehreka</author><text>What makes you think that LLMs will be better at combating spam than they are at creating it? There’s no universal rule that innovations in AI will go hand in hand with innovations in detecting AI, yet I feel like I see people talking all the time like that’s the case.<p>As of right now, LLMs are prolific but unreliable, which makes them extremely well suited for generating spam, but unsuited to detecting it without a large number of false positives and negatives.</text></comment> |
5,479,449 | 5,479,470 | 1 | 3 | 5,479,217 | train | <story><title>Netcat – The swiss army knife of networking</title><url>http://mylinuxbook.com/linux-netcat-command/</url></story><parent_chain></parent_chain><comment><author>kaeso</author><text>I couldn't really survive without socat:
<a href="http://www.dest-unreach.org/socat/doc/socat.html#EXAMPLES" rel="nofollow">http://www.dest-unreach.org/socat/doc/socat.html#EXAMPLES</a></text></comment> | <story><title>Netcat – The swiss army knife of networking</title><url>http://mylinuxbook.com/linux-netcat-command/</url></story><parent_chain></parent_chain><comment><author>c-oreills</author><text>See also ncat - netcat for the 21st century: <a href="http://nmap.org/ncat/" rel="nofollow">http://nmap.org/ncat/</a></text></comment> |
7,575,974 | 7,575,871 | 1 | 3 | 7,575,651 | train | <story><title>The race to contain West Africa's Ebola outbreak</title><url>http://www.wired.co.uk/news/archive/2014-04/11/ebola-open-street-map</url></story><parent_chain></parent_chain><comment><author>mediaman</author><text>There are two companies that have demonstrated cures to the Ebola and Marburg viruses: Tekmira, based in Vancouver BC, and Sarepta, in Boston. Initial data shows that they have 85%-100% cure rates receiving the medicine up to 72hrs from initial exposure, as tested on monkeys.<p>Unfortunately a portion of this research was funded by the DoD, which decided to cut funding in 2012 at least for the Sarepta Marburg cure, which significantly slowed down progress and prevented any stockpiling of the medicine.</text></comment> | <story><title>The race to contain West Africa's Ebola outbreak</title><url>http://www.wired.co.uk/news/archive/2014-04/11/ebola-open-street-map</url></story><parent_chain></parent_chain><comment><author>timr</author><text>This outbreak is scary because it&#x27;s finally hit a major city in Africa.<p>All you need is for one case to hop on an international flight, and things get much harder to control.</text></comment> |
23,092,702 | 23,092,822 | 1 | 2 | 23,090,393 | train | <story><title>No cookie consent walls, scrolling isn’t consent, says EU data protection body</title><url>https://techcrunch.com/2020/05/06/no-cookie-consent-walls-and-no-scrolling-isnt-consent-says-eu-data-protection-body/</url></story><parent_chain><item><author>notechback</author><text>They are not mandatory, it&#x27;s a choice by the site owner to include them. They are only mandatory if you include tracking features that track users across the web (= ads and Google analytics).</text></item><item><author>dcow</author><text>Cookie banners have taken the internet back 20 years. Now every website has a mandatory <i>popup</i>. And you can’t block these new breed because they&#x27;re part of the site.</text></item><item><author>elric</author><text>Tangent: I wish the idiom of &quot;placing&quot; cookies would go away. Websites don&#x27;t &quot;place&quot; cookies. Websites can include cookies in their HTTP responses. Your browser can include them in future requests. But it doesn&#x27;t have to. There is nothing in the HTTP spec that says you have to accept cookies or include them in subsequent requests. There certainly isn&#x27;t any reason to &quot;place&quot; them on your computer.<p>If more browsers were still User Agents in the literal sense, maybe we wouldn&#x27;t have needed this legislation. Browsers could have informed people about what cookies were, and could have presented the user with the option to never accept tracking cookies from Big Advertising. Every browser has the option to reject third party cookies or to clear all cookies at the end of the browser session.<p>This mischaracterization of cookies has, ironically, made life a lot less pleasant for people who don&#x27;t accept cookies. The &quot;opt-out&quot; is just another cookie. There&#x27;s nothing special about them either, they can be used to track return visitors just as well as any other cookie. I&#x27;m sure they&#x27;re not, because that would be against the spirit of the law ...<p>Not tracking people without consent is definitely a Good Thing, but it shouldn&#x27;t require everyone and their grandmother to put annoying cookie banners on every website under the sun. And I think it wouldn&#x27;t have, had people been better informed.</text></item></parent_chain><comment><author>mfontani</author><text>One can include ads and analytics without consent being granted - they&#x27;re &quot;just&quot; restricted to a method of delivering ads and performing analytics which don&#x27;t track the user.<p>IANAL, mind you - but that&#x27;s how we implemented it - you&#x27;re opting-in to the ads that target you and analytics which track you, or you get the non-tracking&#x2F;non-targeting ads and analytics.</text></comment> | <story><title>No cookie consent walls, scrolling isn’t consent, says EU data protection body</title><url>https://techcrunch.com/2020/05/06/no-cookie-consent-walls-and-no-scrolling-isnt-consent-says-eu-data-protection-body/</url></story><parent_chain><item><author>notechback</author><text>They are not mandatory, it&#x27;s a choice by the site owner to include them. They are only mandatory if you include tracking features that track users across the web (= ads and Google analytics).</text></item><item><author>dcow</author><text>Cookie banners have taken the internet back 20 years. Now every website has a mandatory <i>popup</i>. And you can’t block these new breed because they&#x27;re part of the site.</text></item><item><author>elric</author><text>Tangent: I wish the idiom of &quot;placing&quot; cookies would go away. Websites don&#x27;t &quot;place&quot; cookies. Websites can include cookies in their HTTP responses. Your browser can include them in future requests. But it doesn&#x27;t have to. There is nothing in the HTTP spec that says you have to accept cookies or include them in subsequent requests. There certainly isn&#x27;t any reason to &quot;place&quot; them on your computer.<p>If more browsers were still User Agents in the literal sense, maybe we wouldn&#x27;t have needed this legislation. Browsers could have informed people about what cookies were, and could have presented the user with the option to never accept tracking cookies from Big Advertising. Every browser has the option to reject third party cookies or to clear all cookies at the end of the browser session.<p>This mischaracterization of cookies has, ironically, made life a lot less pleasant for people who don&#x27;t accept cookies. The &quot;opt-out&quot; is just another cookie. There&#x27;s nothing special about them either, they can be used to track return visitors just as well as any other cookie. I&#x27;m sure they&#x27;re not, because that would be against the spirit of the law ...<p>Not tracking people without consent is definitely a Good Thing, but it shouldn&#x27;t require everyone and their grandmother to put annoying cookie banners on every website under the sun. And I think it wouldn&#x27;t have, had people been better informed.</text></item></parent_chain><comment><author>nicbou</author><text>&gt; across the web<p>That part is not necessary. They are mandatory if you collect any form of personal data without legitimate interest.</text></comment> |
15,412,051 | 15,410,715 | 1 | 3 | 15,408,800 | train | <story><title>PostgreSQL 10 Released</title><url>https://www.postgresql.org/about/news/1786/</url></story><parent_chain><item><author>thom</author><text>It&#x27;s 2017, and despite all this parallel stuff, query planners are still dumb as bricks. Every day I get annoyed at various obvious-to-the-eye WHERE clause pushdown opportunities, or situations where I can manually copy a view to a table, add indexes, and build a query on top of that quicker than letting the database fill up the entire disk with temporary files trying to do it as a subquery. It&#x27;s utterly maddening if you spend your life on ad-hoc analytics&#x2F;reporting workloads.<p>I strongly believe there are huge opportunities for AI to come along and, given a logical description of the data and query, do a better job of physically organising and serving the data. I realise some database management systems do this to an extent, but it&#x27;s pretty weak sauce in my experience.</text></item></parent_chain><comment><author>anarazel</author><text>I&#x27;d like to see your proposal get a bit more concrete. Your request seems a bit more for &quot;magic&quot; than &quot;AI&quot;.<p>You&#x27;re not likely to succeed with &quot;here&#x27;s the query and schema&quot; and have AI somehow figure out what the best plan there is. There&#x27;s a lot of rules to observe for correctness, and you can&#x27;t just try to execute arbitrary valid plans, because the initial query plans are going to be really bad. So you again need a model of what valid query plans are, and a cost model to evaluate how efficient a plan is, without having executed it.<p>I think there&#x27;s plenty opportunities for ML type improvements, but they&#x27;re in individual parts, less as a wholesale query planner replacement. I think the most likely beneficiaries are around statics collection and evaluation, incremental improvements to query plans, and some execution time -&gt; plan time feedback loops.</text></comment> | <story><title>PostgreSQL 10 Released</title><url>https://www.postgresql.org/about/news/1786/</url></story><parent_chain><item><author>thom</author><text>It&#x27;s 2017, and despite all this parallel stuff, query planners are still dumb as bricks. Every day I get annoyed at various obvious-to-the-eye WHERE clause pushdown opportunities, or situations where I can manually copy a view to a table, add indexes, and build a query on top of that quicker than letting the database fill up the entire disk with temporary files trying to do it as a subquery. It&#x27;s utterly maddening if you spend your life on ad-hoc analytics&#x2F;reporting workloads.<p>I strongly believe there are huge opportunities for AI to come along and, given a logical description of the data and query, do a better job of physically organising and serving the data. I realise some database management systems do this to an extent, but it&#x27;s pretty weak sauce in my experience.</text></item></parent_chain><comment><author>fatbird</author><text>This is interesting. When you see these &#x27;obvious to the eye&#x27; opportunities, do you stop and treat them as an exercise in generalizing the optimization for a query planner?<p>I&#x27;m curious how much query planners have to make tradeoffs between effective optimizations and not overloading the analysis phase.</text></comment> |
27,605,904 | 27,605,897 | 1 | 2 | 27,604,809 | train | <story><title>Costa Rica Has Run on 100% Renewable Electricity for 299 Days</title><url>https://www.under30experiences.com/blog/costa-rica-has-run-on-100-renewable-energy-for-299-days</url></story><parent_chain><item><author>DennisP</author><text>We really need two words for renewables: one for types that scale but don&#x27;t provide power on demand (wind, solar), and the other for types that provide power on demand but aren&#x27;t available at scale everywhere (hydro, geothermal).<p>Costa Rica is blessed with plenty of the second type and uses it. Because we lump it all into &quot;renewables,&quot; it sounds like it should be easy for everyone to follow their example, and that&#x27;s not the case. No country has run on a high percentage of wind&#x2F;solar&#x2F;battery for a long period of time.</text></item></parent_chain><comment><author>cesarb</author><text>&gt; We really need two words for renewables: one for types that scale but don&#x27;t provide power on demand (wind, solar), and the other for types that provide power on demand but aren&#x27;t available at scale everywhere (hydro, geothermal).<p>The wording you are looking for is &quot;dispatchable&quot; (<a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Dispatchable_generation" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Dispatchable_generation</a>).</text></comment> | <story><title>Costa Rica Has Run on 100% Renewable Electricity for 299 Days</title><url>https://www.under30experiences.com/blog/costa-rica-has-run-on-100-renewable-energy-for-299-days</url></story><parent_chain><item><author>DennisP</author><text>We really need two words for renewables: one for types that scale but don&#x27;t provide power on demand (wind, solar), and the other for types that provide power on demand but aren&#x27;t available at scale everywhere (hydro, geothermal).<p>Costa Rica is blessed with plenty of the second type and uses it. Because we lump it all into &quot;renewables,&quot; it sounds like it should be easy for everyone to follow their example, and that&#x27;s not the case. No country has run on a high percentage of wind&#x2F;solar&#x2F;battery for a long period of time.</text></item></parent_chain><comment><author>wp381640</author><text>&gt; No country has run on a high percentage of wind&#x2F;solar&#x2F;battery for a long period of time.<p>Only because it&#x27;s early. The first grid-scale battery only went online 3 years ago.<p>In that example (South Australia) you only have to look at this chart[0] to see what it done to grid renewable demand and where the trend is heading.<p>There are multiple GW more of additional battery coming online over the next few years<p>[0] <a href="https:&#x2F;&#x2F;opennem.org.au&#x2F;stripes&#x2F;sa1&#x2F;?metric=renewablesProportion" rel="nofollow">https:&#x2F;&#x2F;opennem.org.au&#x2F;stripes&#x2F;sa1&#x2F;?metric=renewablesProport...</a></text></comment> |
41,373,979 | 41,373,940 | 1 | 2 | 41,338,581 | train | <story><title>The Simple Algorithm That Ants Use to Build Bridges (2018)</title><url>https://www.quantamagazine.org/the-simple-algorithm-that-ants-use-to-build-bridges-20180226/</url></story><parent_chain><item><author>wonderwonder</author><text>Always found colony based insects but especially ants fascinating.
Its as if the ant is not the organism but rather the colony is.</text></item></parent_chain><comment><author>jsenn</author><text>Deborah Gordon (ant biologist) makes this argument in her book Ant Encounters. Because the ants in a colony are all sisters, the unit of reproduction is actually the colony rather than the individual ant. Great book!</text></comment> | <story><title>The Simple Algorithm That Ants Use to Build Bridges (2018)</title><url>https://www.quantamagazine.org/the-simple-algorithm-that-ants-use-to-build-bridges-20180226/</url></story><parent_chain><item><author>wonderwonder</author><text>Always found colony based insects but especially ants fascinating.
