id
int64 2
42.1M
| by
large_stringlengths 2
15
⌀ | time
timestamp[us] | title
large_stringlengths 0
198
⌀ | text
large_stringlengths 0
27.4k
⌀ | url
large_stringlengths 0
6.6k
⌀ | score
int64 -1
6.02k
⌀ | descendants
int64 -1
7.29k
⌀ | kids
large list | deleted
large list | dead
bool 1
class | scraping_error
large_stringclasses 25
values | scraped_title
large_stringlengths 1
59.3k
⌀ | scraped_published_at
large_stringlengths 4
66
⌀ | scraped_byline
large_stringlengths 1
757
⌀ | scraped_body
large_stringlengths 1
50k
⌀ | scraped_at
timestamp[us] | scraped_language
large_stringclasses 58
values | split
large_stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
42,069,835 | mykeyglobal | 2024-11-06T21:37:25 | null | null | null | 1 | null | [
42069836
] | null | true | null | null | null | null | null | null | null | train |
42,069,842 | melatrmwp | 2024-11-06T21:37:40 | Making $4000/month from a free DNS lookup tool I built | null | https://www.foundernoon.com/casestudies/ruurtjan-pul | 1 | 1 | [
42070405
] | null | null | null | null | null | null | null | null | null | train |
42,069,863 | paulpauper | 2024-11-06T21:39:13 | Woman Couldn't Chew for 2 Years After 'Horror' Dental Surgery, She Says | null | https://www.vice.com/en/article/woman-couldnt-chew-for-2-years-after-horror-dental-surgery-she-says/ | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,069,871 | ivewonyoung | 2024-11-06T21:39:56 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,069,874 | benhoyt | 2024-11-06T21:40:05 | The small web is beautiful (2021) | null | https://benhoyt.com/writings/the-small-web-is-beautiful/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,069,911 | asim | 2024-11-06T21:42:46 | AI Is the Next Platform | null | http://aslam.com/ai | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,069,939 | chaktty | 2024-11-06T21:44:35 | null | null | null | 1 | null | [
42069940
] | null | true | null | null | null | null | null | null | null | train |
42,069,983 | mkaic | 2024-11-06T21:47:47 | Vector math library codegen in Debug | null | https://aras-p.info/blog/2024/09/14/Vector-math-library-codegen-in-Debug/ | 1 | 0 | null | null | null | no_error | Vector math library codegen in Debug · Aras' website | 2024-09-14T21:01:10+03:00 | null |
This will be about how when in your C++ code you have a “vector math library”, and how the choices of code style in there affect non-optimized
build performance.
Backstory
A month ago I got into the rabbit hole of trying to “sanitize” the various
ways that images can be resized within Blender codebase. There were at least 4
different functions to do that, with different filtering modes (expected),
but also different corner case behaviors and other funkiness, that was not
well documented and not well understood.
I combed through all of that, fixed some arguably wrong behaviors of some
of the functions, unified their behavior, etc. etc. Things got faster
and better documented. Yay! (PR)
However. While doing that, I also made the code smaller, primarily
following the guideline of “code should use our C++ math library, not
the legacy C one”. That is, use Blender codebase classes like float4 with
related functions and operators (e.g. float4 c = a + b), instead of
float v[4] c; add_v4_v4v4(c, a, b); and so on. Sounds good? Yes!
But. There’s always a “but”.
Other developers later on noticed that some parts of Blender got slower,
in non-optimized (“Debug”) build. Normally people say “oh it’s a Debug build,
no one should care about performance of it”, and while in some cases it might be
true, when anything becomes slower it is annoying.
In this particular case, it was “saving a file within Blender”. You see, as part
of the saving process, it takes a screenshot of your application, resizes it
to be smaller and embeds that as a “thumbnail” inside the file itself. And yes,
this “resize it” part is exactly what my change affected. Many developers
run their build in Debug mode for easier debugging and/or faster builds;
some run it in Debug mode with Address Sanitizer on as well. If “save a file”,
an operation that you normally do many times, became slower by say 2 seconds,
that is annoying.
What can be done?
How Blender’s C++ math library is written today
It is pretty compact and neat! And perhaps too flexible :)
Base of the math vector types is this struct, with is just a fixed size array
of N entries. For the (most common) case of 2D, 3D and 4D vectors, the struct
is instead entries explicitly named x, y, z, w:
template<typename T, int Size>
struct vec_struct_base { std::array<T, Size> values; };
template<typename T> struct vec_struct_base<T, 2> { T x, y; };
template<typename T> struct vec_struct_base<T, 3> { T x, y, z; };
template<typename T> struct vec_struct_base<T, 4> { T x, y, z, w; };
And then it has functions and operators, where most of their implementations
use an “unroll with a labmda” style. Here’s operator that adds two vectors together:
friend VecBase operator+(const VecBase &a, const VecBase &b)
{
VecBase result;
unroll<Size>([&](auto i) { result[i] = a[i] + b[i]; });
return result;
}
with unroll itself being this:
template<class Fn, size_t... I> void unroll_impl(Fn fn, std::index_sequence<I...>)
{
(fn(I), ...);
}
template<int N, class Fn> void unroll(Fn fn)
{
unroll_impl(fn, std::make_index_sequence<N>());
}
– it takes “how many times to do the lambda” (typically vector dimension), the lambda itself,
and then “calls” the lambda with the index N times.
And then most of the functions use indexing operator to access the element of a vector:
T &operator[](int index)
{
BLI_assert(index >= 0);
BLI_assert(index < Size);
return reinterpret_cast<T *>(this)[index];
}
Pretty compact and hassle free. And given that C++ famously has “zero-cost abstractions”, these are all
zero-cost, right? Let’s find out!
Test case
Let’s do some “simple” image processing code that does not serve a practical purpose, but is simple enough to test things out.
Given an input image (RGBA, byte per channel), blur it by averaging 11x11 square of pixels around each pixel, and overlay
a slight gradient over the whole image. This input, for example, gets turned into that output:
The filter code itself is this:
inline float4 load_pixel(const uint8_t* src, int size, int x, int y)
{
x &= size - 1;
y &= size - 1;
uchar4 bpix(src + (y * size + x) * 4);
float4 pix = float4(bpix) * (1.0f / 255.0f);
return pix;
}
inline void store_pixel(uint8_t* dst, int size, int x, int y, float4 pix)
{
pix = math_max(pix, float4(0.0f));
pix = math_min(pix, float4(1.0f));
pix = math_round(pix * 255.0f);
((uchar4*)dst)[y * size + x] = uchar4(pix);
}
void filter_image(int size, const uint8_t* src, uint8_t* dst)
{
const int kFilter = 5;
int idx = 0;
float blend = 0.2f;
float inv_size = 1.0f / size;
for (int y = 0; y < size; y++)
{
for (int x = 0; x < size; x++)
{
float4 pix(0.0f);
float4 tint(x * inv_size, y * inv_size, 1.0f - x * inv_size, 1.0f);
for (int by = y - kFilter; by <= y + kFilter; by++)
{
for (int bx = x - kFilter; bx <= x + kFilter; bx++)
{
float4 sample = load_pixel(src, size, bx, by);
sample = sample * (1.0f - blend) + tint * blend;
pix += sample;
}
}
pix *= 1.0f / ((kFilter * 2 + 1) * (kFilter * 2 + 1));
store_pixel(dst, size, x, y, pix);
}
}
}
This code uses very few vector math library operations: create a float4,
add them together, multiply them by a scalar, some functions to do min/max/round.
There are no branches (besides the loops themselves), no cross-lane swizzles, fancy
packing or anything like that.
Let’s run this code in the usual “Release” setting, i.e. optimized build (-O2 on gcc/clang,
/O2 on MSVC). Processing a 512x512 input image with the above filter, on Ryzen 5950X, Windows 10,
single threaded, times in milliseconds (lower is better):
MSVC 2022
Clang 17
Clang 14
Gcc 12
Gcc 11
Release (O2)
67
41
45
70
70
Alright, Clang beats the others by a healthy margin here.
Enter Debug builds
At least within Blender (but also elsewhere), besides a build configuration that ships to the
users, during development you often work with two or more other build configurations:
“Debug”, which often means “the compiler does no optimizations at all” (-O0 on gcc/clang, /Od on MSVC).
This is the least confusing debugging experience, since nothing is “optimized out” or “folded together”.
On MSVC, people also sometimes put /JMC (“just my code” debugging), and that is default in recent
MSVC project templates. Blender uses that too in the “Debug” cmake configuration.
“Developer”, which often is the same as “Release” but with some extra checks enabled. In Blender’s case,
besides things like “enable unit tests” and “use a guarded memory allocator”, it also enables assertion checks,
and in Linux/Mac also enables Address Sanitizer (-fsanitize=address).
While some people argue that “Debug” build configuration should pay no attention to performance at all, I’m not sold
on that argument. I’ve seen projects where a non-optimized code build, while it works, produces such bad
performance that using the resulting application is an exercise in frustration. Some places explicitly enable
some compiler optimizations on an otherwise “Debug” build, since otherwise the result is just unusable (e.g. in C++-heavy
codebase, you’d enable function inlining).
However, the “Developer” configuration is an interesting one. It is supposed to be “optimized”, just with “some” extra
safety features. I would normally expect that to be “maybe 5% or 10% slower” than the final “Release” build, but not
more than that.
Let’s find out!
MSVC 2022Clang 17Clang 14Gcc 12Gcc 11
Release (O2) 67 41 45 70 70
Developer (O2 + asserts) 591 42 45 71 71
Debug (-O0 / /Od /JMC)175604965647656105942
Or, phrased in terms of “how many times a build configuration is slower compared to Release”:
MSVC 2022Clang 17Clang 14Gcc 12Gcc 11
Release (O2) 1x 1x 1x 1x 1x
Developer (O2 + asserts) 9x 1x 1x 1x 1x
Debug (-O0 / /Od /JMC)262x121x144x80x85x
On Developer config, gcc and clang are good: assertions being enabled does not cause a slowdown. On MSVC, however, this makes the code run 9 times slower. All of that
only because vector operator[](int index) has asserts in there. And it is only ever called with indices that are statically known to pass the asserts!
So much for an “optimizing compiler”, eh.
The Debug build configuration is just bad everywhere. Yes it is the worst on MSVC, but come on, anything that is more than 10 times slower than optimized code is going
into “too slow to be practical” territory. And here we are talking about things being from 80 times to 250 times slower!
What can be done without changing the code?
Perhaps realizing that “no optimizations at all produce unusably slow result” is true, some compiler developers have added an “a bit optimized, yet still debuggable”
optimization level: -Og. GCC has added that in 2013 (gcc 4.8):
-Og should be the optimization level of choice for the standard edit-compile-debug cycle,
offering a reasonable level of optimization while maintaining fast compilation and a good debugging experience.
Clang followed suit in 2017 (clang 4.0), however their -Og does
exactly the same thing as -O1.
MSVC has no setting like that, but we can at least try to turn off “just my code debugging” (/JMC) flag and see what happens. The slowdown table:
MSVC 2022Clang 17Clang 14Gcc 12Gcc 11
Release (O2) 1x 1x 1x 1x 1x
Faster Debug (-Og / /Od)114x2x2x50x14x
Debug (-O0 / /Od /JMC)262x121x144x80x85x
Alright, so:
On clang -Og makes the performance good. Expected since this is the same as -O1.
On gcc -Og is better than -O0. Curiously gcc 12 is slower than gcc 11 here for some reason.
MSVC without /JMC is better, but still very very slow.
Can we change the code to be more Debug friendly?
Current way that the math library is written is short and concise. If you are used to C++ lambda syntax, things are very clear:
friend VecBase operator+(const VecBase &a, const VecBase &b)
{
VecBase result;
unroll<Size>([&](auto i) { result[i] = a[i] + b[i]; });
return result;
}
however, without compiler optimizations, for a float4 that produces (on clang):
18 function calls,
8 branches,
assembly listing of 150 instructions.
What it actually does, is do four float additions.
Loop instead of unroll lambda
What about, if instead of this unroll+lambda machinery, we used just a simple loop?
friend VecBase operator+(const VecBase &a, const VecBase &b)
{
VecBase result;
for (int i = 0; i < Size; i++) result[i] = a[i] + b[i];
return result;
}
MSVC 2022Clang 17Clang 14Gcc 12Gcc 11
Release unroll 1x 1x 1x 1x 1x
Release loop 1x 1x 1x 3x 10x
Developer unroll 9x
Developer loop 1x
Faster Debug unroll114x2x2x50x14x
Faster Debug loop 65x2x2x31x14x
Debug unroll 262x121x144x80x85x
Debug loop 126x102x108x58x58x
This does help Debug configurations somewhat (12 function calls, 9 branches, 80 assembly instructions). However! It hurts Gcc code generation even in
full Release mode 😱, so that’s probably a no-go. If it were not for the Gcc slowdown, it would be a win: better performance in Debug configuration,
and a simple loop is easier to understand than a variadic template + lambda.