Its as if the ant is not the organism but rather the colony is.</text></item></parent_chain><comment><author>senthil_rajasek</author><text>If you find Ants interesting you would probably like E.O. Wilson&#x27;s works. That&#x27;s all he did his entire life, study ants.<p><a href="https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Journey_to_the_Ants" rel="nofollow">https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Journey_to_the_Ants</a></text></comment> |
29,906,996 | 29,906,560 | 1 | 2 | 29,872,227 | train | <story><title>Articles on Cyber-Warfare from the Modern War Institute at West Point</title><url>https://mwi.usma.edu/full-spectrum/</url></story><parent_chain></parent_chain><comment><author>ya_throw</author><text>&gt;&quot;Ignore the Human Factor at your peril&quot;<p>Indeed. Why bother with weapons (kinetic AND cyber) at all, when the both the spread and the pseudo-anonymous nature of online communication has left the thoughts, desires and opinions of the Western world wide open for study and manipulation.<p>I used to have strong opinions about the necessity for a fully free, open, anonymous, untracked, etc etc, internet. However, from a national security point of view, countries like Russia and China have it right. Why would you allow the above to happen to your citizens?</text></comment> | <story><title>Articles on Cyber-Warfare from the Modern War Institute at West Point</title><url>https://mwi.usma.edu/full-spectrum/</url></story><parent_chain></parent_chain><comment><author>Borrible</author><text>Just in the unlikely event of disarmament negotiations between cyber super powers in the future. You can count nukes and explosive power, but you can&#x27;t count cyber weapons.<p>A new deterrence doctrine,nobody knows what will happen.</text></comment> |
38,415,436 | 38,415,620 | 1 | 2 | 38,408,319 | train | <story><title>Cash Cows</title><url>https://www.ageofinvention.xyz/p/age-of-invention-cash-cows</url></story><parent_chain></parent_chain><comment><author>swalling</author><text>One interesting fact is that Darwin directly referred to Bakewell in the first chapter of On the Origin of Species:<p>“a kind of Selection, which may be called Unconscious, and which results from every one trying to possess and breed from the best individual animals, is more important. Thus, a man who intends keeping pointers naturally tries to get as good dogs as he can, and afterwards breeds from his own best dogs, but he has no wish or expectation of permanently altering the breed. Nevertheless I cannot doubt that this process, continued during centuries, would improve and modify any breed, in the same way as Bakewell, Collins, &amp;c., by this very same process, only carried on more methodically, did greatly modify, even during their own lifetimes, the forms and qualities of their cattle.”</text></comment> | <story><title>Cash Cows</title><url>https://www.ageofinvention.xyz/p/age-of-invention-cash-cows</url></story><parent_chain></parent_chain><comment><author>fbdab103</author><text>Selective breeding always struck me as interesting for the sheer amount of time required for experimentation. Pick your pairs, and then wait 2-5 years before you can repeat the cycle again.<p>Amazing that a single lifetime was able to yield measurable results.<p>Also reminded of the fox domestication program[0] where they were able to go from wild foxes to a dog-like domesticated breed within a decade.<p>&gt;Belyaev was correct that selection on tameness alone leads to the emergence of traits in the domestication syndrome. In less than a decade, some of the domesticated foxes had floppy ears and curly tails (Fig. 2). Their stress hormone levels by generation 15 were about half the stress hormone (glucocorticoid) levels of wild foxes. Over generations, their adrenal gland became smaller and smaller. Serotonin levels also increased, producing “happier” animals. Over the course of the experiment, researchers also found the domesticated foxes displayed mottled “mutt-like” fur patterns, and they had more juvenilized facial features (shorter, rounder, more dog-like snouts) and body shapes (chunkier, rather than gracile limbs)<p>[0] <a href="https:&#x2F;&#x2F;evolution-outreach.biomedcentral.com&#x2F;articles&#x2F;10.1186&#x2F;s12052-018-0090-x" rel="nofollow noreferrer">https:&#x2F;&#x2F;evolution-outreach.biomedcentral.com&#x2F;articles&#x2F;10.118...</a></text></comment> |
39,392,747 | 39,388,914 | 1 | 2 | 39,386,156 | train | <story><title>Sora: Creating video from text</title><url>https://openai.com/sora</url></story><parent_chain><item><author>gerash</author><text>When others create text to video systems (eg. Lumiere from Google) they publish the research (eg. <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;2401.12945.pdf" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;2401.12945.pdf</a>). Open AI is all about commercialization. I don&#x27;t like their attitude</text></item><item><author>crazygringo</author><text>This is insane. But I&#x27;m impressed most of all by the quality of <i>motion</i>. I&#x27;ve quite simply <i>never seen convincing computer-generated motion before</i>. Just look at the way the wooly mammoths connect with the ground, and their lumbering mass feels real.<p>Motion-capture works fine because that&#x27;s real motion, but every time people try to animate humans and animals, even in big-budget CGI movies, it&#x27;s always ultimately obviously fake. There are so many subtle things that happen in terms of acceleration and deceleration of all of the different parts of an organism, that no animator ever gets it 100% right. No animation <i>algorithm</i> gets it to a point where it&#x27;s believable, just where it&#x27;s &quot;less bad&quot;.<p>But these videos seem to be getting it entirely believable for both people and animals. Which is <i>wild</i>.<p>And then of course, not to mention that these are entirely believable 3D spaces, with seemingly full object permanence. As opposed to other efforts I&#x27;ve seen which are basically briefly animating a 2D scene to make it seem vaguely 3D.</text></item></parent_chain><comment><author>comex</author><text>Google is hardly a good actor here. They just announced Gemini 1.5 along with a &quot;technical report&quot; [1] whose entire description of the model architecture is: &quot;Gemini 1.5 Pro is a sparse mixture-of-expert (MoE) Transformer-based model&quot;. Followed by a list of papers that it &quot;builds on&quot;, followed by a definition of MoE. I suppose that&#x27;s more than OpenAI gave in their GPT-4 technical report. But not by much!<p>[1] <a href="https:&#x2F;&#x2F;storage.googleapis.com&#x2F;deepmind-media&#x2F;gemini&#x2F;gemini_v1_5_report.pdf" rel="nofollow">https:&#x2F;&#x2F;storage.googleapis.com&#x2F;deepmind-media&#x2F;gemini&#x2F;gemini_...</a></text></comment> | <story><title>Sora: Creating video from text</title><url>https://openai.com/sora</url></story><parent_chain><item><author>gerash</author><text>When others create text to video systems (eg. Lumiere from Google) they publish the research (eg. <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;2401.12945.pdf" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;2401.12945.pdf</a>). Open AI is all about commercialization. I don&#x27;t like their attitude</text></item><item><author>crazygringo</author><text>This is insane. But I&#x27;m impressed most of all by the quality of <i>motion</i>. I&#x27;ve quite simply <i>never seen convincing computer-generated motion before</i>. Just look at the way the wooly mammoths connect with the ground, and their lumbering mass feels real.<p>Motion-capture works fine because that&#x27;s real motion, but every time people try to animate humans and animals, even in big-budget CGI movies, it&#x27;s always ultimately obviously fake. There are so many subtle things that happen in terms of acceleration and deceleration of all of the different parts of an organism, that no animator ever gets it 100% right. No animation <i>algorithm</i> gets it to a point where it&#x27;s believable, just where it&#x27;s &quot;less bad&quot;.<p>But these videos seem to be getting it entirely believable for both people and animals. Which is <i>wild</i>.<p>And then of course, not to mention that these are entirely believable 3D spaces, with seemingly full object permanence. As opposed to other efforts I&#x27;ve seen which are basically briefly animating a 2D scene to make it seem vaguely 3D.</text></item></parent_chain><comment><author>jstummbillig</author><text>Not to be overly cute, but if the cutting edge research you do is maybe changing the world fundamentally, forever, guarding that tech should be really, really, really far up your list of priorities and everyone else should be really happy about your priorities.<p>And that should probably take precedence over the semantics of your moniker, every single time (even if hn continues to be super sour about it)</text></comment> |
5,954,209 | 5,953,829 | 1 | 2 | 5,953,756 | train | <story><title>Memories of Stasi color Germans’ view of U.S. surveillance programs</title><url>http://www.mcclatchydc.com/2013/06/26/195045/memories-of-stasi-color-germans.html</url></story><parent_chain></parent_chain><comment><author>feral</author><text>The EU, which Germany is a major part of, passed the Data Retention Directive [0] in 2006.<p>That directive required every state in the EU to pass laws that all their citizens telecommunications metadata would be stored for at least 6 months, and often more, up to 2 years.<p>As I understand it, this means all &#x27;metadata&#x27; is required to be stored, including the source and destination for every phonecall and text message, including cell location for cellphones, and information mapping IPs to users for web+email (and perhaps also the source and destination of every e-mail; but I&#x27;m not certain about that?). I believe that the data is stored by service providers, and only passed to law enforcement in the context of a particular investigation <i>in theory</i> (in the Irish implementation, a court order is not required to access an individual&#x27;s records; a request from a high ranking law enforcement officer or tax official is enough). But its all collected and stored.<p>Maybe that&#x27;s a sufficiently big difference that warrants the EU retention laws not being mentioned in this article? But it seems to me that they should still be part of this narrative.<p>There have been challenges to the EU directive, and countries dragging their heels about implementing it. But, by and large, it is an established part of EU law.<p>I don&#x27;t know a lot about this area, but I feel that an understanding of existing European data retention laws seems to be missing from the coverage of the European reaction to the US data collection issues.<p>[0] <a href="http://en.wikipedia.org/wiki/Data_Retention_Directive" rel="nofollow">http:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Data_Retention_Directive</a></text></comment> | <story><title>Memories of Stasi color Germans’ view of U.S. surveillance programs</title><url>http://www.mcclatchydc.com/2013/06/26/195045/memories-of-stasi-color-germans.html</url></story><parent_chain></parent_chain><comment><author>griffordson</author><text>A choice quote:<p>Even Schmidt, 73, who headed one of the more infamous departments in the infamous Stasi, called himself appalled. The dark side to gathering such a broad, seemingly untargeted, amount of information is obvious, he said.<p>“It is the height of naivete to think that once collected this information won’t be used,” he said. “This is the nature of secret government organizations. The only way to protect the people’s privacy is not to allow the government to collect their information in the first place.”</text></comment> |
17,486,712 | 17,485,673 | 1 | 3 | 17,484,692 | train | <story><title>Intercooler.js – Simple AJAX Using HTML Attributes</title><url>http://intercoolerjs.org/</url></story><parent_chain></parent_chain><comment><author>e67f70028a46fba</author><text>i work on intercooler, you can AMA<p>as usual, a lot of negative comments from HN on the approach (which is admittedly against the grain) but I&#x27;ve used it for a relatively large and successful web app (rails back end) and it has worked out well compared to the SPA architectures I&#x27;ve had to deal with<p>i think the philosophical roots in REST&#x2F;HATEOAS are worth considering, if you are open minded:<p><a href="http:&#x2F;&#x2F;intercoolerjs.org&#x2F;2016&#x2F;01&#x2F;18&#x2F;rescuing-rest.html" rel="nofollow">http:&#x2F;&#x2F;intercoolerjs.org&#x2F;2016&#x2F;01&#x2F;18&#x2F;rescuing-rest.html</a><p><a href="http:&#x2F;&#x2F;intercoolerjs.org&#x2F;2016&#x2F;05&#x2F;08&#x2F;hatoeas-is-for-humans.html" rel="nofollow">http:&#x2F;&#x2F;intercoolerjs.org&#x2F;2016&#x2F;05&#x2F;08&#x2F;hatoeas-is-for-humans.ht...</a></text></comment> | <story><title>Intercooler.js – Simple AJAX Using HTML Attributes</title><url>http://intercoolerjs.org/</url></story><parent_chain></parent_chain><comment><author>slow_donkey</author><text>I tried this library before but Vue does pretty much the same thing and is way more powerful. Additionally, starting out with Vue makes transitioning to a SPA much easier when needed.<p>I also encountered some bugs when mixing the 2 during a transition phase that broke my requests. If I had to start over again I would only use Vue starting with the drop in JS library (no webpack&#x2F;preprocessing)</text></comment> |
17,860,305 | 17,859,303 | 1 | 3 | 17,858,842 | train | <story><title>T-Mobile Database Breach Exposes 2M Customers' Data</title><url>https://www.databreachtoday.com/t-mobile-database-breach-exposes-2-million-customers-data-a-11420</url></story><parent_chain><item><author>seibelj</author><text>I have T-Mobile. 6 weeks ago my phone could no longer access the cell network. The support agent told me that someone went into a store, claimed to be me, and was able to change the SIM card. The history showed the employee in the store verified me by my driver&#x27;s license. We changed the SIM back and supposedly locked the account.<p>I use Google Auth OTP for all the accounts that I can, and as far as I can tell nothing was breached or stolen, but I wouldn&#x27;t rely on your cell phone or number for anything whatsoever, it&#x27;s way too easy to socially engineer, or have some easily corruptible retail employee steal from you.</text></item></parent_chain><comment><author>rm999</author><text>T-mobile is horrible with security. They have this service called Digits, which lets users access their phone number from other devices. The problem is, this subverts the security model a lot of the American internet ecosystem is built on, i.e. your phone number will be secure online. Someone got into my t-mobile account and enabled Digits, then had free-reign on my gmail, texts, paypal, etc (I don&#x27;t use the strongest passwords, but always keep 2-factor authentication on important services like these). Every time I got hacked I&#x27;d spend a couple hours of trying to figure out how, and the T-mobile agents would always claim no one accessed my number. Finally, the third time, I escalated up the security chain within T-mobile and they figured out the issue.<p><a href="https:&#x2F;&#x2F;www.t-mobile.com&#x2F;offers&#x2F;t-mobile-digits" rel="nofollow">https:&#x2F;&#x2F;www.t-mobile.com&#x2F;offers&#x2F;t-mobile-digits</a></text></comment> | <story><title>T-Mobile Database Breach Exposes 2M Customers' Data</title><url>https://www.databreachtoday.com/t-mobile-database-breach-exposes-2-million-customers-data-a-11420</url></story><parent_chain><item><author>seibelj</author><text>I have T-Mobile. 6 weeks ago my phone could no longer access the cell network. The support agent told me that someone went into a store, claimed to be me, and was able to change the SIM card. The history showed the employee in the store verified me by my driver&#x27;s license. We changed the SIM back and supposedly locked the account.<p>I use Google Auth OTP for all the accounts that I can, and as far as I can tell nothing was breached or stolen, but I wouldn&#x27;t rely on your cell phone or number for anything whatsoever, it&#x27;s way too easy to socially engineer, or have some easily corruptible retail employee steal from you.</text></item></parent_chain><comment><author>hippich</author><text>Similar thing happened to me about a year ago. The creepiest part was - it happened while I was in the international flight and my gmail WAS on sms &quot;two factory&quot; authentication... Since then everywhere i can I used OTP, but some sites fallback to use your phone number if you can&#x27;t provide OTP password... So I have to enter some completely invalid phone number there to make it impossible.</text></comment> |
21,144,178 | 21,142,834 | 1 | 3 | 21,136,348 | train | <story><title>Schwab Removes U.S. Stock, ETF and Options Commissions</title><url>https://pressroom.aboutschwab.com/press-release/corporate-and-financial-news/conjunction-chuck-schwabs-new-book-invested-schwab-remove</url></story><parent_chain><item><author>StevePerkins</author><text>Similar story with Fidelity.<p>Honestly, I just don&#x27;t understand why people bother with traditional banks at this point. Pretty much all of the major brokers offer cash management accounts that are equivalent of online banking. Pretty much all with standard:<p>- No fees<p>- No minimum balance<p>- Free checkwriting (usually free physical checks for that matter!)<p>- ATM fee reimbursement (at least anywhere in the U.S., if not internationally too)<p>- Free debit cards<p>- Direct deposit<p>- Deposit paper checks by taking a picture with a phone app, etc etc etc.<p>The only possible downside that I can think of is that I&#x27;d have to deal with a little extra hassle if I wanted a cashier&#x27;s check. But even then, it would simply take an extra day or two. And it&#x27;s not like I&#x27;m closing on a new home purchase all that frequently.<p>Nevertheless, people that I talk to get weird and SCARED when I talk about it. Even my wife keeps a separate checking account at a physical bank, because she just likes knowing that a brick-and-mortar building is there. I don&#x27;t get it myself, but human nature can be odd when it comes to money.</text></item><item><author>whalesalad</author><text>As a side note I use Schwab as a checking account and it’s incredible. Zero fees. Reimbursement of ATM fees. Easy overdraft protection from the brokerage account. I love them as a bank.<p>This is even more reason to love them.<p>I’ve used Chase, Citi, was a Simple customer for a long time... and Schwab blows all of them out of the water.</text></item></parent_chain><comment><author>ggm</author><text><i>The only possible downside that I can think of is that I&#x27;d have to deal with a little extra hassle if I wanted a cashier&#x27;s check.</i><p>Almost everything wrong with consumer banking in america is contained, or stems from the ability to refer to cheque as a neccessary and normal thing in modern day finance.<p>I tore up my last cheque book ten years ago, I have had to resort to Western Union three times now (once a year) to send money to Her Majesties Customs and Excise for my UK pension topup, its the only thing I still do that way, and if the UK stopped using a closed-loop domestic clearing house and opened up SWIFT I&#x27;d do this via the IBAN immediately.<p>Cheques are cool. They&#x27;re also horrendously stone age and the &quot;yes but...&quot; are the collected history of insanity that is american banking.<p>Jimmy Stewart, holding the fort at the friendly credit union on christmas day...</text></comment> | <story><title>Schwab Removes U.S. Stock, ETF and Options Commissions</title><url>https://pressroom.aboutschwab.com/press-release/corporate-and-financial-news/conjunction-chuck-schwabs-new-book-invested-schwab-remove</url></story><parent_chain><item><author>StevePerkins</author><text>Similar story with Fidelity.<p>Honestly, I just don&#x27;t understand why people bother with traditional banks at this point. Pretty much all of the major brokers offer cash management accounts that are equivalent of online banking. Pretty much all with standard:<p>- No fees<p>- No minimum balance<p>- Free checkwriting (usually free physical checks for that matter!)<p>- ATM fee reimbursement (at least anywhere in the U.S., if not internationally too)<p>- Free debit cards<p>- Direct deposit<p>- Deposit paper checks by taking a picture with a phone app, etc etc etc.<p>The only possible downside that I can think of is that I&#x27;d have to deal with a little extra hassle if I wanted a cashier&#x27;s check. But even then, it would simply take an extra day or two. And it&#x27;s not like I&#x27;m closing on a new home purchase all that frequently.<p>Nevertheless, people that I talk to get weird and SCARED when I talk about it. Even my wife keeps a separate checking account at a physical bank, because she just likes knowing that a brick-and-mortar building is there. I don&#x27;t get it myself, but human nature can be odd when it comes to money.</text></item><item><author>whalesalad</author><text>As a side note I use Schwab as a checking account and it’s incredible. Zero fees. Reimbursement of ATM fees. Easy overdraft protection from the brokerage account. I love them as a bank.<p>This is even more reason to love them.<p>I’ve used Chase, Citi, was a Simple customer for a long time... and Schwab blows all of them out of the water.</text></item></parent_chain><comment><author>loeg</author><text>Typically, brokerages don&#x27;t give you any interest on checking. Now, banks haven&#x27;t really either since 2008, but prior to that you could make a few dollars a year on checking interest.<p><a href="https:&#x2F;&#x2F;www.kalzumeus.com&#x2F;2019&#x2F;6&#x2F;26&#x2F;how-brokerages-make-money&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.kalzumeus.com&#x2F;2019&#x2F;6&#x2F;26&#x2F;how-brokerages-make-mone...</a> is fascinating if you enjoy this kind of stuff.</text></comment> |
18,498,926 | 18,499,008 | 1 | 2 | 18,498,657 | train | <story><title>Ask HN: Why and how is Gartner Inc. still relevant?</title><text>I just came across one of Gartner&#x27;s most recent publications (predictions, trends, ...).<p>I&#x27;m finding it hard to distill through the fluff to find meaningful insights.
Their trends are dated and it feels as if someone has spilled the buzzword paint on a slide deck.<p>I&#x27;m experiencing first-hand the mess they have caused by suggesting an operational&#x2F;outsourcing model that makes it extremely hard to get anything done other than planning.<p>How are they still relevant?</text></story><parent_chain><item><author>dplgk</author><text>So basically Gartner is there to gobble up money from incompetence.</text></item><item><author>chomp</author><text>Is there a better corporate insights and research firm that you know of?<p>Their trends seem dated because they&#x27;re not going to feed enterprises bleeding edge tech trends; they&#x27;re going to go with trends that are mostly proven, so yes at that point they&#x27;re probably going to be a bit stale. Large enterprises with minimal tech advocacy&#x2F;leadership in the senior leadership team eat this stuff up because they don&#x27;t know much better.<p>Frankly, if your business is struggling with operations to a point where they are bringing in third parties to get told to outsource operations, I feel like your business has a competency problem that you can&#x27;t place squarely on the shoulders of Gartner.</text></item></parent_chain><comment><author>Kalium</author><text>You could look at it that way!<p>Some might describe Gartner as a company that dispenses reasonable advice on demand to companies who need guidance on subjects well outside their core competencies. This is very much making money from incompetence, but it&#x27;s the sort of profiting from incompetence that every expert adviser does.</text></comment> | <story><title>Ask HN: Why and how is Gartner Inc. still relevant?</title><text>I just came across one of Gartner&#x27;s most recent publications (predictions, trends, ...).<p>I&#x27;m finding it hard to distill through the fluff to find meaningful insights.
Their trends are dated and it feels as if someone has spilled the buzzword paint on a slide deck.<p>I&#x27;m experiencing first-hand the mess they have caused by suggesting an operational&#x2F;outsourcing model that makes it extremely hard to get anything done other than planning.<p>How are they still relevant?</text></story><parent_chain><item><author>dplgk</author><text>So basically Gartner is there to gobble up money from incompetence.</text></item><item><author>chomp</author><text>Is there a better corporate insights and research firm that you know of?<p>Their trends seem dated because they&#x27;re not going to feed enterprises bleeding edge tech trends; they&#x27;re going to go with trends that are mostly proven, so yes at that point they&#x27;re probably going to be a bit stale. Large enterprises with minimal tech advocacy&#x2F;leadership in the senior leadership team eat this stuff up because they don&#x27;t know much better.<p>Frankly, if your business is struggling with operations to a point where they are bringing in third parties to get told to outsource operations, I feel like your business has a competency problem that you can&#x27;t place squarely on the shoulders of Gartner.</text></item></parent_chain><comment><author>monkeydreams</author><text>In the same way that the supermarket is there to gobble money up from your inability to grow your own vegetables, mix up your own detergents, manufacture your own batteries, etc.</text></comment> |
9,314,409 | 9,314,402 | 1 | 3 | 9,314,269 | train | <story><title>Earth receiving mysterious radio signals</title><url>http://www.ibtimes.co.uk/earth-receiving-mysterious-radio-signals-outer-space-alien-life-source-possibility-believe-1494686</url></story><parent_chain></parent_chain><comment><author>dfryer</author><text>The paper is much more conservative than the article about it. They&#x27;re saying that the close relationship between the signal and the human-made second, plus the fact that most of these signals were observed at one particular telescope, suggests strongly that it&#x27;s a local, human source.<p>&quot;Paper doesn&#x27;t rule out aliens&quot; isn&#x27;t a particularly strong statement. I&#x27;ve read many papers that don&#x27;t rule out aliens, dragons OR unicorns.</text></comment> | <story><title>Earth receiving mysterious radio signals</title><url>http://www.ibtimes.co.uk/earth-receiving-mysterious-radio-signals-outer-space-alien-life-source-possibility-believe-1494686</url></story><parent_chain></parent_chain><comment><author>FatalLogic</author><text>Maybe this is a better link for this story? <a href="http:&#x2F;&#x2F;www.newscientist.com&#x2F;article&#x2F;mg22630153.600-is-this-et-mystery-of-strange-radio-bursts-from-space.html?full=true" rel="nofollow">http:&#x2F;&#x2F;www.newscientist.com&#x2F;article&#x2F;mg22630153.600-is-this-e...</a></text></comment> |
19,341,343 | 19,339,800 | 1 | 2 | 19,337,638 | train | <story><title>Galaxy Simulations Offer a New Solution to the Fermi Paradox</title><url>https://www.quantamagazine.org/galaxy-simulations-offer-a-new-solution-to-the-fermi-paradox-20190307/</url></story><parent_chain><item><author>lukifer</author><text>My favorite solution: The galaxy is teeming with life, but in the form of quasi-immortal software that lives at such vast timescales, they&#x27;re as uninterested in talking to us as we would be in talking to a fruit fly. (Imagine Space Ents that say hello, and are perfectly content to wait 10,000 years for a reply.) It may well be that becoming such a life form is the only practical way to deal with the speed-of-light constraint, either in travel or communication.</text></item></parent_chain><comment><author>burtonator</author><text>We don&#x27;t have conversations with paramecium.<p>If you wanted to build a Dyson sphere you would capture all the light from your star and might not exhibit ANY radiation.<p>We wouldn&#x27;t see such a species and if there&#x27;s such a small blink of time between the industrial age to the singularity the probability that there is an overlapping species is very very low.<p>We would only have about 20-40 years to actually have the technology to find them before we hit the singularity.<p>We have NO idea what happens after that point but my plan is to just chill with my wife in a virtual Yosemite park once we hit the singularity.</text></comment> | <story><title>Galaxy Simulations Offer a New Solution to the Fermi Paradox</title><url>https://www.quantamagazine.org/galaxy-simulations-offer-a-new-solution-to-the-fermi-paradox-20190307/</url></story><parent_chain><item><author>lukifer</author><text>My favorite solution: The galaxy is teeming with life, but in the form of quasi-immortal software that lives at such vast timescales, they&#x27;re as uninterested in talking to us as we would be in talking to a fruit fly. (Imagine Space Ents that say hello, and are perfectly content to wait 10,000 years for a reply.) It may well be that becoming such a life form is the only practical way to deal with the speed-of-light constraint, either in travel or communication.</text></item></parent_chain><comment><author>joe_the_user</author><text>To continue the speculation, such &quot;life&quot; may be what can travel between stars but life such as our may still be common. Indeed, life such as our may be common enough that interstellar life merely takes note of it and moves on since novelty wears off quickly.</text></comment> |
39,424,214 | 39,422,561 | 1 | 2 | 39,422,417 | train | <story><title>Ask HN: Best resources on starting a lifestyle business?</title><text>After working in RevOps and sales for VC-backed startups for a few years, I&#x27;m considering creating a SaaS business that helps solve problems for a particular niche industry that interests me.<p>What are some resources that are useful to look at while I consider embarking on this journey?<p>My experience in RevOps &amp; sales the last few years has been that so many influencers are trying to pitch you their guides and resources for a fee. It&#x27;s made me suspicious of pretty much anyone selling their insights and made it hard to determine which are of actual value.</text></story><parent_chain></parent_chain><comment><author>stanislavb</author><text>Not exactly a resource but more of an advice - learn software development. Unless you are familiar to some extent, you will always be dependent on a competent software engineer, and that will make it much harder to build a lifestyle business.<p>Second advice, use Ruby on Rails or Django as a web framework and don&#x27;t follow after the complexities of JS and other newcomers. Phoenix (Elixir) is awesome but that will require much more efforts and expertise compared to the established and time-tested Rails and Django.<p>Good luck!<p>p.s. I&#x27;m writing this as a person managing a lifestyle business (LibHunt &amp; SaaSHub).<p>p.p.s. I&#x27;d also recommend getting into the IndieHackers community for inspiration and shared experience.</text></comment> | <story><title>Ask HN: Best resources on starting a lifestyle business?</title><text>After working in RevOps and sales for VC-backed startups for a few years, I&#x27;m considering creating a SaaS business that helps solve problems for a particular niche industry that interests me.<p>What are some resources that are useful to look at while I consider embarking on this journey?<p>My experience in RevOps &amp; sales the last few years has been that so many influencers are trying to pitch you their guides and resources for a fee. It&#x27;s made me suspicious of pretty much anyone selling their insights and made it hard to determine which are of actual value.</text></story><parent_chain></parent_chain><comment><author>hermitcrab</author><text>I&#x27;ve been running a bootstrapped, lifestyle software business since 2005. I&#x27;ve linked to some articles you might find relevant here (for free!):
<a href="https:&#x2F;&#x2F;successfulsoftware.net&#x2F;starting-a-microisv&#x2F;" rel="nofollow">https:&#x2F;&#x2F;successfulsoftware.net&#x2F;starting-a-microisv&#x2F;</a></text></comment> |
35,114,345 | 35,111,937 | 1 | 2 | 35,110,974 | train | <story><title>Repairing a tiny ribbon cable inside a 28 year old IBM ThinkPad 701c</title><url>https://blog.jgc.org/2023/03/repairing-tiny-ribbon-cable-inside-28.html</url></story><parent_chain></parent_chain><comment><author>jacquesm</author><text>I make repairs to cables like these with a very simple method: glue the two pieces together with superglue and some overlap, a few mm will do. Then use conducting glue (with silver particles in it) to restore the connection. Use a tiny brush and a microscope to apply the glue. If the currents are low (as they usually are with cables like these) this will work just fine and give a quick and long lasting result.</text></comment> | <story><title>Repairing a tiny ribbon cable inside a 28 year old IBM ThinkPad 701c</title><url>https://blog.jgc.org/2023/03/repairing-tiny-ribbon-cable-inside-28.html</url></story><parent_chain></parent_chain><comment><author>somat</author><text>A deserved &quot;Well done&quot; to the author, a good end to a tricky job. I wonder if it is possible to solder the two halves of a flex circuit directly together? That is, is it possible without the junction wires?<p>Scrape the top of one, the bottom of the other, apply solder pads to both sides then align the two and reflow the solder.</text></comment> |
32,296,052 | 32,296,032 | 1 | 2 | 32,295,766 | train | <story><title>Riding transit takes almost twice as long as driving (2017)</title><url>https://www.governing.com/archive/gov-transit-driving-times.html</url></story><parent_chain><item><author>bonyt</author><text>&gt; For New York metro residents who take public transportation, a door-to-door commute averages about 51 minutes. That’s much longer than the 29 minutes typically spent by those who drive alone.<p>I mean, is this really an apples to apples comparison that is at all meaningful? The NY metro area includes NYC with its subway, but also suburbs in long island, New Jersey, and Connecticut with long commuter rail trips where you go to a different city, and short car commutes where you don’t.</text></item></parent_chain><comment><author>panarky</author><text>Also, the average bicycle trip is shorter than the average airplane trip, therefore bikes are faster than airplanes.</text></comment> | <story><title>Riding transit takes almost twice as long as driving (2017)</title><url>https://www.governing.com/archive/gov-transit-driving-times.html</url></story><parent_chain><item><author>bonyt</author><text>&gt; For New York metro residents who take public transportation, a door-to-door commute averages about 51 minutes. That’s much longer than the 29 minutes typically spent by those who drive alone.<p>I mean, is this really an apples to apples comparison that is at all meaningful? The NY metro area includes NYC with its subway, but also suburbs in long island, New Jersey, and Connecticut with long commuter rail trips where you go to a different city, and short car commutes where you don’t.</text></item></parent_chain><comment><author>donohoe</author><text>This.<p>I came here to say this. When you bundle NYC (which is one of the largest systems in itself between subway and bus) that represents a skewed average.<p>As a NYC resident of 20 years I can assure you public transportation much much faster than a car in most cases.<p>Don&#x27;t even get me started about the time it takes to find parking in this city... I am guessing that that time consideration is not in this study.</text></comment> |
31,572,804 | 31,571,979 | 1 | 3 | 31,567,345 | train | <story><title>Ask HN: Is there an “uncanny valley” effect with startup MVPs?</title><text>I am asking because I have noticed that as I build better landing pages for my MVPs I get judged more harshly.