Explicit code paths for 2D/3D/4D vector cases
Out of all the possible vector math cases, 2D, 3D and 4D vectors are by far the most common. I’m not sure if other cases even happen within Blender
codebase, TBH. Maybe we could specialize those to help the compiler a bit? For example:
friend VecBase operator+(const VecBase &a, const VecBase &b)
{
if constexpr (Size == 4) {
return VecBase(a.x + b.x, a.y + b.y, a.z + b.z, a.w + b.w);
}
else if constexpr (Size == 3) {
return VecBase(a.x + b.x, a.y + b.y, a.z + b.z);
}
else if constexpr (Size == 2) {
return VecBase(a.x + b.x, a.y + b.y);
}
else {
VecBase result;
unroll<Size>([&](auto i) { result[i] = a[i] + b[i]; });
return result;
}
}
This is very verbose and a bit typo-prone however :( With some C preprocessor help it can be reduced to hide most of the ugliness inside a macro,
and then the actual operator implementation is not terribad:
friend VecBase operator+(const VecBase &a, const VecBase &b)
{
BLI_IMPL_OP_VEC_VEC(+);
}
MSVC 2022Clang 17Clang 14Gcc 12Gcc 11
Release unroll 1x 1x 1x 1x 1x
Release explicit 1x 1x 1x 1x 1x
Developer unroll 9x
Developer explicit 1x
Faster Debug unroll114x2x2x50x14x
Faster Debug explicit 19x2x2x5x3x
Debug unroll 262x121x144x80x85x
Debug explicit 55x18x30x22x21x
This actually helps Debug configurations quite a lot! One downside is that you have to have a handful of C preprocessor macros to hide away all the complexity that specializes
implementations for 2D/3D/4D vectors.
Not using C++ vector math, use C instead
As a thought exercise – what if instead of using the C++ vector math library, we went back to the C-style of writing code?
Within Blender, right now there’s guidance of “use C++ math library for new code, occasionally rewrite old code to use C++ math library” too.
That makes the code more compact and easier to read for sure, but does it have any possible downsides?
Our test image filter code becomes this then (there’s no “math library” used then, just operations on numbers):
inline void load_pixel(const uint8_t* src, int size, int x, int y, float pix[4])
{
x &= size - 1;
y &= size - 1;
const uint8_t* ptr = src + (y * size + x) * 4;
pix[0] = ptr[0] * (1.0f / 255.0f);
pix[1] = ptr[1] * (1.0f / 255.0f);
pix[2] = ptr[2] * (1.0f / 255.0f);
pix[3] = ptr[3] * (1.0f / 255.0f);
}
inline void store_pixel(uint8_t* dst, int size, int x, int y, const float pix[4])
{
float r = std::max(pix[0], 0.0f);
float g = std::max(pix[1], 0.0f);
float b = std::max(pix[2], 0.0f);
float a = std::max(pix[3], 0.0f);
r = std::min(r, 1.0f);
g = std::min(g, 1.0f);
b = std::min(b, 1.0f);
a = std::min(a, 1.0f);
r = std::round(r * 255.0f);
g = std::round(g * 255.0f);
b = std::round(b * 255.0f);
a = std::round(a * 255.0f);
uint8_t* ptr = dst + (y * size + x) * 4;
ptr[0] = uint8_t(r);
ptr[1] = uint8_t(g);
ptr[2] = uint8_t(b);
ptr[3] = uint8_t(a);
}
void filter_image(int size, const uint8_t* src, uint8_t* dst)
{
const int kFilter = 5;
int idx = 0;
float blend = 0.2f;
float inv_size = 1.0f / size;
for (int y = 0; y < size; y++)
{
for (int x = 0; x < size; x++)
{
float pix[4] = { 0,0,0,0 };
float tint[4] = { x * inv_size, y * inv_size, 1.0f - x * inv_size, 1.0f };
for (int by = y - kFilter; by <= y + kFilter; by++)
{
for (int bx = x - kFilter; bx <= x + kFilter; bx++)
{
float sample[4];
load_pixel(src, size, bx, by, sample);
sample[0] = sample[0] * (1.0f - blend) + tint[0] * blend;
sample[1] = sample[1] * (1.0f - blend) + tint[1] * blend;
sample[2] = sample[2] * (1.0f - blend) + tint[2] * blend;
sample[3] = sample[3] * (1.0f - blend) + tint[3] * blend;
pix[0] += sample[0];
pix[1] += sample[1];
pix[2] += sample[2];
pix[3] += sample[3];
}
}
float scale = 1.0f / ((kFilter * 2 + 1) * (kFilter * 2 + 1));
pix[0] *= scale;
pix[1] *= scale;
pix[2] *= scale;
pix[3] *= scale;
store_pixel(dst, size, x, y, pix);
}
}
}
MSVC 2022Clang 17Clang 14Gcc 12Gcc 11
Release unroll 1x 1x 1x 1x 1x
Release C 1x 0.5x 0.5x 1x 1x
Developer unroll 9x
Developer C 1x
Faster Debug unroll114x2x2x50x14x
Faster Debug C 4x0.5x1.5x1x1x
Debug unroll 262x121x144x80x85x
Debug C 5x6x6x4x4x
Writing code in “pure C” style makes Debug build configuration performance really good! But the more interesting thing is… in Release build on clang, this is faster than C++ code.
Even for this very simple vector math library, used on a very simple algorithm, C++ abstraction is not zero-cost!
What about SIMD?
In an ideal world, the compiler would take care of SIMD for us, especially in simple algorithms like the one being tested here. It is just number math, with very clear
“four lanes” being operated on (maps perfectly to SSE or NEON registers), no complex cross-lane shuffles, packing or any of that stuff. Just some loops and some math.
Of course, as Matt Pharr writes in the excellent ISPC blog series, “Auto-vectorization is not a programming model” (post)
(original quote by Theresa Foley).
What if we manually added template specializations to our math library, where for “type is float, dimension is 4” case it would just use SIMD intrinsics directly?
Note that this is not the right model if you want absolute best performance that also scales past 4-wide SIMD; the correct way would be to map one SIMD lane to one
“scalar item” in your algorithm. But that is a whole another topic; let’s limit ourselves to 4-wide SIMD and map a float4 to one SSE register:
template<> struct vec_struct_base<float, 4> { __m128 simd; };
template<>
inline VecBase<float, 4> operator+(const VecBase<float, 4>& a, const VecBase<float, 4>& b)
{
VecBase<float, 4> r;
r.simd = _mm_add_ps(a.simd, b.simd);
return r;
}
MSVC 2022Clang 17Clang 14Gcc 12Gcc 11
Release unroll 1x 1x 1x 1x 1x
Release C 1x 0.5x 0.5x 1x 1x
Release SIMD 0.8x 0.5x 0.5x 0.7x 0.7x
Developer unroll 9x
Developer C 1x
Developer SIMD 0.8x
Faster Debug unroll114x2x2x50x14x
Faster Debug C 4x0.5x1.5x1x1x
Faster Debug SIMD 24x1.3x1.3x2x2x
Debug unroll 262x121x144x80x85x
Debug C 5x6x6x4x4x
Debug SIMD 70x44x51x20x20x
Two surprising things here:
Even for this very simple case that sounds like “of course the compiler would perfectly vectorize this code, it is trivial!”, manually writing SIMD still wins
everywhere except on Clang.
However, in Debug build configuration, SIMD intrinsics incur heavy cost on performance, i.e. code is way slower than written in pure C scalar style. SIMD intrinsics
are still better at performance than our intial code that uses unroll+lambda style.
What about O3 optimization level?
You say “hey you are using -O2 on gcc/clang, you should use -O3!” Yes I’ve tried that, and:
On gcc it does not change anything, except fixes the curious “changing unroll lambda to a simple loop” problem, i.e. under -O3 there is no downside to
using a loop in your vector math class compared to unroll+lambda.
On clang it makes the various C++ approaches from above almost reach the performance of either “raw C” or “SIMD” styles, but not quite.
What about “force inline”?
(Update 2024 Sep 17) Vittorio Romeo asked “what about using attributes that force inlining”? I don’t have a full table, but
a quick summary is:
In MSVC, that is not possible at all. Under /Od nothing is inlined, even if you mark it as “force inline please”. There’s an
open feature request to make that possible,
but is not there (yet?).
In Clang force inline attributes help a bit, but not by much.
Learnings
All of the learnings are based on this particular code, which is “simple loops that do some simple pixel operations”. The learnings may or might not transfer
to other domains.
Clang feels like the best compiler of the three. Consistently fastest code, compared to other compilers, across various coding styles.
“C++ has zero cost abstractions” (compared to raw C code) is not true, unless you’re on Clang.
Debug build (no optimizations at all) performance of any C++ code style is really bad. The only way I could make it acceptable, while still being C++, is
by specializing code for common cases, which I achieved by… using C preprocessor macros 🤦.
It is not true that “MSVC has horrible Debug build performance”. Yes it is the worst of all of them, but the other compilers also produce really
badly performing code in Debug build config.
SIMD intrinsics in a non-optimized build have quite bad performance :(
Using “enable some optimizations” build setting, e.g. -Og, might be worth looking into, if your codebase is C++ and heavy on inlined functions, lambdas
and that stuff.
Using “just my code debugging” (/JMC) on Visual Studio has a high performance cost, on already really bad Debug build performance. I’m not sure if it is worth using
at all, ever, anywhere.
All my test code of the above is in this tiny github repo, and a PR for Blender codebase
that does “explicitly specialize for common cased via C macros” is at #127577. Whether it will
get accepted is up in the air, since it does arguably make the code “more ugly” (the PR got merged into Blender mainline).
| 2024-11-07T22:22:11 | en | train |
42,070,003 | PaulHoule | 2024-11-06T21:49:02 | When bats die, farmers use more pesticides and infant deaths rise, study shows | null | https://news.mongabay.com/2024/10/when-bats-die-farmers-use-more-pesticides-infant-deaths-rise-study-shows/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,007 | lawrencechen | 2024-11-06T21:49:17 | And now we see what is possible | null | https://geohot.github.io//blog/jekyll/update/2024/11/06/what-is-possible.html | 19 | 11 | [
42070244,
42070218,
42070305,
42070918,
42070450,
42070303
] | null | null | null | null | null | null | null | null | null | train |
42,070,048 | rtills | 2024-11-06T21:52:06 | Show HN: Take a Moment and Breathe | A simple web app to help your breathing exercises.<p>If you submit your email, I'll also ask what personalised scenes you might like to meditate to.<p>First 5 free but will have to work out a payment system after that... | https://breaths.life/ | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,058 | cryptozeus | 2024-11-06T21:52:43 | Imagination is more powerful than knowledge | null | https://rh.online/letters-blog1/may-2022-7spfw-37jth-5rmp8-yrxp9 | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,082 | sargstuff | 2024-11-06T21:55:09 | Cello – genetic circuit design automation | null | https://www.cidarlab.org/cello | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,086 | mixeden | 2024-11-06T21:55:21 | Residual Reinforcement Learning for Precise Assembly | null | https://synthical.com/article/From-Imitation-to-Refinement--Residual-RL-for-Precise-Assembly-e79408a4-e728-4f40-8052-807770da4948 | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,106 | emil_priver | 2024-11-06T21:57:03 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,070,153 | xy2_ | 2024-11-06T22:00:09 | Visualizing binary files with ImHex's DSL, the "pattern language" | null | https://xy2i.blogspot.com/2024/11/using-imhexs-pattern-language-to-parse.html | 8 | 0 | null | null | null | no_error | Visualizing binary files with ImHex's DSL, the "pattern language" | null | null |
Viewing my binary file with ImHex. The pattern language pane, on the
right, provides the highlighting and data on the left.
I've got a binary file with a custom made binary format, and a spec for that
binary format. How do I go into the binary quickly and see the data I want ?
This is a problem I ran into while writing some software that operates on this
data. Before, my approach would have probably been to write some Python code
to parse the format, following the spec carefully and see where it
diverges.
Recently though, I heard about ImHex,
an hex editor with advanced features, and decided to give it a go.The feature I used the most in ImHex for my case
was the integrated DSL, the "pattern language". It lets you define structures
that ImHex will match on, and decode the data. The syntax is a mix of C++ and Rust, but with unique semantics and a lot of nice affordances.Many examples that follow will be my real code for parsing the SWF file format.