This got me thinking if there is a way in which having a not particularly attractive landing page might get people to approach the product in a different way.</text></story><parent_chain><item><author>naet</author><text>Checking your actual landing page ( <a href="https:&#x2F;&#x2F;propertywebbuilder.com" rel="nofollow">https:&#x2F;&#x2F;propertywebbuilder.com</a> ) and my opinion is... that it isn&#x27;t a very good landing page. The sizing and spacing are awkward and blocky, your color scheme clashes with your images, the images are generic and tacky, everything is too big and wide, etc. Maybe because you worked on it you feel that it is a great modern style landing page, but that isn&#x27;t how it reads to me (and seemingly to other people as well).<p>Maybe you look at a SIMPLER site like <a href="https:&#x2F;&#x2F;tom.preston-werner.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;tom.preston-werner.com&#x2F;</a> and mistakenly equate that with a &quot;worse&quot; landing page. Actually though this simple page has bullet proof styling, and slapping some stock images on it like you did on your page wouldn&#x27;t make it better, it would make it way worse.<p>Rather than looking for an external factor like a &quot;mythical uncanny valley&quot; to explain your landing page&#x27;s poor performance, look inwards at your individual styling and what could be improved.</text></item></parent_chain><comment><author>pantulis</author><text>To be fair with OP, comparing Preston-Werner&#x27;s blog to a landing page is not really an apples to apples comparison, imho, as the purpose is totally different.<p>Still, OP has lots of work to do on that landing page, but not only on the design (there are a gazillion templates for that) but on how does it convey the value of the service it&#x27;s offering. Also almost all clicks take inmediately to a login page without much explanation.</text></comment> | <story><title>Ask HN: Is there an “uncanny valley” effect with startup MVPs?</title><text>I am asking because I have noticed that as I build better landing pages for my MVPs I get judged more harshly.
This got me thinking if there is a way in which having a not particularly attractive landing page might get people to approach the product in a different way.</text></story><parent_chain><item><author>naet</author><text>Checking your actual landing page ( <a href="https:&#x2F;&#x2F;propertywebbuilder.com" rel="nofollow">https:&#x2F;&#x2F;propertywebbuilder.com</a> ) and my opinion is... that it isn&#x27;t a very good landing page. The sizing and spacing are awkward and blocky, your color scheme clashes with your images, the images are generic and tacky, everything is too big and wide, etc. Maybe because you worked on it you feel that it is a great modern style landing page, but that isn&#x27;t how it reads to me (and seemingly to other people as well).<p>Maybe you look at a SIMPLER site like <a href="https:&#x2F;&#x2F;tom.preston-werner.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;tom.preston-werner.com&#x2F;</a> and mistakenly equate that with a &quot;worse&quot; landing page. Actually though this simple page has bullet proof styling, and slapping some stock images on it like you did on your page wouldn&#x27;t make it better, it would make it way worse.<p>Rather than looking for an external factor like a &quot;mythical uncanny valley&quot; to explain your landing page&#x27;s poor performance, look inwards at your individual styling and what could be improved.</text></item></parent_chain><comment><author>yupper32</author><text>Oh jeez, yeah that first site looks like an auto-generated scam&#x2F;spam website. I&#x27;d never enter anything into that website.</text></comment> |
16,783,563 | 16,782,757 | 1 | 3 | 16,782,051 | train | <story><title>At 88, doctor pursues a long-ignored treatment for strokes, heart attacks</title><url>https://www.statnews.com/2018/04/06/victor-gurewich-harvard-stroke-heart-attack/</url></story><parent_chain></parent_chain><comment><author>arkades</author><text>What this story leaves out is why we are moving away from thrombolytics - besides them kinda sucking.<p>Once a clot has matured they tend to be useless, which is part of why there’s a solid upper cap on time to administration. You cross a line where any clot they’re likely to bust isn’t one you want to bust. So, a window of a couple of hours. Urokinase doesn’t change that, to my knowledge.<p>Surgical removal of the clot, on the other hand, has recently been shown to offer enormous benefit to at-risk tissue (tissue not yet dead but in the watershed area) 24 hours later, without risk of hitting the wrong clot. It’s no miracle either, but evidence is piling up that it’s got more Pros and fewer Cons.<p>Thus, no one really following armchair hypotheses (one old non RCT trial not withstanding) about combining two increasingly undesirable drugs.<p>I respect the guy’s thought and effort, and have to admit he might be right about their combined effectiveness, but this article leaves out a lot of the context as to -why- this isn’t getting attention. Writing a hagiography-by-omission is below what I’ve come to expect from statnews.</text></comment> | <story><title>At 88, doctor pursues a long-ignored treatment for strokes, heart attacks</title><url>https://www.statnews.com/2018/04/06/victor-gurewich-harvard-stroke-heart-attack/</url></story><parent_chain></parent_chain><comment><author>helloworld</author><text>For me, it was the persistence of the researcher -- well into his ninth decade -- that caught my eye, rather than the somewhat arcane details of the thrombolytic therapy that he&#x27;s been pursuing.<p>A strong sense of purpose seems to be a factor in health and longevity. Is it just good fortune that allows some people to develop that sense of purpose, or can it be consciously and effectively nurtured?</text></comment> |
3,765,012 | 3,764,468 | 1 | 2 | 3,764,350 | train | <story><title>Chrome addons hacking: Bye Bye AdBlock filters</title><url>http://blog.kotowicz.net/2012/03/chrome-addons-hacking-bye-bye-adblock.html?spref=tw</url></story><parent_chain></parent_chain><comment><author>cmelbye</author><text>His tone when he talks about extensions modifying the DOM of random sites, injecting CSS, etc was funny; almost as if they have some god-given right to control exactly how their site will appear on my computer. As long as I control the browser software, it's always going to be possible for me to choose how web sites are displayed on my computer.</text></comment> | <story><title>Chrome addons hacking: Bye Bye AdBlock filters</title><url>http://blog.kotowicz.net/2012/03/chrome-addons-hacking-bye-bye-adblock.html?spref=tw</url></story><parent_chain></parent_chain><comment><author>overshard</author><text>Bye bye scripting with noscript and scriptno. This is nifty but if people don't want ads people won't have ads. As someone who makes most of his money off of ads I understand the want to stop ad blockers but I rather just provided a good, clean experience and when an ad is blocked put up a nice message saying how I make my money off of ads and to please disable ad block on my website.<p>This method seems intrusive to me and like an "Ah ha! I stopped your ad blocker from working! TAKE THAT!"</text></comment> |
16,865,725 | 16,865,806 | 1 | 2 | 16,862,925 | train | <story><title>Facebook Container for Firefox</title><url>https://www.mozilla.org/en-US/firefox/facebookcontainer</url></story><parent_chain><item><author>cornholio</author><text>Third party cookies, and any way to fingerprint a specific user starting from high entropy user agents to screen resolution, font fingerprinting or canvas data, should be considered a breach of the browser security model.<p>All sites should run in containers and no advertiser should be able to track you across sessions. When I want 3rd party interaction, I should need to opt in and connect the current site with Facebook or some other 3rd party.</text></item></parent_chain><comment><author>mih</author><text>Ever encountered Google ReCaptcha when you&#x27;ve turned off third party cookies or while in incognito&#x2F;private mode? It&#x27;s a nightmare, even if you&#x27;re logged into a Google account. You can be shown upto 7-8 challenges, painfully slow loading images and Google&#x27;s insistence that they encountered malicious traffic from your IP when attempting to use the audio version. What&#x27;s even worse is they track your mouse movements and fast solvers like me are penalized since they think I&#x27;m not human anymore.<p>As much as I hate third party cookies, turning them on drastically simplifies the captcha solving process. So much so, I now use a separate browser profile with third party cookies allowed, just for the sake of captcha heavy sites.<p>Given that Google benefits by tracking my activities and free labor from my captcha solving, they will always punish privacy conscious users via such dark patterns.</text></comment> | <story><title>Facebook Container for Firefox</title><url>https://www.mozilla.org/en-US/firefox/facebookcontainer</url></story><parent_chain><item><author>cornholio</author><text>Third party cookies, and any way to fingerprint a specific user starting from high entropy user agents to screen resolution, font fingerprinting or canvas data, should be considered a breach of the browser security model.<p>All sites should run in containers and no advertiser should be able to track you across sessions. When I want 3rd party interaction, I should need to opt in and connect the current site with Facebook or some other 3rd party.</text></item></parent_chain><comment><author>simias</author><text>The security model of the internet is whack. Javascript should be opt-in, especially from 3rd party sources. &quot;This website wants&#x2F;requires javascript for an enhanced viewing experience [allow&#x2F;deny]&quot;. Instead not only will any stock browser gladly run anything thrown at it, it will also accept any cookies and now run WebGL code, trigger DRM engines and various other things. All that because the engineers who wrote the standards had enough hubris to think that a turing complete language with access to such a large API surface (from the DOM to the GPU to threads to a lot of the browser state) could be successfully sandboxed.<p>The web is a lost cause as far I&#x27;m concerned, it was under-engineered at the start and as a result people kept pilling mounds of crap on top to make up for it. What started as a way to display basic markup and images has been arm wrestled into handling complex web applications and videogames. The thing is so complex that it would probably take me less time to write a basic operating system and run Firefox on it than implementing my own browser from scratch on Linux. And despite all of this you end up having to use a billion of 3rd party javascript libraries to do anything remotely complex in the browser, things that have been standard in Qt or GTK since forever with a fraction of the code base and memory footprint.<p>And now people seem to enjoy reimplementing all the internet protocols on top of HTTP. The world has gone mad. I&#x27;m going to get a &quot;gopher should&#x27;ve won&quot; tatoo and live in as an hermit on some nearby mountain.<p>I&#x27;m glad that there appears to be a lot of discussion surrounding internet privacy lately, including in the mainstream. However I think that focusing all the discussion of Facebook is becoming counter-productive. Facebook exploited the weaknesses of the web very successfully but it only exploited tools that existed long before it was created. I have absolutely no illusion that Google &amp; friends are doing pretty much the same thing, and if you have an Android phone the amount of info Google can harvest is nothing short of terrifying when you think about it from an Orwellian perspective. Don&#x27;t miss the forest for the Facetree.<p>Assuming that internet giants will adhere to some &quot;code of conduct&quot; is naive at best. &quot;Do no evil&quot;, yeah right. We let them have to tools to do these things, we need to pry some of them away from their hands through technology and regulation (probably in that order).</text></comment> |
4,526,639 | 4,526,671 | 1 | 2 | 4,526,243 | train | <story><title>The Pirate Bay - 9 years and still bloody runnin</title><url>https://thepiratebay.se/blog/222</url></story><parent_chain><item><author>klrr</author><text>I don't support illegal activity, but I believe this is a quest of freedom, not for getting stuff free as in free beer.</text></item></parent_chain><comment><author>why-el</author><text>I don't agree. Sometimes it makes a lot of sense to support illegal activity (It's too trivial to site examples from US history).</text></comment> | <story><title>The Pirate Bay - 9 years and still bloody runnin</title><url>https://thepiratebay.se/blog/222</url></story><parent_chain><item><author>klrr</author><text>I don't support illegal activity, but I believe this is a quest of freedom, not for getting stuff free as in free beer.</text></item></parent_chain><comment><author>lumberjack</author><text>I disagree. Free software projects like the GNU foundation or the OpenBSD foundation are striving towards freedom.<p>ThePirateBay.se et al are simply filling in on market demand created by the utter absurdity of treating digital data as if it was a can of tomatoes or a bag of crisps.</text></comment> |
17,681,167 | 17,681,412 | 1 | 2 | 17,680,833 | train | <story><title>Google Maps is no longer a flat map</title><url>https://techcrunch.com/2018/08/03/google-maps-is-no-longer-flatearth/</url></story><parent_chain><item><author>Bud</author><text>&quot;On flat maps, it’s impossible to represent land mass size on a relative scale.&quot;<p>Um, no it&#x27;s not. TechCrunch needs better editors. You just need to use a non-Mercator projection.<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Gall%E2%80%93Peters_projection" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Gall%E2%80%93Peters_projection</a></text></item></parent_chain><comment><author>astrodust</author><text>You can represent size <i>or</i> shape correctly, not both.<p>Anyone who uses the Gall-Peters projection is a cop.</text></comment> | <story><title>Google Maps is no longer a flat map</title><url>https://techcrunch.com/2018/08/03/google-maps-is-no-longer-flatearth/</url></story><parent_chain><item><author>Bud</author><text>&quot;On flat maps, it’s impossible to represent land mass size on a relative scale.&quot;<p>Um, no it&#x27;s not. TechCrunch needs better editors. You just need to use a non-Mercator projection.<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Gall%E2%80%93Peters_projection" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Gall%E2%80%93Peters_projection</a></text></item></parent_chain><comment><author>jbob2000</author><text>That&#x27;s still not an accurate map. 3D maps don&#x27;t have projection issues, they can have near perfect representation of the world.</text></comment> |