An overview of the pattern languageHere's a pattern which parses a char, followed by the "WS" string in memory, an unsigned byte, and a four byte number after:
import type.magic;
struct Header {
char compressionSignature;
type::Magic<"WS"> signature;
u8 swfVersion;
u32 bytesSize;
};
Header h @ 0x0;
In ImHex, there are two steps to patterns:
first, you define the variable that you want to place, with a type. This can be one of the many primitive types, like u8, char, an array, or a compound type, like a struct or a bitfield.then, you can place that variable somewhere in memory, using the @ address syntax. Placing the pattern at 0x0 in memory. ImHex fills out the data in the Pattern Data pane, and shows the value in the types you chose.Patterns can be made more complex, nesting structs, or if you don't have a spec of your data and are reverse engineering, you can progressively place things, until you make more sense of it.Logic in patternsA value may exist or not based on a previous value, or even be sized differently based on a prior value that tells you the size. When parsing SWFs, I encountered this case with a bitfield, where the first field decides the size of the remaining fields. These cases are easy to write, since in the pattern language, you can reuse previous values, even if it's declared in the same structure:bitfield Rect {
// First field has a size of 5 bits
nSize : 5;
// The remaining fields have a size of nSize bits,
// based on the value of nSize.
signed xMin : nSize;
signed xMax : nSize;
signed yMin : nSize;
signed yMax : nSize;
};You can even put conditionals and other logic constructs inside. Here there are two fields indicating the length of a structure, of different size, and the second one exists only if the first has some magic value:bitfield RecordHeaderShort {
Length: 6;
Tag Tag: 10;
};
struct RecordHeader {
// Including a structure inside another one
RecordHeaderShort short;
// Defines a variable, that does not show up in the view,
// but can be accessed like a field
Tag Tag = short.Tag;
u32 len = short.Length;
if (short.Length == 0x3f) {
u32 LongLength;
len = LongLength;
}
};I defined a variable len that's present on this structure, but doesn't show up in the ImHex view. It is accesible however if RecordHeader is nested in another struct, so it's useful for getting the real length, without duplicating the logic each time.Match statementsThis feature is neat. First let's define a enum:enum Tag: u8 {
End = 0,
ShowFrame = 1,
SetBackgroundColor = 9,
DoAction = 12,
FileAttributes = 69,
Metadata = 77,
};Now I can use a Rust like match statement to match on the values extensively, defining values in the struct based on the matched value:struct TagRecord {
RecordHeader RecordHeader; // see above
match (RecordHeader.Tag) {
// To define a field, you can also just declare its type,
// there is no need to name it. This declares a field
// of type `SetBackgroundColor`.
(Tag::SetBackgroundColor): SetBackgroundColor;
(Tag::DoAction): DoAction;
(Tag::FileAttributes): FileAttributes;
(Tag::Metadata): Metadata;
// The `padding` keyword is treated specially by the language:
// Creates a padding, with its length set to
// a field of the RecordHeader structure. This will not show up in
// the pattern view, because it's declared as padding.
(_): padding[RecordHeader.len];
}
// An attribute. In the data pane, this structure's name will be
// the enum's name.
} [[name(RecordHeader.Tag)]];In the data view, each tag shows up with the enum's name.Array stop conditionsAn issue I encountered was that my file format is just an array of TagRecords, until the end of the file. This means there is no length: and using a null terminated array doesn't work, since there are tags with null values. Thankfully, the pattern language has support for custom stop conditions, using a "loop sized array":struct File {
Header;
// This function returns true when
// the end of the file has been reached
TagRecord Tags[while(!std::mem::eof())];
};ImHex also has a special $ (dollar) operator, which always points to the current offset in the file. This variable is even modifiable inside your pattern to change the position. One good use for this is a loop sized array like this one, that will stop on a 0xFF value:u8 string[while(std::mem::read_unsigned($, 1) != 0xFF)];Decompressing files inside ImHex directly Most SWF files in the wild come compressed, with a non trivial decompression procedure, which means my pattern can't be applied directly: I need a decompression step first. In the past I would have reached for a Python script to decompress the file with the custom logic needed.But the pattern language also supports this case, and has built-in decompression functions, as well as a virtual file system to store the output file:struct Compressed {
// Create a section in memory for the decompressed contents. Only contents
// starting from a certain offset are compressed
UncompressedHeader h;
std::mem::Section decompressed = std::mem::create_section("Zlib decompressed");
u8 compressedContents[std::mem::size() - 8] @ 0x8;
// Do the actual decompression.
hex::dec::zlib_decompress(compressedContents, decompressed, 15);
// Combine both the uncompressed header and decompressed part to recreated
// the full decompressed file.
std::mem::Section full = std::mem::create_section("Full decompressed SWF");
std::mem::copy_value_to_section(h, full, 0x0);
std::mem::copy_section_to_section(decompressed, 0x0, full, 0x8, h.decompressedSize - 8);
// Write decompressed flag, so next time we open the pattern knows it's decompressed
std::mem::copy_value_to_section("Z", full, 0x0);
// Write this file to the virtual file system.
u8 d[h.bytesSize] @ 0x00 in full;
builtin::hex::core::add_virtual_file(std::format("dec-{}", hex::prv::get_information("file_name")), d);
std::warning("This SWF is ZLib-compressed, grab the decompressed save from\nthe Virtual Filesystem tab and use this pattern on it.");
};
struct Main {
char compressed @ 0x0;
match (compressed) {
('C'): Compressed;
('Z'): File;
}
};By running this pattern once, the resulting file ends up in a new tab, in ImHex's "virtual filesystem".The decompressed file in the virtual file systemConclusionOverall I was very impressed by the pattern language. Not only does it make my task much simpler, it also is well thought out for the tasks it sets out to do,, both in its design and in its day to day conveniences.However, currently the language doesn't have a lot of documentation or tutorials to get started. Thankfully it was easy to get started with some help, the developers on discord are very nice (and thanks again to paxcut for helping debug some issues!) Hopefully this post was interesting even if you've never had this problem before :)You can find my full code here, if you're curious of what it looks like all put together.Discussion on reddit, hn, lobsters
| 2024-11-08T10:24:43 | en | train |
42,070,231 | haunter | 2024-11-06T22:08:19 | War Is a Racket | null | https://en.wikipedia.org/wiki/War_Is_a_Racket | 1 | 0 | null | null | null | missing_parsing | War Is a Racket | 2004-06-15T02:37:50Z | Contributors to Wikimedia projects |
From Wikipedia, the free encyclopedia
War Is a Racket 1935 cover from the first printingAuthorSmedley D. Butler(Major general (Ret.), USMC)LanguageEnglishSubjectMilitary–industrial complexMoralityWarfarePublisherRound Table PressPublication date1935Publication placeUnited StatesPages51 (first edition)ISBN9780922915866OCLC3015073Dewey Decimal172.4LC ClassHB195 .B8[1]
War Is a Racket is a speech and a 1935 short book by Smedley D. Butler, a retired United States Marine Corps major general and two-time Medal of Honor recipient.[2][3] Based on his career military experience, Butler discusses how business interests commercially benefit from warfare. He had been appointed commanding officer of the Gendarmerie during the 1915–1934 United States occupation of Haiti.
After Butler retired from the US Marine Corps in October 1931, he made a nationwide tour in the early 1930s giving his speech "War Is a Racket". The speech was so well received that he wrote a longer version as a short book published in 1935. His work was condensed in Reader's Digest as a book supplement, which helped popularize his message. In an introduction to the Reader's Digest version, Lowell Thomas, who wrote Butler's oral autobiography, praised Butler's "moral as well as physical courage".[4]
In War Is a Racket, Butler points to a variety of examples, mostly from World War I, where industrialists, whose operations were subsidized by public funding, were able to generate substantial profits, making money from mass human suffering.
The work is divided into five chapters:
War is a racket
Who makes the profits?
Who pays the bills?
How to smash this racket!
To hell with war!
It contains this summary:
War is a racket. It always has been. It is possibly the oldest, easily the most profitable, surely the most vicious. It is the only one international in scope. It is the only one in which the profits are reckoned in dollars and the losses in lives. A racket is best described, I believe, as something that is not what it seems to the majority of the people. Only a small "inside" group knows what it is about. It is conducted for the benefit of the very few, at the expense of the very many. Out of war a few people make huge fortunes.
Butler confesses that during his decades of service in the United States Marine Corps:
I helped make Mexico, especially Tampico, safe for American oil interests in 1914. I helped make Haiti and Cuba a decent place for the National City Bank boys to collect revenues in. I helped in the raping of half a dozen Central American republics for the benefits of Wall Street. The record of racketeering is long. I helped purify Nicaragua for the international banking house of Brown Brothers in 1909–1912 (where have I heard that name before?). I brought light to the Dominican Republic for American sugar interests in 1916. In China I helped see to it that Standard Oil went its way unmolested.
In the booklet's penultimate chapter, Butler recommends three steps to disrupt the war racket:
Making war unprofitable. Butler suggests that the means for war should be "conscripted" before those who would fight the war: It can be smashed effectively only by taking the profit out of war. The only way to smash this racket is to conscript capital and industry and labour before the nation's manhood can be conscripted. […] Let the officers and the directors and the high-powered executives of our armament factories and our steel companies and our munitions makers and our ship-builders and our airplane builders and the manufacturers of all other things that provide profit in war time as well as the bankers and the speculators, be conscripted — to get $30 a month, the same wage as the lads in the trenches get.
Acts of war to be decided by those who fight it. He also suggests a limited referendum to determine if the war is to be fought. Eligible to vote would be those who risk death on the front lines.
Limitation of militaries to self-defense. For the United States, Butler recommends that the Navy be limited, by law, to operating within 200 miles of the coastline, and the Army restricted to the territorial limits of the country, ensuring that war, if fought, can never be one of aggression.
Arms industry
Merchants of death
Military–industrial complex
Perpetual war
Confessions of an Economic Hit Man
^ "Item Information – War Is a Racket". United States Library of Congress. Retrieved 1 July 2018.
^ "Major General Smedley D. Butler, USMC (Deceased)". Who's Who in Marine Corps History. History Division, United States Marine Corps. Archived from the original on 20 March 2014. Retrieved 28 July 2024.
^ "Smedley Darlington Butler". Hall of Valor. Military Times. Retrieved 28 July 2024.
^ Thomas, Lowell (1933). Old Gimlet Eye: Adventures of Smedley D. Butler. Farrar & Rinehart.
War Is a Racket public domain audiobook at LibriVox
War Is a Racket at the Internet Archive
War Is a Racket (online version)
War Is a Racket
Scanned copy of the original 1935 printing
"War Is A Racket" by Maj. Gen. Smedley D. Butler, read By Jon Gold
"War is still 'a Racket'", Foreign Policy Magazine column by David James
| 2024-11-08T20:21:19 | null | train |
42,070,232 | NavinF | 2024-11-06T22:08:21 | Starship – Fifth Flight Test (SpaceX channel) [video] | null | https://www.youtube.com/watch?v=hI9HQfCAw64 | 1 | 0 | null | null | null | no_article | null | null | null | null | 2024-11-08T00:57:16 | null | train |
42,070,240 | emot | 2024-11-06T22:08:56 | Exploring Internet traffic shifts and cyber attacks during the 2024 US election | null | https://blog.cloudflare.com/exploring-internet-traffic-shifts-and-cyber-attacks-during-the-2024-us-election/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,255 | edward | 2024-11-06T22:10:10 | Starship flight test 6 – November 18, 2024, at 22:00 UTC | null | https://en.wikipedia.org/wiki/Starship_flight_test_6 | 2 | 1 | [
42071145
] | null | null | null | null | null | null | null | null | null | train |
42,070,262 | ta8645 | 2024-11-06T22:10:29 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,070,263 | sandwichsphinx | 2024-11-06T22:10:34 | Managing Cryptography with CBOMkit | null | https://research.ibm.com/blog/quantum-safe-cbomkit | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,291 | mdp2021 | 2024-11-06T22:12:28 | U.S. Election Is Over. EFF Is Ready for What's Next | null | https://www.eff.org/deeplinks/2024/11/2024-us-election-over-eff-ready-whats-next | 8 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,293 | talboren | 2024-11-06T22:12:36 | OpenAI bought the web domain Chat.com | null | https://www.engadget.com/ai/openai-bought-the-web-domain-chatcom-213638986.html | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,307 | springjben | 2024-11-06T22:13:36 | Don't Work at My Company | null | https://benspring.com/p/working-at-tryhackme-isnt-for-everyone | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,320 | ishannagpal | 2024-11-06T22:14:33 | Let Claude control your computer (at your own risk) | null | https://github.com/suitedaces/computer-agent | 2 | 0 | null | null | null | no_error | GitHub - suitedaces/computer-agent: Desktop app powered by Claude’s computer use capability to control your computer | null | suitedaces | 👨🏽💻 Grunty
Self-hosted desktop app to have AI control your computer, powered by the new Claude computer use capability. Allow Claude to take over your laptop and do your tasks for you (or at least attempt to, lol). Written in Python, using PyQt.
Demo
Here, I asked it to use vim to create a game in Python, run it, and play it.
grunty.8x.compressed.mp4
Video was sped up 8x btw. Computer use is pretty slow as of today.
⚠️ Important Disclaimers
This is experimental software - It gives an AI control of your mouse and keyboard. Things can and will go wrong.