31,488,326 | 31,487,912 | 1 | 3 | 31,486,590 | train | <story><title>I made an AI write a story about AI then I made another AI illustrate it</title><url>https://tristrumtuttle.medium.com/the-day-i-became-friends-with-a-robot-86433f707fad</url></story><parent_chain><item><author>turtledove</author><text>How many attempts and how much human curation was involved. Every &quot;I made AI do...&quot; story seems to necessarily omit the &quot;and I filtered out a LOT of rubbish before picking this one good example.&quot;</text></item></parent_chain><comment><author>goodside</author><text>The following is an un-curated text-davinci-002 generation, completing the text:<p>&gt; Write an insightful reply to this Hacker News comment on a link to an article titled &quot;I made an AI write a story about AI then I made another AI illustrate it&quot;. The comment is left by user &quot;turtledove&quot;, and reads, &quot;How many attempts and how much human curation was involved. Every &#x27;I made AI do...&#x27; story seems to necessarily omit the &#x27;and I filtered out a LOT of rubbish before picking this one good example.&#x27;&quot;<p>The completion:<p>&gt; I&#x27;m not sure how many attempts were made, but I imagine it wasn&#x27;t too many. As for human curation, I&#x27;m sure there was some involved in choosing which story and illustration to publish. But overall, I think the point of the article is that the AI was able to produce something fairly coherent on its own.</text></comment> | <story><title>I made an AI write a story about AI then I made another AI illustrate it</title><url>https://tristrumtuttle.medium.com/the-day-i-became-friends-with-a-robot-86433f707fad</url></story><parent_chain><item><author>turtledove</author><text>How many attempts and how much human curation was involved. Every &quot;I made AI do...&quot; story seems to necessarily omit the &quot;and I filtered out a LOT of rubbish before picking this one good example.&quot;</text></item></parent_chain><comment><author>qgin</author><text>That could be the job description of the editor of the New Yorker</text></comment> |
21,465,990 | 21,464,866 | 1 | 2 | 21,462,120 | train | <story><title>Uber’s self-driving car could not detect pedestrians outside of a crosswalk</title><url>https://www.theregister.co.uk/2019/11/06/uber_self_driving_car_death/</url></story><parent_chain><item><author>Strilanc</author><text>&gt; <i>5.6 seconds before impact, it classified her as a vehicle. Then it changed its mind to “other,” then to vehicle again, back to “other,” then to bicycle, then to “other” again, and finally back to bicycle.</i><p>System can&#x27;t decide what&#x27;s happening.<p>&gt; <i>It wasn’t until 1.2 seconds before the impact that the system recognized that the SUV was going to hit Herzberg</i><p>System is too slow to realize something serious is happening.<p>&gt; <i>That triggered what Uber called “action suppression,” in which the system held off braking for one second</i><p>A hardcoded 1 second delay during a potential emergency situation. Horrifying.<p>I bet they added it because the system kept randomly thinking something serious was going to happen for a few milliseconds when everything was going fine. If you ever find yourself doing that for a safety critical piece of software, you should stop and reconsider what you are doing. This is a hacky patch over a serious underlying classification issue. You need to fix the underlying problem, not hackily patch over it.<p>How is this not the title of the story? This is so much worse than the &quot;it couldn&#x27;t see her as a person, only as a bicycle&quot;. At least the car would still try to avoid a bicycle, in principle, instead of blindly gliding into it while hoping for the best.<p>&gt; <i>with 0.2 seconds left before impact, the car sounded an audio alarm, and Vasquez took the steering wheel, disengaging the autonomous system. Nearly a full second after striking Herzberg, Vasquez hit the brakes.</i><p>And then top it off with systemic issues around the backup driver not actually being ready to react.</text></item></parent_chain><comment><author>chooseaname</author><text>&gt;&gt; 5.6 seconds before impact, it classified her as a vehicle. Then it changed its mind to “other,” then to vehicle again, back to “other,” then to bicycle, then to “other” again, and finally back to bicycle.<p>The system should have started applying brake at this point. If a 3500lb vehicle can&#x27;t decide what it is about to impact, it needs to slow down (to a stop if necessary).<p>&gt; That triggered what Uber called “action suppression,” in which the system held off braking for one second<p>This is borderline criminal negligence.<p>&gt; with 0.2 seconds left before impact, the car sounded an audio alarm, and Vasquez took the steering wheel, disengaging the autonomous system. Nearly a full second after striking Herzberg, Vasquez hit the brakes.<p>Why were there no alarms going off at 5.6 seconds when the vehicle was confused!!!??<p>SMH. This is just ... I&#x27;m flabbergasted.</text></comment> | <story><title>Uber’s self-driving car could not detect pedestrians outside of a crosswalk</title><url>https://www.theregister.co.uk/2019/11/06/uber_self_driving_car_death/</url></story><parent_chain><item><author>Strilanc</author><text>&gt; <i>5.6 seconds before impact, it classified her as a vehicle. Then it changed its mind to “other,” then to vehicle again, back to “other,” then to bicycle, then to “other” again, and finally back to bicycle.</i><p>System can&#x27;t decide what&#x27;s happening.<p>&gt; <i>It wasn’t until 1.2 seconds before the impact that the system recognized that the SUV was going to hit Herzberg</i><p>System is too slow to realize something serious is happening.<p>&gt; <i>That triggered what Uber called “action suppression,” in which the system held off braking for one second</i><p>A hardcoded 1 second delay during a potential emergency situation. Horrifying.<p>I bet they added it because the system kept randomly thinking something serious was going to happen for a few milliseconds when everything was going fine. If you ever find yourself doing that for a safety critical piece of software, you should stop and reconsider what you are doing. This is a hacky patch over a serious underlying classification issue. You need to fix the underlying problem, not hackily patch over it.<p>How is this not the title of the story? This is so much worse than the &quot;it couldn&#x27;t see her as a person, only as a bicycle&quot;. At least the car would still try to avoid a bicycle, in principle, instead of blindly gliding into it while hoping for the best.<p>&gt; <i>with 0.2 seconds left before impact, the car sounded an audio alarm, and Vasquez took the steering wheel, disengaging the autonomous system. Nearly a full second after striking Herzberg, Vasquez hit the brakes.</i><p>And then top it off with systemic issues around the backup driver not actually being ready to react.</text></item></parent_chain><comment><author>Strilanc</author><text>To rant a bit more about this one second delay thing.<p>This reeks of a type of thinking where you are relying on other parts of the system to compensate. You might expect to hear things like &quot;it&#x27;s okay, the safety driver will catch it&quot;. Speaking for myself personally, this type of thinking comes very naturally. I like to come up with little proofs that a problem is handled by one part of a program, and then forget about that problem. But in my experience (which does not involve writing anything safety critical) this strategy kinda sucks at getting things right. Dependencies change, assumptions become invalid, holes in your intuitive proof become apparent, etc, etc, etc, and the thing falls over.<p>If you are designing a safety critical system, something you <i>really</i> want to work, I don&#x27;t think you should be thinking in the mode where each problem is handled by one part of the system. You need to be thinking in terms of defense in depth. When something goes wrong, <i>many</i> parts of the system should <i>all</i> be able to detect and correct the problem. And then when something bad does come up, and 9 out of 10 of those defensive layers each individually were sufficient to save the day so there was no disaster, you should go figure out what the hell went wrong in the tenth.</text></comment> |
17,855,909 | 17,855,520 | 1 | 2 | 17,855,104 | train | <story><title>Pandoc</title><url>https://pandoc.org/</url></story><parent_chain><item><author>koolba</author><text>My favorite pandoc hack is using it to convert word docs into markdown which can then be diffed similar to source code. Works great for legal redlining.</text></item></parent_chain><comment><author>d99kris</author><text>Agree, and it can be nicely integrated with Git:
<a href="http:&#x2F;&#x2F;blog.martinfenner.org&#x2F;2014&#x2F;08&#x2F;25&#x2F;using-microsoft-word-with-git&#x2F;" rel="nofollow">http:&#x2F;&#x2F;blog.martinfenner.org&#x2F;2014&#x2F;08&#x2F;25&#x2F;using-microsoft-word...</a></text></comment> | <story><title>Pandoc</title><url>https://pandoc.org/</url></story><parent_chain><item><author>koolba</author><text>My favorite pandoc hack is using it to convert word docs into markdown which can then be diffed similar to source code. Works great for legal redlining.</text></item></parent_chain><comment><author>flatline</author><text>Do Word’s native diff features not work for you?</text></comment> |
3,134,736 | 3,134,764 | 1 | 2 | 3,134,240 | train | <story><title>Immigrant entrepreneur arrested to meet quotas</title><url>http://www.tenthcave.com/2011/10/plea-for-help.html</url></story><parent_chain><item><author>rdtsc</author><text>An interesting factoid. There is no legal way for anyone to just come to US and make it their home country.<p>Say someone from Africa or South America. They can't simply save enough money, get on a plane, land at JFK and then after 10 years or whatever time become citizens. They would have to be persecuted in a terrible way before being granted asylum, have to marry a US citizen, find a company that would hire them.<p>I think that is very interesting given that this country was founded on immigrants just hoping on the boat and coming here, either to build a better life or because things got terribly bad where they were.<p>I personally went through the whole process and it is a fucking mess. I knew how to deal with it, because I dealt with bureaucracy in the old country. I can sense an annoyed low level clerk's power trip coming on before they notice it. Such individual need special treatment. You never want to fall into anyone's cross-hairs, it seems like they would destroy you just because they can.<p>Also know someone who works for immigration. It is quite a dysfunctional entity. Very inefficient. Lots of internal mistakes, unpaid interns do the job searching through databases and picking people out for violations &#38; deportation. Next time you know someone was picked out -- chances are some kid thought their last name sounded funny and it caught their attention.</text></item></parent_chain><comment><author>shin_lao</author><text>If you invest at least $ 1,000,000 in a company, you can get a visa. This gets down to $ 500,000 for some areas. The investment must create or maintain at least 10 jobs.<p>It's the EB-5 visa.</text></comment> | <story><title>Immigrant entrepreneur arrested to meet quotas</title><url>http://www.tenthcave.com/2011/10/plea-for-help.html</url></story><parent_chain><item><author>rdtsc</author><text>An interesting factoid. There is no legal way for anyone to just come to US and make it their home country.<p>Say someone from Africa or South America. They can't simply save enough money, get on a plane, land at JFK and then after 10 years or whatever time become citizens. They would have to be persecuted in a terrible way before being granted asylum, have to marry a US citizen, find a company that would hire them.<p>I think that is very interesting given that this country was founded on immigrants just hoping on the boat and coming here, either to build a better life or because things got terribly bad where they were.<p>I personally went through the whole process and it is a fucking mess. I knew how to deal with it, because I dealt with bureaucracy in the old country. I can sense an annoyed low level clerk's power trip coming on before they notice it. Such individual need special treatment. You never want to fall into anyone's cross-hairs, it seems like they would destroy you just because they can.<p>Also know someone who works for immigration. It is quite a dysfunctional entity. Very inefficient. Lots of internal mistakes, unpaid interns do the job searching through databases and picking people out for violations &#38; deportation. Next time you know someone was picked out -- chances are some kid thought their last name sounded funny and it caught their attention.</text></item></parent_chain><comment><author>blahedo</author><text>&#62; <i>There is no legal way for anyone to just come to US and make it their home country.</i><p>Are there countries where this is possible? This is a real question, I'm not snarking---is there some country X where I could just up and say, "I want to move to X" and just go there and make it my permanent residence? (I know that there are some countries that <i>in practice</i> don't <i>enforce</i> their requirements, but I thought all countries had them.)</text></comment> |