Tread Lightly - If it wipes your computer, sends weird emails, or orders 100 pizzas... that's on you.
Anthropic can see your screen through screenshots during actions. Hide sensitive information or private stuff.
🎯 Features
Literally ask AI to do ANYTHING on your computer that you do with a mouse and keyboard. Browse the web, write code, blah blah.
💻 Platforms
Anything you can run Python on: MacOS, Windows, Linux, etc.
🛠️ Setup
Get an Anthropic API key here.
# Python 3.10+ recommended
python -m venv venv
source venv/bin/activate # or `venv\Scripts\activate` on Windows
pip install -r requirements.txt
# Add API key to .env
echo "ANTHROPIC_API_KEY=your-key-here" > .env
# Run
python run.py
🔑 Productivity Keybindings
Ctrl + Enter: Execute the current instruction
Ctrl + C: Stop the current agent action
Ctrl + W: Minimize to system tray
Ctrl + Q: Quit application
💡 Tips
Claude really loves Firefox. You might want to install it for better UI detection and accurate mouse clicks.
Be specific and explicit, help it out a bit
Always monitor the agent's actions
🐛 Known Issues
Sometimes, it doesn't take a screenshot to validate that the input is selected, and types stuff in the wrong place.. Press CMD+C to end the action when this happens, and quit and restart the agent. I'm working on a fix.
🤝 Contributing
Issues and PRs are most welcome! Made this is in a day so don't really have a roadmap in mind. Hmu on Twitter @ishanxnagpal if you're got interesting ideas you wanna share.
📄 License
Apache License 2.0
| 2024-11-08T04:10:18 | en | train |
42,070,367 | bookofjoe | 2024-11-06T22:17:46 | Murray McCory, 80, Dies; JanSport Founder Created the School Backpack | null | https://www.nytimes.com/2024/11/04/business/murray-mccory-dead.html | 3 | 1 | [
42070374
] | null | null | null | null | null | null | null | null | null | train |
42,070,415 | RyeCombinator | 2024-11-06T22:21:05 | Overengineering this blog's preview site with Kubernetes | null | https://xeiaso.net/blog/2024/overengineering-preview-site/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,431 | gslin | 2024-11-06T22:21:55 | IPTV Piracy Blocking at the Internet's Core Routers Undergoes Testing | null | https://torrentfreak.com/iptv-piracy-blocking-at-the-internets-core-routers-undergoes-testing-241106/ | 2 | 1 | [
42070626
] | null | null | no_error | IPTV Piracy Blocking at the Internet's Core Routers Undergoes Testing * TorrentFreak | null | null |
During 2010/2011, opportunity arose for Hollywood to convince the High Court in London that site-blocking would be a proportionate response to tackle a single Usenet indexing site called Newzbin.
As rightsholders offered assurances that the action would be carefully targeted and strictly limited in scope, the requested injunction was granted in October 2011. Within 14 days, ISP BT would implement blocking to prevent six million customers from accessing the site in the UK. That was a landmark win for the studios; it also laid the foundations for something bigger.
Whether the High Court would’ve acted any differently is unclear, but it certainly wasn’t informed in advance that its decision would effectively seed site-blocking on a global scale, while acting as an official seal of approval.
Not only did the injunction eventually lead to the blocking of tens of thousands of domains locally, the High Court’s decision was used to convince courts all around the world to do the same. Even in countries where blocking already assists mass censorship, governments are routinely encouraged to block more than they do already.
Of course, it’s never enough. Blocking is easily circumvented, which prompts calls for even more blocking. When faster blocking fails to produce results, preemptive and in some cases perpetual blocking is now accepted as normal. News that testing is underway, to block pirate IPTV devices by meddling with the internet’s core routers, is certainly depressing. What it definitely is not, however, is any kind of surprise.
Brazil Embraces Blocking
When Elon Musk and a Brazilian judge became embroiled in a bitter dispute over what can (and cannot) be said online, Brazil’s Supreme Court ordered local ISPs to take action. Using tools developed in recent years to block pirate sites and piracy-configured set-top boxes, the entire X platform was rendered inaccessible in Brazil. When a blocking mechanism is so readily available, the likelihood of it being used to stifle dissent is just a button press away.
Urged on by movie studios in the United States and the global recording industry, Brazil has now fully embraced site-blocking as a convenient anti-piracy solution. Courts have been issuing orders with such frequency it’s now almost impossible to keep up. Details of the entities subjected to blocking aren’t for public consumption, a common trait of site-blocking systems which prevents accountability.
Yet, those who somehow gain access to the blocklist will discover it currently contains around 11,800 domains. The majority are related to piracy and at some others concern outlawed gambling platforms that don’t appear on Brazil’s official whitelist.
The whitelist approach also applies to Android-style set-top boxes. All such devices are now illegal by default, pending state certification authorizing their use.
Building/Blocking Communications Infrastructure
Ensuring that only authorized platforms and devices are accessible in Brazil falls to telecoms regulator Anatel.
In an interview with Tele.Sintese, outgoing Anatel board member Artur Coimbra recalls the lack of internet infrastructure in Brazil as recently as 2010. As head of the National Broadband Plan under the Ministry of Communications, that’s something he personally addressed. For Anatel today, blocking access to pirate websites and preventing unauthorized devices from communicating online is all in a day’s work.
“The topic of combating piracy has evolved significantly. We can already see the impact of this work on customer satisfaction indicators for IPTV boxes. Pirate box brands are receiving worse reviews as time goes by,” Coimbra says.
“This means that the service is getting worse, users are becoming more dissatisfied, and as a result, one day they will no longer use that pirated service. This is a great indicator of the work that Anatel has been doing.”
Automated Site-Blocking Incoming
While blocking a few sites, services, or devices can be managed manually, Coimbra says that automation is the preferred option.
“Today, orders to block pirate boxes are issued manually. We work on call and send the orders to the operators. The operators receive this and implement the IP blocking,” he explains.
“What we are going to do at this point is that these orders will no longer be manual, they will have a common system in which everyone [operators and providers] will have access to the system at the same time.”
While it can be argued that manual systems are prone to errors, automated systems are designed to need much less oversight. Whether that means fewer checks and balances remains to be seen. In general, however, limited oversight is considered a plus in the world of site-blocking.
Oversight Makes Blocking Less Efficient
After many years of putting Brazil under enormous pressure to block pirate sites, the current system involving the courts now blocks thousands of them. Yet in a typical display of incremental demands for improvement, rightsholders now want more.
In a January report to the USTR (pdf), major rightsholders urged ANATEL to “implement an effective system to tackle online piracy within Internet applications and sites based on Bill of Law #3696/2023, which was signed by the President on January 15, 2024, and sets forth an administrative site-blocking provision.”
When a country’s ISPs receive their first request to block a single pirate site, using carefully targeted, strictly limited measures under the supervision of the courts, administrative blocking of tens of thousands of sites is the long-term goal.
This often means blocking measures discussed behind closed doors between mostly commercial entities, with limited or even no oversight from local courts. This is the system preferred by major rightsholders but having inspired Brazil to do more, why should it stop there?
Targeting the Internet’s Backbone
In broad terms, the ‘internet backbone’ is the core infrastructure that combines to form the foundations of the global internet. It is comprised of the fastest, most capable networks, and data travels via high capacity fiber-optics and advanced ‘core’ routers. Operated by commercial companies, government, military and educational institutions, effective backbone networks are critical to the functioning of the wider internet.
Considering the ongoing crisis in Italy where the Piracy Shield system has already caused considerable damage with nothing like the same level of access, the idea of messing with the backbone of the internet seems like a bad dream; now it’s wake-up time.
“The second step, which we still need to evaluate because some companies want it, and others are more hesitant, is to allow Anatel to have access to the core routers to place a direct order on the router,” Coimbra reveals, referencing IPTV blocking.
“In these cases, these companies do not need to have someone on call to receive the [blocking] order and then implement it.”
Already Targeting the Internet’s Backbone
Thanks to the interviewer at Tele.Sintese pressing Coimbra after his initial response, what initially sounds like a plan for the future is suddenly revealed as already underway. Will Anatel really access core routers to block IP addresses used for piracy?
“Companies that deem it convenient can give us limited access, not full access, so that we can perform these blocks directly for non-certified and non-approved equipment. Limited access so that we can only perform these blocks remotely. It would be a kind of virtual seal,” Coimbra adds.
Recall, all devices (mostly Android devices) are illegal without certification, regardless of whether they’re configured for piracy or not. So how long before something like this is actually implemented?
“Participation is voluntary. We are still testing with some companies. So, it will take some time until it actually happens,” Coimbra says. “I can’t say [how long]. Our inspection team is carrying out tests with some operators, I can’t say which ones.”
Limited to Brazil? Or Not…
Most likely quite surprised at the revelation, the interviewer inquires whether this is also happening in other countries.
“I don’t know. Maybe in Spain and Portugal, which are more advanced countries in this fight. But I don’t have that information,” Coimbra responds, randomly naming two countries with which Brazil has consulted extensively on blocking matters.
“It’s critical infrastructure, so it has to be done with great care, with a limited scope. That’s why it has to have the support of the company that feels comfortable,” he concludes.
Blocking with great care and with limited scope are the same arguments presented in London during 2010/11. Due to a lack of transparency, how many domains and IP addresses are currently blocked around the world is impossible to say. That means that it’s impossible to say whether it’s carried out with great care or not.
As for limited scope, Brazil doesn’t appear to be targeting all core routers at the moment. Spain and Portugal, of which nothing is known, may or may not have tested anything at all. By definition, then, the scope is indeed limited and arguing otherwise risks being portrayed as an alarmist who lacks respect for the creative industries.
| 2024-11-08T17:25:52 | en | train |
42,070,457 | mfiguiere | 2024-11-06T22:24:18 | iRobot lays off another 105 employees | null | https://techcrunch.com/2024/11/06/irobot-lays-off-another-105-employees/ | 3 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,459 | raykyri | 2024-11-06T22:24:31 | How the Trump Whale Correctly Called the Election | null | https://www.wsj.com/finance/how-the-trump-whale-correctly-called-the-election-cb7eef1d | 5 | 0 | [
42070941
] | null | null | null | null | null | null | null | null | null | train |
42,070,466 | zicxor | 2024-11-06T22:24:58 | Show HN: Reddit cold outreach platform as mobile app | Introducing Allwent - Your ultimate Reddit cold outreach platform!<p>Perfect for marketers, recruiters, and anyone looking to expand their Reddit connections. Allwent helps you find relevant posts and engage with potential collaborators.<p>Note: This app adheres to Reddit's terms of service. Use responsibly and respect user privacy.<p>App link: <a href="https://apps.apple.com/us/app/allwent/id6475793145" rel="nofollow">https://apps.apple.com/us/app/allwent/id6475793145</a> | null | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,489 | hhs | 2024-11-06T22:26:38 | Sin taxes are suffering from a shortage of sinners | null | https://www.economist.com/finance-and-economics/2024/10/31/sin-taxes-are-suffering-from-a-shortage-of-sinners | 34 | 40 | [
42070830,
42070992,
42070901,
42070820,
42071348,
42071436,
42071379,
42071135,
42071322,
42071262,
42071117,
42071058
] | null | null | null | null | null | null | null | null | null | train |
42,070,490 | rpgbr | 2024-11-06T22:26:40 | New AI experiences for Paint and Notepad begin rolling out to Windows Insiders | null | https://blogs.windows.com/windows-insider/2024/11/06/new-ai-experiences-for-paint-and-notepad-begin-rolling-out-to-windows-insiders/ | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,499 | jjgreen | 2024-11-06T22:27:14 | The Perfect AI Companion | null | https://cinemasojourns.com/2024/11/06/the-perfect-ai-companion/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,515 | null | 2024-11-06T22:28:23 | null | null | null | null | null | null | [
"true"
] | null | null | null | null | null | null | null | null | train |
42,070,540 | ZeljkoS | 2024-11-06T22:30:06 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,070,544 | lyricsongation | 2024-11-06T22:30:28 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,070,574 | davemoore | 2024-11-06T22:32:47 | We Need a New IfC Approach to Meet HPC Challenges | null | https://thenewstack.io/why-we-need-a-new-ifc-approach-to-meet-hpc-challenges/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,592 | Michelangelo11 | 2024-11-06T22:34:18 | Revenge of the Silent Male Voter | null | https://quillette.com/2024/11/06/the-revenge-of-the-silent-male-voter-trump-vance-musk/ | 6 | 6 | [
42071294,
42070839
] | null | null | null | null | null | null | null | null | null | train |
42,070,702 | peter_d_sherman | 2024-11-06T22:43:05 | About Those Gaussian Envelopes (2014) | null | http://marty-green.blogspot.com/2014/01/about-those-gaussian-envelopes.html | 1 | 1 | [
42070857
] | null | null | null | null | null | null | null | null | null | train |
42,070,717 | EMIRELADERO | 2024-11-06T22:44:44 | iSH, JIT, and EU | null | https://ish.app/blog/ish-jit-and-eu | 3 | 0 | null | null | null | no_error | iSH | null | null |
iSH, JIT, and EU
2024-10-16
Earlier this year, Apple announced a sweeping set of changes to iOS aimed at bringing the platform in compliance with the European Union’s Digital Markets Act. Naturally, such a significant reform to app capabilities and distribution has left some wondering what this means for iSH users in the EU. We’ve been working to answer this question and have some updates to share.
While Apple’s DMA concessions are broad, the parts most relevant to iSH are new APIs to support just-in-time (JIT) compilation of code. Hardware support for this feature has been present in every iPhone going back to the original one released in 2007, but it took until 2011 for Apple to start taking advantage of it in their own software. That software is, of course, Apple’s Safari browser engine–the first and to date only software on iPhone that is permitted to use JIT compilation technology. Until now, that is.
JIT compilation plays a key part of high performance interpreters–including the ones present in every modern web engine–and Safari’s WebKit is no exception. While Apple has had a long-standing policy to disallow shipping of alternative browser engines on iOS, gating JIT to system webviews provides a technological barrier to using one, effectively dooming their interpreters to be unable to compete on performance with Apple’s. To level the playing field, the EU has asked Apple to make this functionality available to others. For the first time, third-party iOS apps will be able to use JIT compilation if Apple deems them in compliance with their requirements for browsers. It is unclear what the exact policy is for granting these requests, given that there do not appear to be any apps that are using this API yet, but we leave those questions to be answered by other browser vendors.
Apple’s JIT API is clearly designed to support browsers, and in many ways mirrors the architecture of their own engine. However, the JIT technology it exposes is generally useful to any interpreter, including the one that powers iSH. For the past several months, we have been privately evaluating whether we can use this API in our app. This has proven challenging: officially, even developing against this API officially requires permission from Apple, and iSH does not meet the requirements for us to make such a request. Still, we’ve been able to use workarounds to test some prototypes. This has allowed us to confirm that such an approach would significantly improve the performance of iSH–in some cases, by an order of magnitude. This is in line with academic results and findings from other related projects, and thus is not particularly surprising.
Under the DMA, third parties can lodge requests for interoperability with gatekeepers (Apple, in this case) to request free and fair access to hardware or software features available to the platform’s owner. From our testing, we were able to demonstrate that the requisite JIT features we needed are in fact already available and useful to iSH. The only obstacle preventing us from using them, besides updating our app with support, is Apple’s policy limiting its use to browsers.
In July, we filed an interoperability request with Apple to expand the scope of JIT compilation to non-browser apps. A full copy of the request we submitted can be viewed here. Unfortunately, after two months of deliberation Apple has denied our request. This means we cannot provide JIT support in iSH to our users in the EU at this time.
This decision is regrettable, as we feel that JIT compilation would be unambiguously positive for iSH. This feature is a major win for performance and battery life. In addition, we’ve already done a security review of our proposed implementation and found that it meets or exceeds Apple’s requirements for access to this feature. And, ultimately, we believe that Apple’s rejection ties the hands of third party developers looking to ship interesting software for their platforms. We get a lot of feedback from our users and it’s been our top feature request since we started working on the app.
Ripping out the asbestos
Of course, most of these requests do not actually arrive in the form of “please add a just-in-time compilation tier” (though, perhaps surprisingly, some do). Instead, we get many variants–using more colorful language than others–along the lines of “Why is iSH so slow? Can you make it faster?” The concerns aren’t misplaced: iSH typically benchmarks at 5-100x slower than native code, depending on workload. This performance hit is inherent to the environment iSH operates in. JIT compilation would unlock powerful ways to cut down on overhead, but to see why, it’s important to understand how iSH works today.
The premise of iSH is a simple one: it provides access to a command-line shell on your iPhone, one that runs locally and supports all the software you expect to use. While conceptually straightforward, its implementation is anything but. On a personal computer the shell has broad privileges by design. As an extension of the user, it can split itself into multiple processes, read and write directly to the filesystem, and communicate with other applications. Most importantly, it can execute code: not just some prepared commands, but real code, written by the user or downloaded from the internet or generated from yet another program. This is, to put it plainly, not how phones work. An app runs in a strict sandbox, limiting its capabilities. On iOS, an app cannot spawn multiple processes; it is assigned a narrow slice of the disk with which to work with. And critically, it cannot execute new native code: the OS prohibits it.
iSH, of course, works anyway. It leans into the restrictions, offering a convenient “Linux in an app” that is isolated from the rest of the device, so it cannot break anything, but still accessible enough to get real work done. It achieves this by turning to a standard technique in computer science: emulation. Two layers of it, in fact, each working around different limitations on iOS.
The first layer of emulation is the kernel layer. Apps cannot spawn new processes; they cannot schedule them, nor can they offer or gate access to resources. This is traditionally the job of an operating system’s kernel, and iOS already has its own. In iSH, we emulate our own kernel, translating the needs of the programs we run and wrapping them into requests for the real iOS kernel to service for us. Our translation chooses to implement a Linux interface, which is vaguely similar to the XNU kernel that iOS uses. This lets us map Linux processes to groups of iOS threads and forward file operations to a location inside the app sandbox. The process is very similar to Wine, or WSL1.
Our second emulation layer works around iOS restrictions on executing native code. We can’t run the code directly, so we don’t: we interpret it, as a script, using a custom interpreter called Asbestos. It’s largely written in assembly and uses techniques such as direct threading and branch prediction for speed. For the interpreter, the architecture being emulated is largely irrelevant, so 32-bit x86 was chosen to balance ease of implementation with software compatibility.
Emulation is an effective way to bridge dissimilar systems. It’s a wonder iSH works at all, really, but it does, in a way that is useful to many thousands of people. But it’s not free. This translation has overhead, and in the case of iSH the overhead is quite severe. There are still some ways we can speed up the interpreter, which we will continue to implement (as time permits–making fast interpreters is a billion dollar problem, and we’re just hobbyists). But those improvements are nebulous and have diminishing returns.
Just-in-time compilers substantially reduce emulation overhead by generating native code that can run at nearly full speed. There is still some extra work required to do this translation, but the gains are typically quite substantial for most applications. This is the reason why all major browsers on every modern platform have a JIT compiler for interpreting JavaScript code, or why so much research is put into optimizations applying just-in-time compilers to other languages. Apple uses it for accelerating not only JavaScript but WebAssembly, CSS, and regular expressions in their Safari browser on their mobile and desktop OSes. It also forms a critical part of their Rosetta 2 emulator on macOS, which is very similar in design to iSH’s interpreter. And Swift Playgrounds for iPadOS uses a limited form of native code generation to ensure the performance of apps it builds–a capability no other app on the platform enjoys.
We can’t speak to exact numbers because it’s not easy to test a JIT on iOS. However, our tests show easy wins of 2-5x on almost all tasks, even with very simple code generation. Depending on the implementation–we took Apple’s BrowerEngineKit as an example–we can improve performance even further by optimizing address translation and instruction lowering. The ability to generate native code just-in-time is unquestionably a major performance win for iSH, and essential if it is to compete with other scripting runtimes that have access to these features.
Industry-leading security
The DMA provides gatekeepers broad latitude to prioritize platform security over interoperability requests. Apple has repeatedly claimed that they believe the changes they are being asked to make in the EU compromise the safety and privacy of their users. Answering the question “is [x] secure” is never straightforward, and evaluating just-in-time compilers is no exception. Still, we have thoroughly reviewed iSH’s proposed implementation and we believe we can say that it incontrovertibly matches or surpasses comparable efforts, including Apple’s own.
Evaluation of the security properties of native code is a complex subject that is out of scope for this post, but Saagar has written a few words on the topic of to threat modeling against code generation. We will instead focus here on the specific security measures iSH would employ.
iSH, like all third-party apps on iOS, is subject to the platform sandbox. This serves as the first line of defense against malicious software, and it’s a very effective one: barring exploits or the occasional oversight, apps cannot strew cruft all over your device, trivially access your private data, or hog system resources. As far as we are aware, there are no serious calls to remove this sandbox, nor do we think it should go anywhere. iSH remains sandboxed today and we fully intend for it to be sandboxed forever. This is a valuable security feature which would be foolish to squander.
Apple’s architecture for just-in-time compilers limits code generation to heavily locked-down, isolated subprocesses of the main app. We welcome the change and wholly support this design. It confers useful security properties while also simplifying bookkeeping. In fact, we would suggest the following changes to the model to make it even more secure: the new processes actually have access to too much, as they are essentially a copy of Safari’s profile for its web content process. Third party browsers have no need to talk to as many things as Safari does, and iSH has fewer needs still. Locking down the surface area would increase the security of this design. In addition, while not quite related to just-in-time compilation, it seems reasonable to offer the process isolation model to any app, not just those needing code generation. Many apps handle untrusted content and perform complicated parsing that can open the door to dangerously easy compromises if done in the same address space as critical user data. Apple already splits off these tasks into specialized “quarantined” processes for its own software, and it’s overdue to make this functionality available to third parties as well. In fact, iSH is less likely to need such functionality, because it is less likely to be used in targeted attacks but also because it processes less untrusted content. Of course, we plan to use this design anyway because it’s good security hygiene.
Unlike today’s major browsers, iSH is a small project. It’s a lot simpler, and more amenable to adopting best practices for writing secure software. We aim for the iSH JIT to be written in a memory safe language, which would put it well ahead of attempts by the industry in this area. This would entirely eliminate or greatly eliminate a large source of vulnerabilities in our software.
The anti-Sherlock
In late September, we finally received a reply to our interoperability request. The response was terse but the actual rationale was shorter still:
After review, we have determined that it does not fall in scope of article 6(7) DMA. Apple does not itself offer emulation functionalities on iOS and it does not offer JIT compilation for non-browser apps on iOS.
There is a term in the Apple community that refers to the all-too-common practice of an indie app suddenly finding that their software has become a bullet point in the latest OS release: Sherlocking. Despite likely holding the dubious honor of the app that its users most want to be Sherlocked, iSH’s core value proposition has never truly been integrated into the platform. iSH uses iOS technologies but it does not compete with what iOS offers, because Apple does not provide comparable functionality. It’s unique. Or at least as unique as it can be in a market with a dozen terminal emulator apps with different tradeoffs of their own.
The rejection rationale is truly an anti-Sherlock: instead of Apple deciding to compete with our product, they believe that they do not have to offer their technology to us because they don’t compete with our app. They say, plain as day, that they do not offer emulation facilities on iOS. They don’t use JIT except for browsers and so nobody may use JITs except for browsers. The technology is there. The hardware for it is present, the software for it is already written. The only thing that prevents its use in iSH is that Apple will not grant us an entitlement to use the API. And as they have said, they do not grant this to any app that is not a browser, because they do not use it for anything that is not a browser.
Whether this is a valid response in light of article 6(7) of the DMA, we do not know. That is for the European Commission to decide. Our request was obviously submitted in good faith and is, to our understanding, valid. On the other hand, Apple feels that they are not required to expose this API because they do not compete in this area.
Perhaps this is true. In any case, however, it is somewhat disheartening to see this response. Apple does not think they have to give us access to this API because they do not compete in this area. But perhaps Apple does not feel they have to compete in this area because we do not have access to this API. We’ve poured a lot of love and sweat into making iSH. If what we can make is only “ok”, and never great, does that mean this market can be held permanently depressed by the gatekeeper never entering it? Not needing to enter, because everything can be limited to never being worth competing against?