20,594,305 | 20,594,344 | 1 | 2 | 20,593,563 | train | <story><title>‘Just four dudes’: Inside EasyList, a community-run ad-blocking list</title><url>https://digiday.com/media/just-four-dudes-inside-easylist-community-run-adblocking-list-disrupting-internet/</url></story><parent_chain><item><author>GordonS</author><text>I had an issue with this recently, where buttons simply linking to our Twitter and Facebook pages were being blocked.<p>I&#x27;m no fan of obtrusive, JavaScript laden ads, and have used adblockers myself for years, but why block &quot;legitimate&quot; content?<p>I didn&#x27;t know how to get removed from the list (and assumed it wouldn&#x27;t happen anyway), so changed a couple of class names and it was working again. But I don&#x27;t know how long the issue was there before I realised it :&#x2F;</text></item><item><author>patd</author><text>I did run into the same issue as mentioned in the article and the Easylist forum is full of publishers having complaining that their site doesn&#x27;t work anymore.<p>I started working on <a href="http:&#x2F;&#x2F;blockedby.com" rel="nofollow">http:&#x2F;&#x2F;blockedby.com</a> , a monitoring tool to warn you when a rule would impact your website.<p>But after contacting a few websites that did have the issue on the forum, I couldn&#x27;t find any that actually wanted to prevent this. A few discussions with publishers told me that they&#x27;re more interested in convincing their users to give up on ad blockers rather than face the fact that people are not going to abandon ad blockers.</text></item></parent_chain><comment><author>Kalium</author><text>Depending on how they&#x27;re implemented, buttons linking to your Twitter and Facebook pages can double as trackers for Twitter and Facebook.<p>Anyway. People rarely intend to block legitimate content. At the same time, it&#x27;s not always easy to tell what is and isn&#x27;t legitimate when you&#x27;re writing fairly general rules.</text></comment> | <story><title>‘Just four dudes’: Inside EasyList, a community-run ad-blocking list</title><url>https://digiday.com/media/just-four-dudes-inside-easylist-community-run-adblocking-list-disrupting-internet/</url></story><parent_chain><item><author>GordonS</author><text>I had an issue with this recently, where buttons simply linking to our Twitter and Facebook pages were being blocked.<p>I&#x27;m no fan of obtrusive, JavaScript laden ads, and have used adblockers myself for years, but why block &quot;legitimate&quot; content?<p>I didn&#x27;t know how to get removed from the list (and assumed it wouldn&#x27;t happen anyway), so changed a couple of class names and it was working again. But I don&#x27;t know how long the issue was there before I realised it :&#x2F;</text></item><item><author>patd</author><text>I did run into the same issue as mentioned in the article and the Easylist forum is full of publishers having complaining that their site doesn&#x27;t work anymore.<p>I started working on <a href="http:&#x2F;&#x2F;blockedby.com" rel="nofollow">http:&#x2F;&#x2F;blockedby.com</a> , a monitoring tool to warn you when a rule would impact your website.<p>But after contacting a few websites that did have the issue on the forum, I couldn&#x27;t find any that actually wanted to prevent this. A few discussions with publishers told me that they&#x27;re more interested in convincing their users to give up on ad blockers rather than face the fact that people are not going to abandon ad blockers.</text></item></parent_chain><comment><author>nodja</author><text>&gt; buttons simply linking to our Twitter and Facebook pages were being blocked<p>Was everything hosted on your domain there? If they have images hosted by twitter&#x2F;facebook that&#x27;s enough for them to track users which is what these lists don&#x27;t like.</text></comment> |
3,832,884 | 3,832,906 | 1 | 2 | 3,832,575 | train | <story><title>Why you shouldn't invite Yehuda Katz to your user group meeting</title><url>http://thewebivore.com/why-you-shouldnt-invite-yehuda-katz-to-your-u</url></story><parent_chain><item><author>tmorton</author><text>Wow. A post about etiquette that calls someone out by name, with an insulting headline?<p>Face-to-face is the right channel for this sort of pointed criticism. Failing that, email might work. To pass the lesson along to the larger community, leave the specific names out of it - talk about "a big name in the community" who behaved badly. People who were there will know who you're talking about, but it doesn't become part of the internet paper trail.</text></item></parent_chain><comment><author>tomelders</author><text>I don't know. I'm not coming down on either side, but I can see the point that this Katz fellow didn't extend the same courtesy to the poor speaker. An embarrassing taste of their own medicine might do the trick.<p>We can't be nice to people who aren't nice all the time.</text></comment> | <story><title>Why you shouldn't invite Yehuda Katz to your user group meeting</title><url>http://thewebivore.com/why-you-shouldnt-invite-yehuda-katz-to-your-u</url></story><parent_chain><item><author>tmorton</author><text>Wow. A post about etiquette that calls someone out by name, with an insulting headline?<p>Face-to-face is the right channel for this sort of pointed criticism. Failing that, email might work. To pass the lesson along to the larger community, leave the specific names out of it - talk about "a big name in the community" who behaved badly. People who were there will know who you're talking about, but it doesn't become part of the internet paper trail.</text></item></parent_chain><comment><author>shawnc</author><text>I agree with your point about face=to-face or direct communication first. Barring that not accomplishing anything however, I'm not so sure I agree with the last part anymore. I used to, fully. However, after years of watching this 'accepted etiquette' go on and on, i've realized the ones who've done the stupid thing, keep on doing it - with new people none the wiser.<p>Why not call someone out by name? This person seems that they'd have no problem doing so if it came to that. I'm honestly tired of the wise, polite people keeping their mouths shut because it's rude, while the rude people never stop being idiots.</text></comment> |
40,313,348 | 40,313,383 | 1 | 3 | 40,313,171 | train | <story><title>Google Cloud accidentally deletes customer's account</title><url>https://www.theregister.com/2024/05/09/unisuper_google_cloud_outage_caused/</url></story><parent_chain></parent_chain><comment><author>j45</author><text>The cloud is someone else&#x27;s computer. Convenience isn&#x27;t always reliability and security.<p>For anyone who is semi-technical, or not as technical as they wished when it came to file storage wondering what they can do...<p>- Whether it&#x27;s been corporate clients, small business, or individuals: I universally recommend everyone owning a small QNAP or Synology that is storage as a zero maintenance appliance, running software to maintain a 2 way sync of your cloud drives.<p>- Even if you&#x27;re using Google Cloud, MSFT, etc, continue to use it as you please, just siphon off a local data backup in case the internet or the cloud is down. It also can make some kinds of disaster recovery much quicker. For example if you back up your computers locally to a NAS, and then from there to the cloud, it can be a lot more manageable.<p>- Throwing something like Tailscale on it makes it invisible and hyper secure to have on all your devices too. I typically never use the NAS&#x27; network connection tools, as it&#x27;s likely a juicy target to break.<p>- Last but not least, setup a different format of backup, automatically. Backup your file appliance to something like sync.com, tarsnap.com, backblaze, etc to back up elsewhere that you can access.<p>I&#x27;d be happy to learn what anyone else is doing. As someone who lost a ton of data on a Microsoft account once, never twice.<p>*I am perfectly capable of building a NAS myself running an open source package, but storage should be an appliance at home to focus on other things.</text></comment> | <story><title>Google Cloud accidentally deletes customer's account</title><url>https://www.theregister.com/2024/05/09/unisuper_google_cloud_outage_caused/</url></story><parent_chain></parent_chain><comment><author>doctor_eval</author><text>&gt; In the meantime, UniSuper&#x27;s woes remain a lesson for companies leaping cloudwards. Someone clicking the wrong button, a previously unknown bug, an unforeseen series of events…<p>Here’s the thing though. Back in the Dark Ages before cloud services, everyone had to self host. We had a Data General AViiON server (DG&#x2F;UX FTW!) in a dedicated room, and one of the first RAID arrays in Australia (predecessor of CLARiiON).<p>The cover was off the front of the array for some reason, and I had to squeeze past a coworker to get out of the room.<p>Sitting down at my Wyse60 terminal to do some work, a bunch of errors started appearing on my screen. Turns out, I had also “squeezed past” the power button on the RAID array, which was normally recessed - but not when the cover was off. I’d inadvertently shut down the whole system. Fortunately we were in preproduction so nobody really noticed. But it scared the crap out of me.<p>I knew someone at a small bank who told me they were also susceptible to similar problems. Just one big server, let’s hope it doesn’t go down.<p>Cloud services - especially IaaS - _enable_ diversification, and it sounds like UniSuper’s IT team should be congratulated for understanding what this really means in the context of networked services. Diverse networks, diverse suppliers, diverse geography.<p>Without cloud services, none of this is feasible for most SMEs.<p>There are plenty of things we can complain about with the cloud but “someone clicking the wrong button” is even more of a risk if you run your gear in house.</text></comment> |
10,159,908 | 10,160,031 | 1 | 2 | 10,159,244 | train | <story><title>How Pocket Hit 20M Users with 20 People</title><url>http://firstround.com/review/the-story-behind-how-pocket-hit-20m-users-with-20-people/</url></story><parent_chain></parent_chain><comment><author>eliben</author><text>I&#x27;m one of the first paying customers of Pocket.<p>Alas, their quality has been going down recently. The most important feature - keeping the location <i>inside</i> an article between app invocations, is not working. For long articles, if I stop in the middle and want to resume later, there&#x27;s a 80% chance that Pocket will happily set me right in the beginning.<p>Besides, their rendering for articles with code sucks, so I almost always use &quot;web view&quot;, a decision Pocket also forgets every other time.<p>So I end up using Pocket as a convenient keyboard-shortcut to save articles, but on my phone actually open them into Chrome, which has no problem remembering the location in a tab.</text></comment> | <story><title>How Pocket Hit 20M Users with 20 People</title><url>http://firstround.com/review/the-story-behind-how-pocket-hit-20m-users-with-20-people/</url></story><parent_chain></parent_chain><comment><author>11thEarlOfMar</author><text>When I see stats like this, or like the 400MM users&#x2F;40 devs, that WhatsApp had at one point, I can&#x27;t help but think back to, say, 1985. What would it take to develop and scale a software product to that number of users?<p>I worked for MultiScope in 1991. We had to order discs and have disc labels printed, copy the compiler onto the discs, have manuals and boxes printed, stuff and shrink-wrap the boxes, ship to Ingram Micro for distribution, and then wait 2-4 weeks for our product to show up on the shelves at Egghead. I recall 5 developers, and we were ecstatic to ship 4,000 copies of a major new version.<p>That gets me thinking in terms of leverage. The leverage that 2015 Internet technology affords a single developer is a potent economic force.</text></comment> |
34,328,403 | 34,328,393 | 1 | 2 | 34,327,763 | train | <story><title>Surveillance Footage of Tesla Crash on SF’s Bay Bridge</title><url>https://theintercept.com/2023/01/10/tesla-crash-footage-autopilot/</url></story><parent_chain><item><author>a_cardboard_box</author><text>In this case the Tesla slammed on the brakes <i>and</i> changed lanes, so there wasn&#x27;t a realistic way for the other car to prevent the collision.</text></item><item><author>markisus</author><text>The Tesla should not have stopped but there is a cultural problem of driving way too close to the car in front of you. There are other justified scenarios why a car would have had to slam on the brakes at that same spot and the pileup would have still occurred.</text></item></parent_chain><comment><author>markisus</author><text>You&#x27;re right. The first rear-ending may have been unavoidable. However, you can see from the rear surveillance camera that subsequent vehicles could have stopped if everyone agreed to keep larger following distances. As mentioned elsewhere in the thread, there is the game-theoretical issue of other drivers cutting in when large gaps form so it would take some coordination on a technology &#x2F; policy &#x2F; cultural level to resolve the issue.</text></comment> | <story><title>Surveillance Footage of Tesla Crash on SF’s Bay Bridge</title><url>https://theintercept.com/2023/01/10/tesla-crash-footage-autopilot/</url></story><parent_chain><item><author>a_cardboard_box</author><text>In this case the Tesla slammed on the brakes <i>and</i> changed lanes, so there wasn&#x27;t a realistic way for the other car to prevent the collision.</text></item><item><author>markisus</author><text>The Tesla should not have stopped but there is a cultural problem of driving way too close to the car in front of you. There are other justified scenarios why a car would have had to slam on the brakes at that same spot and the pileup would have still occurred.</text></item></parent_chain><comment><author>newaccount74</author><text>Curiously the vehicle behind the Tesla barely hit it whereas the drivers of the cars further back in the pileup seemed to have paid less attention.<p>There would have been multiple ways for the cars further back to prevent a pileup: Drive slower, keep more distance. If you speed, have both hands on the wheel, look at the road, and be ready to brake quickly.<p>It does sound like there were no severe injuries, so you could argue that people kept enough distance to avoid serious injuries.</text></comment> |