We love working on iSH, because we love working on cool things. iSH is unassailably, indubitably cool. But it would certainly be cooler if it was great.
| 2024-11-08T00:43:00 | en | train |
42,070,722 | luu | 2024-11-06T22:45:10 | Fighting Crime or Raising Revenue? [pdf] | null | https://ij.org/wp-content/uploads/2019/06/Fighting-Crime-or-Raising-Revenue-7.20.2020-revision.pdf | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,729 | PaulHoule | 2024-11-06T22:45:23 | Can Indigenous urbanism counter Vancouver's pressing lack of housing? | null | https://english.elpais.com/economy-and-business/2024-10-26/can-indigenous-urbanism-counter-vancouvers-pressing-lack-of-housing.html | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,738 | mooreds | 2024-11-06T22:46:11 | Let's Ease into the Holiday Season | null | https://ashleyjanssen.com/lets-ease-into-the-holiday-season/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,740 | whack | 2024-11-06T22:46:19 | The Anti-Anxiety Election Day Menu | null | https://www.nytimes.com/2024/11/05/opinion/election-day-anti-anxiety-menu.html | 1 | 0 | [
42070933
] | null | null | null | null | null | null | null | null | null | train |
42,070,764 | Turboblack | 2024-11-06T22:48:05 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,070,770 | davidgomes | 2024-11-06T22:48:15 | Just Have AI Build an App for That | null | https://davidgomes.com/just-have-ai-build-an-app-for-that/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,793 | rogerbukuru | 2024-11-06T22:50:21 | What do you think of Building AI with AI? | null | https://www.envole.ai/ | 2 | 0 | [
42070795
] | null | null | null | null | null | null | null | null | null | train |
42,070,807 | aard | 2024-11-06T22:51:35 | Writing Tiny Desktop Apps in C and GTK3 | null | https://danielc.dev/blog/tiny-windows-linux-gtk-apps/ | 3 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,808 | mangoseo | 2024-11-06T22:51:42 | Show HN: Heterogeneous writer that passes AI detectors | null | https://sigmaseo.io/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,812 | handfuloflight | 2024-11-06T22:51:49 | ElevenLabs: Conversational AI | null | https://elevenlabs.io/conversational-ai | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,832 | fabiofederici | 2024-11-06T22:53:22 | Empowering AI Agents on Solana | null | https://twitter.com/glamsystems/status/1854253062208233574 | 1 | 1 | [
42070833
] | null | null | null | null | null | null | null | null | null | train |
42,070,845 | aard | 2024-11-06T22:54:58 | Jon Stewart told Jeff Bezos at a dinner workers want fulfillment, not errands | null | https://www.businessinsider.com/jon-stewart-jeff-bezos-economic-vision-revolution-obama-dinner-2022-1 | 3 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,866 | handfuloflight | 2024-11-06T22:57:05 | Daily Bots | null | https://bots.daily.co/sign-in | 1 | 0 | null | null | null | missing_parsing | Daily Bots Dashboard | null | null | By signing up, you agree to our Terms of Service and Privacy Policy. | 2024-11-08T17:13:15 | null | train |
42,070,873 | throwaway888abc | 2024-11-06T22:57:44 | Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks | null | https://www.microsoft.com/en-us/research/articles/magentic-one-a-generalist-multi-agent-system-for-solving-complex-tasks/ | 2 | 0 | null | null | null | no_error | Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks - Microsoft Research | null | null |
By Adam Fourney, Principal Researcher; Gagan Bansal, Senior Researcher; Hussein Mozannar, Senior Researcher; Victor Dibia, Principal Research Software Engineer; Saleema Amershi, Partner Research Manager
Contributors: Adam Fourney, Gagan Bansal, Hussein Mozannar, Cheng Tan, Eduardo Salinas, Erkang (Eric) Zhu, Friederike Niedtner, Grace Proebsting, Griffin Bassman, Jack Gerrits, Jacob Alber, Peter Chang, Ricky Loynd, Robert West, Victor Dibia, Ahmed Awadallah, Ece Kamar, Rafah Hosn, Saleema Amershi
We are introducing Magentic-One, our new generalist multi-agent system for solving open-ended web and file-based tasks across a variety of domains. Magentic-One represents a significant step towards developing agents that can complete tasks that people encounter in their work and personal lives. We are also releasing an open-source implementation of Magentic-One (opens in new tab) on Microsoft AutoGen, our popular open-source framework for developing multi-agent applications.
The future of AI is agentic. AI systems are evolving from having conversations to getting things done—this is where we expect much of AI’s value to shine. It’s the difference between generative AI recommending dinner options to agentic assistants that can autonomously place your order and arrange delivery. It’s the shift from summarizing research papers to actively searching for and organizing relevant studies in a comprehensive literature review.
Modern AI agents, capable of perceiving, reasoning, and acting on our behalf, are demonstrating remarkable performance in areas such as software engineering, data analysis, scientific research, and web navigation. Still, to fully realize the long-held vision of agentic systems that can enhance our productivity and transform our lives, we need advances in generalist agentic systems. These systems must reliably complete complex, multi-step tasks across a wide range of scenarios people encounter in their daily lives.
Introducing Magentic-One (opens in new tab), a high-performing generalist agentic system designed to solve such tasks. Magentic-One employs a multi-agent architecture where a lead agent, the Orchestrator, directs four other agents to solve tasks. The Orchestrator plans, tracks progress, and re-plans to recover from errors, while directing specialized agents to perform tasks like operating a web browser, navigating local files, or writing and executing Python code.
Magentic-One achieves statistically competitive performance to the state-of-the-art on multiple challenging agentic benchmarks, without requiring modifications to its core capabilities or architecture. Implemented using AutoGen (opens in new tab), our popular open-source multi-agent framework, Magentic-One benefits from the modular and flexible multi-agent paradigm. This approach offers numerous advantages over monolithic single-agent systems. For example, encapsulating distinct skills in separate agents simplifies development and reuse, much like object-oriented programming. Magentic-One’s plug-and-play design further supports easy adaptation and extensibility by enabling agents to be added or removed without altering other agents or the overall architecture, unlike single-agent systems that often struggle with constrained and inflexible workflows.
We’re making Magentic-One open-source for researchers and developers. While Magentic-One shows strong generalist capabilities, it’s still far from human-level performance and can make mistakes. Moreover, as agentic systems grow more powerful, their risks—like taking undesirable actions or enabling malicious use-cases—can also increase. While we’re still in the early days of modern agentic AI, we’re inviting the community to help tackle these open challenges and ensure our future agentic systems are both helpful and safe. To this end, we’re also releasing AutoGenBench (opens in new tab), an agentic evaluation tool with built-in controls for repetition and isolation to rigorously test agentic benchmarks and tasks while minimizing undesirable side-effects.
How it Works
Magentic-One features an Orchestrator agent that implements two loops: an outer loop and an inner loop. The outer loop (lighter background with solid arrows) manages the task ledger (containing facts, guesses, and plan) and the inner loop (darker background with dotted arrows) manages the progress ledger (containing current progress, task assignment to agents).
Magentic-One work is based on a multi-agent architecture where a lead Orchestrator agent is responsible for high-level planning, directing other agents and tracking task progress. The Orchestrator begins by creating a plan to tackle the task, gathering needed facts and educated guesses in a Task Ledger that is maintained. At each step of its plan, the Orchestrator creates a Progress Ledger where it self-reflects on task progress and checks whether the task is completed. If the task is not yet completed, it assigns one of Magentic-One other agents a subtask to complete. After the assigned agent completes its subtask, the Orchestrator updates the Progress Ledger and continues in this way until the task is complete. If the Orchestrator finds that progress is not being made for enough steps, it can update the Task Ledger and create a new plan. This is illustrated in the figure above; the Orchestrator work is thus divided into an outer loop where it updates the Task Ledger and an inner loop to update the Progress Ledger.
Overall, Magentic-One consists of the following agents:
Orchestrator: the lead agent responsible for task decomposition and planning, directing other agents in executing subtasks, tracking overall progress, and taking corrective actions as needed
WebSurfer: This is an LLM-based agent that is proficient in commanding and managing the state of a Chromium-based web browser. With each incoming request, the WebSurfer performs an action on the browser then reports on the new state of the web page The action space of the WebSurfer includes navigation (e.g. visiting a URL, performing a web search); web page actions (e.g., clicking and typing); and reading actions (e.g., summarizing or answering questions). The WebSurfer relies on the accessibility tree of the browser and on set-of-marks prompting to perform its actions.
FileSurfer: This is an LLM-based agent that commands a markdown-based file preview application to read local files of most types. The FileSurfer can also perform common navigation tasks such as listing the contents of directories and navigating a folder structure.
Coder: This is an LLM-based agent specialized through its system prompt for writing code, analyzing information collected from the other agents, or creating new artifacts.
ComputerTerminal: Finally, ComputerTerminal provides the team with access to a console shell where the Coder’s programs can be executed, and where new programming libraries can be installed.
Together, Magentic-One’s agents provide the Orchestrator with the tools and capabilities that it needs to solve a broad variety of open-ended problems, as well as the ability to autonomously adapt to, and act in, dynamic and ever-changing web and file-system environments.
While the default multimodal LLM we use for all agents is GPT-4o, Magentic-One is model agnostic and can incorporate heterogonous models to support different capabilities or meet different cost requirements when getting tasks done. For example, it can use different LLMs and SLMs and their specialized versions to power different agents. We recommend a strong reasoning model for the Orchestrator agent such as GPT-4o. In a different configuration of Magentic-One, we also experiment with using OpenAI o1-preview for the outer loop of the Orchestrator and for the Coder, while other agents continue to use GPT-4o.
Evaluation
To rigorously evaluate Magentic-One’s performance, we introduce AutoGenBench, an open-source standalone tool for running agentic benchmarks that allows repetition and isolation, e.g., to control for variance of stochastic LLM calls and side-effects of agents taking actions in the world. AutoGenBench facilitates agentic evaluation and allows adding new benchmarks. Using AutoGenBench, we can evaluate Magentic-One on a variety of benchmarks. Our criterion for selecting benchmarks is that they should involve complex multi-step tasks, with at least some steps requiring planning and tool use, including using web browsers to act on real or simulated webpages. We consider three benchmarks in this work that satisfy this criterion: GAIA, AssistantBench, and WebArena.
In the Figure below we show the performance of Magentic-One on the three benchmarks and compare with GPT-4 operating on its own and the per-benchmark highest-performing open-source baseline and non open-source benchmark specific baseline according to the public leaderboards as of October 21, 2024. Magentic-One (GPT-4o, o1) achieves statistically comparable performance to previous SOTA methods on both GAIA and AssistantBench and competitive performance on WebArena. Note that GAIA and AssistantBench have a hidden test set while WebArena does not, and thus WebArena results are self-reported. Together, these results establish Magentic-One as a strong generalist agentic system for completing complex tasks.
Evaluation results of Magentic-One on the GAIA, AssistantBench and WebArena. Error bars indicate 95% confidence intervals. Note that WebArena results are self-reported.
Risks & Mitigations
Agentic systems like Magentic-One represent a phase transition in the opportunities and risks of having AI systems in the world. Magentic-One interacts with a digital world designed for, and inhabited by, humans. It can take actions, change the state of the world and result in consequences that might be irreversible. This carries inherent and undeniable risks and we observed examples of emerging risks during our testing. For example, during development, a misconfiguration prevented agents from successfully logging in to a particular WebArena website. The agents attempted to log in to that website until the repeated attempts caused the account to be temporarily suspended. The agents then attempted to reset the account’s password. More worryingly, in a handful of cases — and until prompted otherwise — the agents occasionally attempted to recruit other humans for help (e.g., by posting to social media, emailing textbook authors, or, in one case, drafting a freedom of information request to a government entity). In each of these cases, the agents failed because they did not have access to the requisite tools or accounts, and/or were stopped by human observers.
In accordance with Microsoft’s commitments to Responsible AI, we worked to identify, measure, and mitigate potential risks of Magentic-One prior to deployment. In particular, we performed red-teaming exercises for potential harmful content, jailbreak, and prompt injection attacks finding no increased risk from our design. In addition, we provide cautionary notices about how to use Magentic-One safely along with guidance, examples, and appropriate defaults for using Magentic-One in ways that minimize risks including how to bring humans-in-the-loop for monitoring and oversight, and ensuring all examples involving code execution, and our evaluation and benchmarking tools run in sandboxed docker containers.
We recommend using Magentic-One with models with strong alignment and pre- and post-generation filtering, and closely monitoring logs during and after execution. Indeed, in our own use, we follow the strict principle of least privilege, and maximum oversight. We acknowledge that minimizing the potential risks from agentic AI will require new techniques and much research is needed in both understanding these emerging risks and developing techniques. We will continue sharing our learnings with the community and continue evolving Magentic-One with the latest on safety research.
We see valuable new directions in agentic, safety and Responsible AI research: In terms of anticipating new risks from agentic systems, it is possible that agents will be subject to the same phishing, social engineering, and misinformation attacks that target human web surfers when they are acting on the public web. On cross-cutting mitigations, we anticipate an important direction will be equipping agents with an understanding of which actions are easily reversible, which are reversible with some effort, and which cannot be undone. As an example, deleting files, sending emails, and filing forms, are unlikely to be easily reversed. When faced with a high-cost or irreversible action, systems should be designed to pause, and to seek human input.
To conclude, in this work we introduced Magentic-One, a generalist multi-agent system represents a significant development in agentic systems capable of solving open-ended tasks.