24,874,036 | 24,874,322 | 1 | 2 | 24,871,762 | train | <story><title>Four features that justify a new Unix shell</title><url>http://www.oilshell.org/blog/2020/10/osh-features.html</url></story><parent_chain><item><author>chubot</author><text>I should turn this into a FAQ, but PowerShell is natural on Windows, where the OS provides objects (either via the .NET VM, or COM and .DLLs, etc.)<p>A Unix shell like bash or Oil is natural on Unix, where the OS uses text files. And in distributed systems where data is JSON, YAML, XML, protobuf, msgpack, etc. not objects.<p>So basically shell is a &quot;situated&quot; language, and the easy of accomplishing any given task depends a lot on the environment.<p>Windows is more tightly coupled and trying to provide something nice for you. Unix is messier but doesn&#x27;t limit you, and it&#x27;s what basically all big systems are made of these days. A major strength of shell is to glue things together that nobody thought should be glued together.<p>---<p>Good example here from Paul Bucheit: <a href="http:&#x2F;&#x2F;www.oilshell.org&#x2F;blog&#x2F;2020&#x2F;01&#x2F;simplest-explanation.html#productive-programmers-secretly-use-shell" rel="nofollow">http:&#x2F;&#x2F;www.oilshell.org&#x2F;blog&#x2F;2020&#x2F;01&#x2F;simplest-explanation.ht...</a><p><a href="http:&#x2F;&#x2F;paulbuchheit.blogspot.com&#x2F;2009&#x2F;01&#x2F;communicating-with-code.html" rel="nofollow">http:&#x2F;&#x2F;paulbuchheit.blogspot.com&#x2F;2009&#x2F;01&#x2F;communicating-with-...</a><p><i>However, we needed a way for Gmail to make money, and Sanjeev Singh kept talking about using relevant ads, even though it was obviously a &quot;bad idea&quot;. I remained skeptical, but thought that it might be a fun experiment, so I connected to that ads database (I assure you, random engineers can no longer do this!), copied out all of the ads+keywords, and did a little bit of sorting and filtering with some UNIX SHELL COMMANDS</i><p><i>I then hacked up the &quot;adult content&quot; classifier that Matt Cutts and I had written for safe-search, linked that into the Gmail prototype, and then loaded the ads data into the classifier. My change to the classifier (which completely broke its original functionality, but this was a separate code branch) changed it from classifying pages as &quot;adult&quot;, to classifying them according to which ad was most relevant. The resulting ad was then displayed in a little box on our Gmail prototype ui. The code was rather ugly and hackish, but more importantly, it only took a few hours to write!</i><p>----<p>Also, use whatever&#x27;s best for you, but Unix people tend to hate PowerShell:<p><a href="https:&#x2F;&#x2F;medium.com&#x2F;@octskyward&#x2F;the-woes-of-powershell-8737e5346b1" rel="nofollow">https:&#x2F;&#x2F;medium.com&#x2F;@octskyward&#x2F;the-woes-of-powershell-8737e5...</a><p><i>PowerShell feels like it was built by people who had heard about command lines a long time ago and tried to recreate one based on stories passed down through generations of their ancestors</i><p>The fact that they cargo-culted operators like -eq, -le, -lt, while not maintaining compatibility is just silly to me. In Oil it&#x27;s &quot;x == y&quot; for string equality.</text></item><item><author>solarkraft</author><text>OSH is probably a good incremental improvement over bash, but I also enjoy using the significantly more tradition-breaking Powershell with its object oriented nature.<p>It feels a lot more like programming and actually gives you useful suggestions right inside the terminal!<p>On Linux the auto completion behavior is luckily less obnoxious than on Windows and it doesn&#x27;t have the multi second startup delay either.</text></item></parent_chain><comment><author>com2kid</author><text>&gt; I connected to that ads database (I assure you, random engineers can no longer do this!), copied out all of the ads+keywords, and did a little bit of sorting and filtering with some UNIX SHELL COMMANDS<p>One of PowerShell&#x27;s nifty tricks includes being able to walk a SQL DB like its a file system, see <a href="https:&#x2F;&#x2F;docs.microsoft.com&#x2F;en-us&#x2F;sql&#x2F;powershell&#x2F;navigate-sql-server-powershell-paths?view=sql-server-ver15" rel="nofollow">https:&#x2F;&#x2F;docs.microsoft.com&#x2F;en-us&#x2F;sql&#x2F;powershell&#x2F;navigate-sql...</a><p>PowerShell is <i>insanely</i> flexible.</text></comment> | <story><title>Four features that justify a new Unix shell</title><url>http://www.oilshell.org/blog/2020/10/osh-features.html</url></story><parent_chain><item><author>chubot</author><text>I should turn this into a FAQ, but PowerShell is natural on Windows, where the OS provides objects (either via the .NET VM, or COM and .DLLs, etc.)<p>A Unix shell like bash or Oil is natural on Unix, where the OS uses text files. And in distributed systems where data is JSON, YAML, XML, protobuf, msgpack, etc. not objects.<p>So basically shell is a &quot;situated&quot; language, and the easy of accomplishing any given task depends a lot on the environment.<p>Windows is more tightly coupled and trying to provide something nice for you. Unix is messier but doesn&#x27;t limit you, and it&#x27;s what basically all big systems are made of these days. A major strength of shell is to glue things together that nobody thought should be glued together.<p>---<p>Good example here from Paul Bucheit: <a href="http:&#x2F;&#x2F;www.oilshell.org&#x2F;blog&#x2F;2020&#x2F;01&#x2F;simplest-explanation.html#productive-programmers-secretly-use-shell" rel="nofollow">http:&#x2F;&#x2F;www.oilshell.org&#x2F;blog&#x2F;2020&#x2F;01&#x2F;simplest-explanation.ht...</a><p><a href="http:&#x2F;&#x2F;paulbuchheit.blogspot.com&#x2F;2009&#x2F;01&#x2F;communicating-with-code.html" rel="nofollow">http:&#x2F;&#x2F;paulbuchheit.blogspot.com&#x2F;2009&#x2F;01&#x2F;communicating-with-...</a><p><i>However, we needed a way for Gmail to make money, and Sanjeev Singh kept talking about using relevant ads, even though it was obviously a &quot;bad idea&quot;. I remained skeptical, but thought that it might be a fun experiment, so I connected to that ads database (I assure you, random engineers can no longer do this!), copied out all of the ads+keywords, and did a little bit of sorting and filtering with some UNIX SHELL COMMANDS</i><p><i>I then hacked up the &quot;adult content&quot; classifier that Matt Cutts and I had written for safe-search, linked that into the Gmail prototype, and then loaded the ads data into the classifier. My change to the classifier (which completely broke its original functionality, but this was a separate code branch) changed it from classifying pages as &quot;adult&quot;, to classifying them according to which ad was most relevant. The resulting ad was then displayed in a little box on our Gmail prototype ui. The code was rather ugly and hackish, but more importantly, it only took a few hours to write!</i><p>----<p>Also, use whatever&#x27;s best for you, but Unix people tend to hate PowerShell:<p><a href="https:&#x2F;&#x2F;medium.com&#x2F;@octskyward&#x2F;the-woes-of-powershell-8737e5346b1" rel="nofollow">https:&#x2F;&#x2F;medium.com&#x2F;@octskyward&#x2F;the-woes-of-powershell-8737e5...</a><p><i>PowerShell feels like it was built by people who had heard about command lines a long time ago and tried to recreate one based on stories passed down through generations of their ancestors</i><p>The fact that they cargo-culted operators like -eq, -le, -lt, while not maintaining compatibility is just silly to me. In Oil it&#x27;s &quot;x == y&quot; for string equality.</text></item><item><author>solarkraft</author><text>OSH is probably a good incremental improvement over bash, but I also enjoy using the significantly more tradition-breaking Powershell with its object oriented nature.<p>It feels a lot more like programming and actually gives you useful suggestions right inside the terminal!<p>On Linux the auto completion behavior is luckily less obnoxious than on Windows and it doesn&#x27;t have the multi second startup delay either.</text></item></parent_chain><comment><author>Denvercoder9</author><text>There&#x27;s nothing inherent about an object-based design that limits what you can glue together. Yes, it might require a bit more effort to transform object formats, but today&#x27;s untyped shell also creates problems that don&#x27;t exist in a typed environment.</text></comment> |
35,684,180 | 35,683,772 | 1 | 2 | 35,683,254 | train | <story><title>DuckDuckGo removed the ability to filter search results</title><url>https://github.com/duckduckgo/duckduckgo-help-pages/commit/d35d03e532c7618bcdd2be10ea67e9e1f021dd96</url></story><parent_chain><item><author>hedora</author><text>I’ve also noticed that “”’s stopped working as expected there recently.<p>Sample query: “lowes pet bedding”. The quotation marks do something; they remove some results from tractor supply and home depot. However, many results don’t contain the word “lowes” or “bedding”. For example:<p><a href="https:&#x2F;&#x2F;animalcare.lacounty.gov&#x2F;licensing&#x2F;" rel="nofollow">https:&#x2F;&#x2F;animalcare.lacounty.gov&#x2F;licensing&#x2F;</a><p>And, none contain the exact phrase. Google does the expected thing (return one result). Annoyingly, the last time I checked, the situation was reversed, and Google search is now useless for other reasons.<p><i>Thirty seven</i> spam results surround the one organic google search result for the query. For the love of all that is good and proper, <i>thirty seven</i>?!? Who thought that was a good number?<p>That result page would make 90’s era pay-for-placement search engines blush!</text></item></parent_chain><comment><author>Animats</author><text>I tried<p><pre><code> graphics card -&quot;nvidia&quot;
graphics card -&quot;radeon&quot;
</code></pre>
on DuckDuckGo, Google, and Bing. Google is currently the only one which appeared to exclude the requested term. Bing and DuckDuckGo not only didn&#x27;t exclude it, they highlighted it in results.</text></comment> | <story><title>DuckDuckGo removed the ability to filter search results</title><url>https://github.com/duckduckgo/duckduckgo-help-pages/commit/d35d03e532c7618bcdd2be10ea67e9e1f021dd96</url></story><parent_chain><item><author>hedora</author><text>I’ve also noticed that “”’s stopped working as expected there recently.<p>Sample query: “lowes pet bedding”. The quotation marks do something; they remove some results from tractor supply and home depot. However, many results don’t contain the word “lowes” or “bedding”. For example:<p><a href="https:&#x2F;&#x2F;animalcare.lacounty.gov&#x2F;licensing&#x2F;" rel="nofollow">https:&#x2F;&#x2F;animalcare.lacounty.gov&#x2F;licensing&#x2F;</a><p>And, none contain the exact phrase. Google does the expected thing (return one result). Annoyingly, the last time I checked, the situation was reversed, and Google search is now useless for other reasons.<p><i>Thirty seven</i> spam results surround the one organic google search result for the query. For the love of all that is good and proper, <i>thirty seven</i>?!? Who thought that was a good number?<p>That result page would make 90’s era pay-for-placement search engines blush!</text></item></parent_chain><comment><author>Svip</author><text>The quotation marks work for me:<p><a href="https:&#x2F;&#x2F;duckduckgo.com&#x2F;?t=ffab&amp;q=%22lowes+pet+bedding%22&amp;atb=v101-1&amp;ia=web" rel="nofollow">https:&#x2F;&#x2F;duckduckgo.com&#x2F;?t=ffab&amp;q=%22lowes+pet+bedding%22&amp;atb...</a><p>I notice in your HN post, you&#x27;re using fancy quotation marks; I assume you&#x27;re not using the same while searching DDG.</text></comment> |
40,577,993 | 40,578,111 | 1 | 3 | 40,571,727 | train | <story><title>Sphere Rendering: Flat Planets</title><url>https://emildziewanowski.com/flat-planets/</url></story><parent_chain></parent_chain><comment><author>mrsharpoblunto</author><text>The author dismisses cubemaps pretty quickly, but imo it’s the simplest solution &amp; it’s what I did when rendering dynamic gas giants on my own personal project a number of years back* . Using a cubemap doesn’t result in a 6x increase in memory usage, you’re just splitting the texture size from one large rectangular face into 6 smaller rectangular faces, but the total texture detail is the same. The nice part about a cubemap is you don’t have to worry about pole pinching at all + you can use a 3 or 4 dimensional noise function to easily create a seamless flow field for texture animation&#x2F;distortion.<p>* <a href="https:&#x2F;&#x2F;www.junkship.net&#x2F;News&#x2F;2016&#x2F;06&#x2F;09&#x2F;jupiter-jazz" rel="nofollow">https:&#x2F;&#x2F;www.junkship.net&#x2F;News&#x2F;2016&#x2F;06&#x2F;09&#x2F;jupiter-jazz</a></text></comment> | <story><title>Sphere Rendering: Flat Planets</title><url>https://emildziewanowski.com/flat-planets/</url></story><parent_chain></parent_chain><comment><author>JKCalhoun</author><text>Reminds me to revisit displacement mapping — probably not going to be a replacement for the problem that author is trying to solve but simpler and kind of fun.<p>I wrote a kind of cool music visualizer for SoundJam perhaps 25 years ago I called &quot;Eclipse&quot;. Input data was an array of levels (probably integers) across some range of audible frequencies — a left and right channel.<p>Think of the eclipsed sun with corona ejections — that was what I was going for. The music data was the &quot;ejections&quot;. The frequency of the data determined where around the disc of the sun it would appear.<p>Over time the ejecta moved away from the sun and soon disappeared as they &quot;cooled&quot; to black — the initial color of the ejecta being white for the strongest signals — yellow, orange, red, brown when weaker. (Think of the black body curve.)<p>I had to keep a circular buffer of the sound data values (an array of arrays) large enough to represent how much time the ejecta would &quot;live&quot; before disappearing to black.<p>In any event, the whole display of the &quot;eclipse&quot;, ejecta, was just a displacement map. I had pre-calculated a bitmap where the value for each &quot;pixel&quot; was an offset into the buffer of sound level values. &quot;pixels&quot; close to the surface of the sun would have offsets to the new data coming in, pixels further out would have offsets into the tail of the buffer that was about to expire. With the circular, radial aspect of the ejecta, there was some math involved in generating the displacement values in order to map from essentially a radial space to a cartesian one.<p>With that established as described, the main loop simply pulled in new sound values, over-wrote the oldest values in the circular buffer with the new and then iterated row and column-wise over the displacement map, grabbing the corresponding sound data value, mapped it to a color in a fixed palette and pushed that color into display buffer.<p>Although not as &quot;flashy&quot; as other visualizers, there was I thought a calm beauty to it. And it very much represented the music data (you know, as opposed to later visualizers where, even when presented with silence, they seemed to be unable to settle down).</text></comment> |