For further information, results and discussion, please see our technical report. (opens in new tab)
| 2024-11-08T09:36:15 | en | train |
42,070,875 | aard | 2024-11-06T22:57:58 | Natural Language Outlines for Code: Literate Programming in the LLM Era | null | https://arxiv.org/abs/2408.04820 | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,882 | Patfernand | 2024-11-06T22:58:26 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,070,897 | wumeow | 2024-11-06T22:59:30 | US Bond Yields Surge as Trump Win Stokes Inflation Expectations | null | https://www.bloomberg.com/news/articles/2024-11-06/us-bonds-slide-most-since-pandemic-as-trump-renews-inflation-bet | 5 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,899 | hershyb_ | 2024-11-06T22:59:39 | Ask HN: Hosting on Digital Ocean, any advice for monitoring and deployments? | I'm moving over from Lambda to Digital Ocean since I found it much easier to test locally and iterate. Unfortunately, I'm not too savvy with setting up monitoring so I can see how often failures are happening and how to smoothly deploy changes on DO. I currently have a setup where I pull each individual change into my instance and re-run docker-compose build etc. Would love to learn more about what tools have helped you for logging/monitoring/deployments | null | 4 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,925 | robaato | 2024-11-06T23:02:27 | Sysadmin shock as Windows Server 2025 installs itself after labeling error | null | https://www.theregister.com/2024/11/06/windows_server_2025_surprise/ | 11 | 1 | [
42071463
] | null | null | no_error | Sysadmin shock as Windows Server 2025 installs itself after update labeling error | 2024-11-06T14:36:08Z | Richard Speed |
Administrators are reporting unexpected appearances of Windows Server 2025 after what was published as a security update turned out to be a complete operating system upgrade.
The problem was flagged by a customer of web app security biz Heimdal. Arriving at the office on the morning of November 5, they found, to their horror, that every Windows Server 2022 system had either upgraded itself to Windows Server 2025 or was about to.
Sysadmins are cautious by nature, so an unplanned operating system upgrade could easily result in morning coffee being sprayed over a keyboard.
Heimdal's services include patch management, and it relies on Microsoft to label patches accurately to ensure the correct update is applied to the correct software at the correct time. In this instance, what should have been a security update turned out to be Windows Server 2025.
It took Heimdal a while to trace the problem. According to a post on Reddit: "Due to the limited initial footprint, identifying the root cause took some time. By 18:05 UTC, we traced the issue to the Windows Update API, where Microsoft had mistakenly labeled the Windows Server 2025 upgrade as KB5044284."
Buckle up, admins – Windows Server 2025 officially hits GA
Productivity suites, Exchange servers in path of Microsoft's end-of-support wave
Saying goodbye to the tech dreams Microsoft abandoned with Windows 11 24H2
Admins using Windows Server Update Services up in arms as Microsoft deprecates feature
It added: "Our team discovered this discrepancy in our patching repository, as the GUID for the Windows Server 2025 upgrade does not match the usual entries for KB5044284 associated with Windows 11. This appears to be an error on Microsoft's side, affecting both the speed of release and the classification of the update. After cross-checking with Microsoft's KB repository, we confirmed that the KB number indeed references Windows 11, not Windows Server 2025."
The Register has contacted Heimdal for more information and will update this piece should the security organization respond. We also asked Microsoft to comment almost 24 hours ago. Since then? Crickets.
As of last night, Heimdal estimated that the unexpected upgrade had affected 7 percent of customers – it said it had blocked KB5044284 across all server group policies. However, this is of little comfort to administrators finding themselves receiving an unexpected upgrade.
Since rolling back to the previous configuration will present a challenge, affected users will be faced with finding out just how effective their backup strategy is or paying for the required license and dealing with all the changes that come with Windows Server 2025. ®
| 2024-11-08T02:42:26 | en | train |
42,070,932 | B1FF_PSUVM | 2024-11-06T23:02:58 | Bathos: The Art of Sinking | null | https://en.wikipedia.org/wiki/Bathos | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,934 | teleforce | 2024-11-06T23:03:15 | Panasonic Connect Certifies Toughbook Devices on Red Hat Enterprise Linux | null | https://na.panasonic.com/news/panasonic-connect-certifies-toughbook-devices-on-red-hat-enterprise-linux-for-enhanced-flexibility-and-security | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,940 | aard | 2024-11-06T23:03:36 | One Company A/B Tested Hybrid Work. Here's What They Found | null | https://hbr.org/2024/10/one-company-a-b-tested-hybrid-work-heres-what-they-found | 1 | 1 | [
42071098
] | null | null | no_error | One Company A/B Tested Hybrid Work. Here’s What They Found. | 2024-10-29T12:05:08Z | by
Nicholas Bloom,
James Liang,
and
Ruobing Han |
The six-month experiment reversed managers’ opinions on hybrid work — and offers three important lessons.
October 29, 2024
Westend61/Getty Images
Post
Post
Share
Annotate
Save
Get PDF
Buy Copies
Print
Since the pandemic, executives have had to rethink their work-from-home policies to better support their companies’ bottom line. Recent research conducted in a real company showed that employees who worked from home three days a week experienced higher satisfaction and lower attrition rates compared with their colleagues who worked from the office. This reduction in turnover saved millions of dollars in recruiting and training costs, thereby increasing profits for the company. Business leaders can learn valuable lessons from this study to implement a successful hybrid work model: establishing rigorous performance management systems, coordinating team or company-level hybrid schedules, and securing support from firm leadership. Additionally, executives should A/B test their own management practices to find what works best for them.
Post
Post
Share
Annotate
Save
Get PDF
Buy Copies
Print
Amazon’s recent call for employees to return to the office (RTO) five days a week is the latest example of high-profile companies pulling back from their remote-work policies. RTO advocates often cite the importance of in-person connections, with former Google CEO Eric Schmidt even claiming, “Google decided that work-life balance and going home early and working from home was more important than winning.”
Nicholas Bloom is a professor of economics at Stanford University.
James Liang is a leading scholar of demographic economics, entrepreneurship, and innovation research. He co-founded Trip.com Group and currently serves as its Executive Chairman. He is also Research Professor of Applied Economics at Peking University’s Guanghua School of Management.
Ruobing Han is an assistant professor at the Chinese University of Hong Kong, Shenzhen and a recent graduate of Stanford’s Ph.D. program in Economics, specializingin Quantitative Marketing. His intellectual curiosity extends to Industrial Organization and Applied Microeconomics and the dynamics of the Chinese Economy. In addition to his academic pursuits, Ruobing is an avid reader and enjoys playing basketball in his spare time.
Post
Post
Share
Annotate
Save
Get PDF
Buy Copies
Print
| 2024-11-08T10:52:53 | en | train |
42,070,942 | bizlinkers | 2024-11-06T23:04:15 | null | null | null | 1 | null | [
42070943
] | null | true | null | null | null | null | null | null | null | train |
42,070,946 | empressplay | 2024-11-06T23:04:39 | Trudeau government bans TikTok from operating in Canada | null | https://www.cbc.ca/news/politics/tiktok-canada-review-1.7375965 | 108 | 43 | [
42071299,
42071124,
42071295,
42071382,
42071085,
42071389,
42071106,
42071282,
42071407,
42071068,
42071108,
42071314,
42071017,
42071359,
42070966
] | null | null | null | null | null | null | null | null | null | train |
42,070,969 | mecarc | 2024-11-06T23:06:25 | Show HN: Galliass v2 | Galliass is now a no-code link engagement tool. QR code generator, Biolink, short links, and more. | https://galliass.com/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,975 | teleforce | 2024-11-06T23:06:47 | Pentagon taps Hughes to develop 5G O-RAN prototype at Fort Bliss | null | https://defensescoop.com/2024/11/04/hughes-5g-oran-prototype-fort-bliss/ | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,976 | geox | 2024-11-06T23:06:53 | Comb jellies are capable of rapidly fusing into a single entity | null | https://www.npr.org/2024/10/08/nx-s1-5144689/comb-jelly-superpower-absorb-become-one | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,070,989 | hettygreen | 2024-11-06T23:08:01 | Man Charged with Attempting to Use a WMD to Destroy Energy Facility in Nashville | null | https://www.justice.gov/opa/pr/man-arrested-and-charged-attempting-use-weapon-mass-destruction-and-destroy-energy-facility | 2 | 2 | [
42071311,
42071133
] | null | null | null | null | null | null | null | null | null | train |
42,070,995 | donadev | 2024-11-06T23:08:30 | Show HN: QABuddy – Test your mobile app flows with natural language and AI | null | https://qabuddy.io/ | 2 | 1 | [
42070998
] | null | null | null | null | null | null | null | null | null | train |
42,071,072 | jsmith21 | 2024-11-06T23:18:16 | Show HN: A online music hosting platform using Palantir AIP as the back end [video] | null | https://www.youtube.com/watch?v=YKkURPjPHRw | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,083 | che_shr_cat | 2024-11-06T23:19:07 | Make Softmax Great Again | null | https://gonzoml.substack.com/p/make-softmax-great-again | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,114 | bikenaga | 2024-11-06T23:21:14 | For fame or a death wish? Kids' TikTok challenge injuries stump psychiatrists | null | https://arstechnica.com/health/2024/11/for-fame-or-a-death-wish-kids-tiktok-challenge-injuries-stump-psychiatrists/ | 2 | 2 | [
42071184,
42071307
] | null | null | null | null | null | null | null | null | null | train |
42,071,147 | quinto_quarto | 2024-11-06T23:24:05 | Show HN: A minimal iPhone widget for people who love sharing links | null | https://eyeball.wtf/ | 1 | 0 | null | null | null | missing_parsing | Eyeball, an iPhone widget | null | null | is a delightful iphone widget for people who love sharing links. made by kabir in new york.you and your amigos get the widget.you discover a fascinating thing on the world wide web. eyeball sends it straight to their screens.privacy policyterms and conditionscontact | 2024-11-08T07:11:15 | null | train |
42,071,148 | teleforce | 2024-11-06T23:24:13 | End-to-End Open-Source 5G and O-Ran Prototyping in Agriculture Application [pdf] | null | https://openairinterface.org/wp-content/uploads/2024/09/OAI-10th-Anniversary-Workshop-Iowa-State-University.pdf | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,189 | adamzerner | 2024-11-06T23:28:22 | Working With Monsters | null | https://www.lesswrong.com/posts/o4cgvYmNZnfS4xhxL/working-with-monsters | 2 | 0 | null | null | null | no_error | Working With Monsters | 2021-07-20T15:23:20.762Z | johnswentworth | Working With MonstersThis is a fictional piece based on Sort By Controversial. You do not need to read that first, though it may make Scissor Statements feel more real. Content Warning: semipolitical. Views expressed by characters in this piece are not necessarily the views of the author.I stared out at a parking lot, the pavement cracked and growing grass. A few cars could still be seen, every one with a shattered windshield or no tires or bashed-in roof, one even laying on its side. Of the buildings in sight, two had clearly burned, only blackened reinforced concrete skeletons left behind. To the left, an overpass had collapsed. To the right, the road was cut by a hole four meters across. Everywhere, trees and vines climbed the remains of the small city. The collapsed ceilings and shattered windows and nests of small animals in the once-hospital behind me seemed remarkably minor damage, relatively speaking.Eighty years of cryonic freeze, and I woke to a post-apocalyptic dystopia.“It’s all like that,” said a voice behind me. One of my… rescuers? Awakeners. He went by Red. “Whole world’s like that.”“What happened?” I asked. “Bioweapon?”“Scissor,” replied a woman, walking through the empty doorway behind Red. Judge, he’d called her earlier.I raised an eyebrow, and waited for elaboration. Apparently they expected a long conversation - both took a few seconds to get comfortable, Red leaning up against the wall in a patch of shade, Judge righting an overturned bench to sit on. It was Red who took up the conversation thread.“Let’s start with an ethical question,” he began, then laid out a simple scenario. “So,” he asked once finished, “blue or green?”.“Blue,” I replied. “Obviously. Is this one of those things where you try to draw an analogy from this nice obvious case to a more complicated one where it isn’t so obvious?”“No,” Judge cut in, “It’s just that question. But you need some more background.”“There was a writer in your time who coined the term ‘scissor statement’,” Red explained, “It’s a statement optimized to be as controversial as possible, to generate maximum conflict. To get a really powerful scissor, you need AI, but the media environment of your time was already selecting for controversy in order to draw clicks.”“Oh no,” I said, “I read about that… and the question you asked, green or blue, it seems completely obvious, like anyone who’d say green would have to be trolling or delusional or a threat to society or something… but that’s exactly how scissor statements work…”“Exactly,” replied Judge. “The answer seems completely obvious to everyone, yet people disagree about which answer is obviously-correct. And someone with the opposite answer seems like a monster, a threat to the world, like a serial killer or a child torturer or a war criminal. They need to be put down for the good of society.”I hesitated. I knew I shouldn’t ask, but… “So, you two…”Judge casually shifted position, placing a hand on some kind of weapon on her belt. I glanced at Red, and only then noticed that his body was slightly tensed, as if ready to run. Or fight.“I’m a blue, same as you,” said Judge. Then she pointed to Red. “He’s a green.”I felt a wave of disbelief, then disgust, then fury. It was so wrong, how could anyone even consider green... I took a step toward him, intent on punching his empty face even if I got shot in the process.“Stop,” said Judge, “unless you want to get tazed.” She was holding her weapon aimed at me, now. Red hadn’t moved. If he had, I’d probably have charged him. But Judge wasn’t the monster here… wait.I turned to Judge, and felt a different sort of anger.“How can you just stand there?”, I asked. “You know that he’s in the wrong, that he’s a monster, that he deserves to be put down, preferably slowly and painfully!” I was yelling at Judge, now, pointing at Red with one hand and gesticulating with the other. “How can you work with him!?”Judge held my eyes for a moment, unruffled, before replying. “Take a deep breath,” she finally said, “calm yourself down, take a seat, and I’ll explain.”I looked down, eyed the tazer for a moment, closed my eyes, then did as she asked. Breathe in, breathe out. After a few slow breaths, I glanced around, then chose a fallen tree for a seat - positioning Judge between Red and myself. Judge raised an eyebrow, I nodded, and she resumed her explanation.