40,610,614 | 40,610,361 | 1 | 3 | 40,608,413 | train | <story><title>σ-GPTs: A new approach to autoregressive models</title><url>https://arxiv.org/abs/2404.09562</url></story><parent_chain><item><author>cs702</author><text>This looks <i>great</i>.<p>The authors randomly permute (i.e., shuffle) input tokens in training and add two positional encodings to each token: one with the token&#x27;s position and another with the position of the token to be predicted. Otherwise, the model is a standard autoregressive GPT. The consequences of this seemingly &quot;simple&quot; modification are significant:<p>* The authors can prompt the trained model with part of a sequence and then decode the missing tokens, all at once, in parallel, regardless of order -- i.e., the model can in-fill in parallel.<p>* The authors can compute conditional probability densities for every missing token in a sequence, again in parallel, i.e., densities for all missing tokens at once.<p>* The authors propose a rejection-sampling method for generating in-fill tokens, again in parallel. Their method seems to work well in practice.<p>I&#x27;ve added this to my reading list. Thank you for sharing it on HN.</text></item></parent_chain><comment><author>toxik</author><text>This problem formulation has been around for a while, it’s kind of the holy grail of modeling. What is new compared to PixelCNN and related is this position embedding idea.</text></comment> | <story><title>σ-GPTs: A new approach to autoregressive models</title><url>https://arxiv.org/abs/2404.09562</url></story><parent_chain><item><author>cs702</author><text>This looks <i>great</i>.<p>The authors randomly permute (i.e., shuffle) input tokens in training and add two positional encodings to each token: one with the token&#x27;s position and another with the position of the token to be predicted. Otherwise, the model is a standard autoregressive GPT. The consequences of this seemingly &quot;simple&quot; modification are significant:<p>* The authors can prompt the trained model with part of a sequence and then decode the missing tokens, all at once, in parallel, regardless of order -- i.e., the model can in-fill in parallel.<p>* The authors can compute conditional probability densities for every missing token in a sequence, again in parallel, i.e., densities for all missing tokens at once.<p>* The authors propose a rejection-sampling method for generating in-fill tokens, again in parallel. Their method seems to work well in practice.<p>I&#x27;ve added this to my reading list. Thank you for sharing it on HN.</text></item></parent_chain><comment><author>taneq</author><text>Wow, if that works that&#x27;s wild (and also has that &quot;damn, now you say it it&#x27;s obvious&quot; flavour that so many really cool discoveries share...)</text></comment> |
32,300,772 | 32,300,888 | 1 | 2 | 32,298,096 | train | <story><title>A looming copper crunch and why recycling can’t fix it</title><url>https://www.mining.com/the-looming-copper-crunch-and-why-recycling-cant-fix-it/</url></story><parent_chain><item><author>ianai</author><text>Who&#x27;s expecting a decline in construction demand in China? I think I&#x27;ve heard this a few places, granted, but have they really built up enough for their whole population (that would want it)?</text></item><item><author>jbay808</author><text>I&#x27;m not an economist, but I&#x27;m going to go out on a limb here: my expectation is that between now and 2025, a decline in construction demand in China will have a more significant impact on global copper consumption than an increase due to EV demand. I think the copper supplies will be okay.</text></item><item><author>amluto</author><text>Some cursory searching suggests that between 40 and 50% of copper is used in building construction. I don’t know the further breakdown, but:<p>Copper is widely used for flashing. For this application, galvanized steel, aluminum, and stainless steel can substitute. All are less expensive.<p>Copper is used for pipes. They are <i>much</i> more expensive than plastics. Arguably, depending on the particular application, one or more plastic options are as good or better. (Copper is unharmed by moderate chlorine concentrations and sunlight. It’s mechanically strong. It’s inert to water at appropriate pH. It is quite reactive to water at the wrong pH. Boiler condensate will quickly destroy it.)<p>Copper is used for heavy-gauge electrical wire. For many of these applications, aluminum is much less expensive and arguable superior (it’s lighter and more flexible).<p>Copper is used for 12 and 14 gauge branch circuits. Aluminum branch circuits are currently strongly discouraged.<p>In any event, a lot of copper is consumed for applications that don’t need it. If prices go up, the industry can adapt.</text></item></parent_chain><comment><author>thematrixturtle</author><text>China&#x27;s population will start shrinking around 2025, and there&#x27;s a real estate bubble that&#x27;s clearly starting to deflate right now.<p>Obviously China is still a poor country and there&#x27;s a lot of housing stock being upgraded, so it&#x27;s not going to be a complete collapse, but I do expect construction to slow down both in the short and long term.</text></comment> | <story><title>A looming copper crunch and why recycling can’t fix it</title><url>https://www.mining.com/the-looming-copper-crunch-and-why-recycling-cant-fix-it/</url></story><parent_chain><item><author>ianai</author><text>Who&#x27;s expecting a decline in construction demand in China? I think I&#x27;ve heard this a few places, granted, but have they really built up enough for their whole population (that would want it)?</text></item><item><author>jbay808</author><text>I&#x27;m not an economist, but I&#x27;m going to go out on a limb here: my expectation is that between now and 2025, a decline in construction demand in China will have a more significant impact on global copper consumption than an increase due to EV demand. I think the copper supplies will be okay.</text></item><item><author>amluto</author><text>Some cursory searching suggests that between 40 and 50% of copper is used in building construction. I don’t know the further breakdown, but:<p>Copper is widely used for flashing. For this application, galvanized steel, aluminum, and stainless steel can substitute. All are less expensive.<p>Copper is used for pipes. They are <i>much</i> more expensive than plastics. Arguably, depending on the particular application, one or more plastic options are as good or better. (Copper is unharmed by moderate chlorine concentrations and sunlight. It’s mechanically strong. It’s inert to water at appropriate pH. It is quite reactive to water at the wrong pH. Boiler condensate will quickly destroy it.)<p>Copper is used for heavy-gauge electrical wire. For many of these applications, aluminum is much less expensive and arguable superior (it’s lighter and more flexible).<p>Copper is used for 12 and 14 gauge branch circuits. Aluminum branch circuits are currently strongly discouraged.<p>In any event, a lot of copper is consumed for applications that don’t need it. If prices go up, the industry can adapt.</text></item></parent_chain><comment><author>atwood22</author><text>The Chinese government is cracking down on debt-fueled property developers.</text></comment> |
38,349,985 | 38,350,167 | 1 | 2 | 38,348,556 | train | <story><title>Canada Inches Closer to 'Right to Repair' Reform</title><url>https://www.techdirt.com/2023/11/20/canada-inches-closer-to-right-to-repair-reform/</url></story><parent_chain></parent_chain><comment><author>thomastjeffery</author><text>It&#x27;s absolutely insane that Copyright can have anything to do with repairs in the first place.<p>We have structured practically every legal framework in the digital world around the DMCA. What&#x27;s our answer to fraud? Copyright. What&#x27;s our answer to privacy violations? Copyright. What&#x27;s our answer to libel? Copyright. Free software? Copyright.<p>This system only helps for a handful of corporations; at the expense of the rest of us. It&#x27;s time to scrap Copyright and start over.</text></comment> | <story><title>Canada Inches Closer to 'Right to Repair' Reform</title><url>https://www.techdirt.com/2023/11/20/canada-inches-closer-to-right-to-repair-reform/</url></story><parent_chain></parent_chain><comment><author>markhahn</author><text>Once people get woke to the right-to-repair their devices, maybe we can have a discussion on how open-source is the software version of the same thing. Perhaps even broach the discussion of interoperability and standards (vs proprietary and locked-in).</text></comment> |
21,631,936 | 21,627,806 | 1 | 2 | 21,627,166 | train | <story><title>Novelty and Heresy</title><url>http://paulgraham.com/nov.html</url></story><parent_chain><item><author>classified</author><text>Just look at HN comment threads: Any truth that&#x27;s inconvenient for the majority in the respective thread will be brainlessly downvoted. Killing the messenger is just as popular here (where all the smartest know-it-alls convene) as it is there (where all the ignoramuses reside).</text></item></parent_chain><comment><author>dang</author><text>That&#x27;s true of any large public community, almost by definition. It bothers me too, and I have to remind myself not to fret too much over things on HN that are true of humans in general: <a href="https:&#x2F;&#x2F;hn.algolia.com&#x2F;?dateRange=all&amp;page=0&amp;prefix=false&amp;query=by%3Adang%20%22humans%20in%20general%22&amp;sort=byDate&amp;type=comment" rel="nofollow">https:&#x2F;&#x2F;hn.algolia.com&#x2F;?dateRange=all&amp;page=0&amp;prefix=false&amp;qu...</a><p>The only way you could get different phenomena is with a small, closed community that would therefore be able to have locally different properties. Just like you need a closed system if you want different entropy.<p>Actually, this has an interesting consequence along the lines of PG&#x27;s essay: when a novel&#x2F;heretical idea does begin to be seen as true, this will probably happen first in small communities, not large ones. Moreover the small community probably needs to be elite somehow, or it won&#x27;t have the standing to deliver the bit flip (heretical -&gt; serious) to the larger community.</text></comment> | <story><title>Novelty and Heresy</title><url>http://paulgraham.com/nov.html</url></story><parent_chain><item><author>classified</author><text>Just look at HN comment threads: Any truth that&#x27;s inconvenient for the majority in the respective thread will be brainlessly downvoted. Killing the messenger is just as popular here (where all the smartest know-it-alls convene) as it is there (where all the ignoramuses reside).</text></item></parent_chain><comment><author>Ma8ee</author><text>Could you give me some examples?</text></comment> |
13,180,207 | 13,180,186 | 1 | 2 | 13,178,305 | train | <story><title>Does It Make Sense for Programmers to Move to the Bay Area?</title><url>http://blog.triplebyte.com/does-it-make-sense-for-programmers-to-move-to-the-bay-area</url></story><parent_chain><item><author>qntty</author><text>This is really insightful. Does this fallacy have an well known name?</text></item><item><author>harterrt</author><text>This could also be an artifact of the metric. I call this the &quot;Geico&quot; fallacy.<p>Geico advertises that &quot;People who switch to Geico save 15% or more&quot;. Notice that people who would lose money by switching to Geico should rarely switch. Accordingly, this metric will usually be positive.<p>Consider that people moving to a new city for a job may need more incentive to move.</text></item><item><author>tom_b</author><text>Interesting that Bay Area hackers make more than local hackers when they relocate outside the Bay Area.<p>FTA:<p><pre><code> A 2015 report by Hired found that when engineers from
the Bay Area relocate to other areas, they out-earn
engineers on the local market. Experience in the Bay
Area seems to advance careers. Engineers moving from
San Francisco to Seattle make an average of $9,000
more than others who get offers in Seattle. This Bay
Area premium is even higher in other cities: $16,000
in Boston, $17,000 in Chicago, and $19,000 in San Diego.
</code></pre>
[found slide at <a href="http:&#x2F;&#x2F;get.hired.com&#x2F;rs&#x2F;348-IPO-044&#x2F;images&#x2F;Hired-State-of-Salaries.pdf" rel="nofollow">http:&#x2F;&#x2F;get.hired.com&#x2F;rs&#x2F;348-IPO-044&#x2F;images&#x2F;Hired-State-of-Sa...</a>]<p>Bay Area hackers are more valued in different markets than local hackers. I would love to see the raw data for the &quot;relocating&quot; hackers and local hackers. Is it a question of applied experience opportunities in the Bay Area hackers? Is just startup afterglow? Are relocating hackers better than average pre-Bay Area experience to begin with and this shows up when they migrate away from the Bay Area?</text></item></parent_chain><comment><author>sjbp</author><text>Self-selection bias : <a href="https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Self-selection_bias" rel="nofollow">https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Self-selection_bias</a></text></comment> | <story><title>Does It Make Sense for Programmers to Move to the Bay Area?</title><url>http://blog.triplebyte.com/does-it-make-sense-for-programmers-to-move-to-the-bay-area</url></story><parent_chain><item><author>qntty</author><text>This is really insightful. Does this fallacy have an well known name?</text></item><item><author>harterrt</author><text>This could also be an artifact of the metric. I call this the &quot;Geico&quot; fallacy.<p>Geico advertises that &quot;People who switch to Geico save 15% or more&quot;. Notice that people who would lose money by switching to Geico should rarely switch. Accordingly, this metric will usually be positive.<p>Consider that people moving to a new city for a job may need more incentive to move.</text></item><item><author>tom_b</author><text>Interesting that Bay Area hackers make more than local hackers when they relocate outside the Bay Area.<p>FTA:<p><pre><code> A 2015 report by Hired found that when engineers from
the Bay Area relocate to other areas, they out-earn
engineers on the local market. Experience in the Bay
Area seems to advance careers. Engineers moving from
San Francisco to Seattle make an average of $9,000
more than others who get offers in Seattle. This Bay
Area premium is even higher in other cities: $16,000
in Boston, $17,000 in Chicago, and $19,000 in San Diego.
</code></pre>
[found slide at <a href="http:&#x2F;&#x2F;get.hired.com&#x2F;rs&#x2F;348-IPO-044&#x2F;images&#x2F;Hired-State-of-Salaries.pdf" rel="nofollow">http:&#x2F;&#x2F;get.hired.com&#x2F;rs&#x2F;348-IPO-044&#x2F;images&#x2F;Hired-State-of-Sa...</a>]<p>Bay Area hackers are more valued in different markets than local hackers. I would love to see the raw data for the &quot;relocating&quot; hackers and local hackers. Is it a question of applied experience opportunities in the Bay Area hackers? Is just startup afterglow? Are relocating hackers better than average pre-Bay Area experience to begin with and this shows up when they migrate away from the Bay Area?</text></item></parent_chain><comment><author>evincarofautumn</author><text>I think it’s attrition bias, a form of selection bias.<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Selection_bias#Attrition" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Selection_bias#Attrition</a></text></comment> |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.