“You can guess, now, how it went down. There were warning shots, controversies which were bad but not bad enough to destroy the world. But then the green/blue question came along, the same question you just heard. It was almost perfectly split, 50/50, cutting across political and geographical and cultural lines. Brothers and sisters came to blows. Bosses fired employees, and employees sued. Everyone thought they were in the right, that the other side was blatantly lying, that the other side deserved punishment while their side deserved an apology for the other side’s punishments. That they had to stand for what was right, bravely fight injustice, that it would be wrong to back down.”I could imagine it. What I felt, toward Red - it felt wrong to overlook that, to back down. To let injustice pass unanswered.“It just kept escalating, until bodies started to pile up, and soon ninety-five percent of the world population was dead. Most people didn’t even try to hole up and ride out the storm - they wanted to fight for what was right, to bring justice, to keep the light in the world.”Judge shrugged, then continued. “There are still pockets here and there, where one side or the other gained the upper hand and built a stronghold. Those groups still fight each other. But most of what’s left is ruins, and people like us who pick over them.”“So why aren’t you fighting?” I asked. “How can you overlook it?”Judge sighed. “I was a lawyer, before Scissor.” She jerked her head toward Red. “He was too. We even came across each other, from time to time. We were both criminal defense attorneys, with similar clients in some ways, though very different motivations.“Red was… not exactly a bleeding heart, but definitely a man of principles. He’d made a lot of money early on, and mostly did pro-bono work. He defended the people nobody else would take. Child abusers, serial killers, monsters who everyone knew were guilty. Even Red thought they were guilty, and deserved life in prison, maybe even a death sentence. But he was one of those people who believed that even the worst criminals had to have a proper trial and a strong defense, because it was the only way our system could work. So he defended the monsters. Man of principle.“As for me, I was a mob lawyer. I defended gangsters, loan sharks, arms dealers… and their friends and families. It was the families who were the worst - the brothers and sons who sought sadistic thrills, knowing they’d be protected. But it was interesting work, the challenge of defending the undefendable, and it paid a fortune.“We hated each other, back in the day. Still do, on some level. He was the martyr, the white knight putting on airs of morality while defending monsters. And I was the straightforward villain, fighting for money and kicks. But when Scissor came, we had one thing in common: we were both willing to work with monsters. And that turned out to be the only thing which mattered.”I nodded. “So you hated each other, but you’d both spent years working with people you hated, so working with each other was… viable. You even had a basis to trust one another, in some weird way, because you each knew that the other could work with people they hated.”“Exactly. In the post-scissor world, people who can work with monsters are basically the only people left. We form mixed groups - Red negotiates with Greens for us, I negotiate with Blues. They can tell, when they ask whether you’re Blue or Green - few people can lie convincingly, with that much emotion wrapped up in it. A single-color group would eventually encounter the opposite single-color group, and they’d kill each other. So when we meet other groups, they have some Blues and some Greens, and we don’t fight about it. We talk, we trade, we go our separate ways. We let the injustice sit, work with the monsters, because that’s the only way to survive in this world.“And now you have to make a choice. You can go out in a blaze of glory, fight for what you know is right, and maybe take down a few moral monsters in the process. Or you can choose to live and let live, to let injustice go unanswered, to work with the monsters you hate. It’s up to you.”244Surely the story would be more believable if the POV character endorsed green rather than blue. As it is, I don't find them a very sympathetic character, and I can't imagine a reasonable audience would either.Nice meta-comment. But it doesn't really work; green was very well chosen so that any right person with a modicum of brains and heart immediately detects it as both wrong and morally repugnant. To such an extent that I found it broke my suspension of disbelief that half of the future society would believe in green.[-][anonymous]3y70Which is a meta comment on present day I think, where the blue red divide is such that one of those sides clearly aligns with facts and physical reality while another side relies on made up stories. Except of course present day, the reason for one of these sides seems almost to be a self identity thing, where they don't really believe in their color's precepts, they just identify with the people in it more so that's the color they fly."The reason for one of these sides seems almost to be a self identity thing, where they don't really believe in their color's precepts, they just identify with the people in it"
Based on that, I know exactly the bastards you're talking about, and I don't believe anyone would be able to tolerate them as compatriots if they weren't totally dark triad, at least to some degree. So we're agreed we need to stand up for what's right and shut them all down before something serious happens?
Interestingly, if one looks at this story in terms of "what message is this story sending", then it feels like the explicit and implicit message are the opposites of each other.The explicit message seems to be something like "cooperation with the other side is good, it can be the only way to survive".But then if we think of this representing a "pro-cooperation side", we might notice that the story doesn't really give any real voice to the "anti-cooperation side" - the one which would point out that actually, there are quite a few situations when you absolutely shouldn't cooperate with monsters. The setup of the story is such that it can present a view from which the pro-cooperation side is simply correct, as opposed to looking at a situation where it's more questionable.In the context of a fictional story making a point about the real world, I would interpret "cooperating with the other side" to mean something like "making an honest attempt to fairly present the case for the opposite position". Since this story doesn't do that, it reads to me like it's saying that we should cooperate with those who disagree with us... while at the same time not cooperating with the side that it disagrees with. I would word the intended message as "whether or not someone shares our values is not directly relevant to whether one should cooperate with them". Moral alignment is not directly relevant to the decision; it enters only indirectly, in reasoning about things like the need for enforcement or reputational costs. Monstrous morals should not be an immediate deal-breaker in their own right; they should weigh on the scales via trust and reputation costs, but that weight is not infinite.I don't really think of it as "pro-cooperation" or "anti-cooperation"; there is no "pro-cooperation" "side" which I'm trying to advocate here.Makes one wonder what kind of story could justify the opposite moral. I do think that moral would be "All that evil needs to win is for the good people to do nothing"A story could also...give us the actual statement. I say 'If scissor statements are real, name 3.'It’s peace vs conflict. Peace is cooperation and conflict is anti-cooperation. Is peace always the right answer? Maybe not, but it’s the one I’m going to pick most of the time.I think it makes a pretty good case for the anti-cooperation side: you might get to kill some of your enemies before you get killed in turn. However, the correctness of any argument can only be judged by those who remain alive.the correctness of any argument can only be judged by those who remain alive.2+2=4It can be judged by those who are alive, those who were alive, those who will be alive...need I say more?If you think dead people can do arithmetic, I think you need to explain how that would work.While they are dead no. While they were alive - yes, they could. (This is interesting in that, properly performed, (a specified) computation gets the same result, whatever the circumstances. More generally, Fermat argued that: for integers a, b, c, and n, where n>2, a^n+b^n=c^N:had no solutionswas provableHe might have been wrong about the difficulty of proving it, but he was right about the above. If perhaps for the wrong reasons. (Can we prove Fermat didn't have a proof?))A while ago, someone encouraged me read Homage to Catalonia, citing it as a book that’d dissuade people from revolutionary justice. And in particular, dissuade people from the notion that they should carefully guard who they work with, like a blue working with a green.In fact, I found the book had the opposite effect. It describes a what amounts to a three-way war between anarchists, communists, and fascists during the Spanish Civil War. During that war, foreign communists and capitalists both benefited from continuing top-down company-owned business models in certain countries, and so strongly dissuaded a Spanish worker’s revolution, an agenda which Spanish stalinists cooperated with to continue receiving funding. The anarchists wanted that revolution, but were willing to team up with the stalinist bloc against the fascists, it seems, because they couldn’t fight both, and they saw the fascists as a greater threat. The stalinists (who did not want revolution) took advantage of the anarchists comparatively worse position to neuter them, rolling back worker-controlled factories and local-run governments, which were a threat to foreign interests.The stalinist block would frame “winning the war” as a means to get the anarchists to surrender on all their hard won progress, saying, “well, we can fight over worker owned factories, or we can fight together against the fascists,” essentially holding the country hostage, using the fascists as a threat to get what they wanted. And in the end, they both lost to Franco. This example seems to be a primary reason for not working with people who aren’t value-aligned: they’ll undermine your position, using the excuse of “unity against the enemy.” Once you give ground on local worker-led patrols instead of police, the non-value-aligned group will start pressing for a return to centralized government, imperially-owned factories, and worker exploitation. Give them an inch, they take a mile. Moloch says, "throw what you love into the fire and I will grant you victory," but any such bargain is made under false pretenses. In making the deal, you've already lost. My model is that a blue and green working together would constantly undermine the other's cause, and when that cause is life and death, this is tantamount to working with someone towards your own end. Some things matter enough that you shouldn't capitulate, where capitulation is the admission that you don't really hold the values you claim to hold -- it would be like saying you believe in gravity, while stepping off a cliff with the expectation that you'll float.Good example. I'll use this to talk about what I think is the right way to think about this.First things first: true zero-sum games are ridiculously rare in the real world. There's always some way to achieve mutual gains - even if it's just "avoid mutual losses" (as in e.g. mutual assured destruction). Of course, that does not mean that an enemy can be trusted to keep a deal. As with any deal, it's not a good deal if we don't expect the enemy to keep it.The mutual gains do have to be real in order for "working with monsters" to make sense.That said... I think people tend to have a gut-level desire to not work with monsters. This cashes out as motivated stopping: someone thinks "ah, but I can't really trust the enemy to uphold their end of the deal, can I?"... and they use that as an excuse to not make any deal at all, without actually considering (a) whether there is actually any evidence that the enemy is likely to break the deal (e.g. track record), (b) whether it would actually be in the enemy's interest to break the deal, or (c) whether the deal can be structured so that the enemy has no incentive to break it. People just sort of horns-effect, and assume the Bad Person will of course break a deal because that would be Bad.(There's a similar thing with reputational effects, which I expect someone will also bring up at some point. Reputational effects are real and need to be taken into consideration when thinking about whether a deal is actually net-positive-expected-value. But I think people tend to say "ah, but dealing with this person will ruin my reputation"... then use that as an excuse to not make a deal, without considering (a) how highly-visible/salient this deal actually is to others, (b) how much reputational damage is actually likely, or (c) whether the deal can plausibly be kept secret.)true zero-sum games are ridiculously rare in the real world. There's always some way to achieve mutual gains - even if it's just "avoid mutual losses"I disagree.I think you're underestimating how deep value differences can be, and how those values play into everything a person does. Countries with nuclear weapons who have opposing interests are actively trying to destroy each other without destroying themselves in the process, and if you're curious about the failures of MAD, I'd suggest reading The Doomsday Machine, by Daniel Ellsberg. If that book is to be taken as mostly true, and the MWI is to be taken as true, then I suspect that many, many worlds were destroyed by nuclear missiles. When I found this unintuitive, I spent a day thinking about quantum suicide to build that intuition: most instances of all of us are dead because we relied on MAD. We're having this experience now where I'm writing this comment and you're reading it because everything that can happen will happen in some branch of the multiverse, meaning our existance is only weak evidence for the efficacy of MAD, and all of those very close calls are stronger evidence for our destruction in other branches. This doesn't mean we're in the magic branch where MAD works, it means we've gotten lucky so far. Our futures are infinite split branches of parallel mes and yous, and in most of those where we rely on strategies like MAD, we die....Scissor statements reveal pre-existing differences in values, they don't create them. There really are people out there who have values that result in them doing terrible things. Furthermore, beliefs and values aren't just clothes we wear -- we act on them, and live by them. So it's reasonable to assume that if someone has a particularly heinous belief, and particularly heinous values, that they act on those beliefs and values.In the ssc short story, scissor statements are used to tear apart mozambique, and in real life, we see propagandists using scissor statements to split up activist coalitions. It's not hypothetical, divide and conquer is a useful strategy that has been used probably since the dawn of time. But not all divides are created equal.In the 1300s in rural France, peasants revolted against the enclosure of the commons, and since many of these revolts were led by women, the nascent state officials focused their efforts on driving a (false) wedge between men and women, accusing those women of being witches & followers of satan. Scissor statements (from what I can tell) are similar in that they're a tactic used to split up a coalition, but different in that they're not inventing conflict. It doesn't seem to make much of a difference in terms of outcome (conflict) once people have sorted themselves into opposing groups, but equating the two is a mistake. You're losing something real if you ally yourself with someone you're not value-aligned with, and you're not losing something real if you're allying yourself with someone you are value-aligned with, but mistakenly think is your enemy. The amount of power people like you with your value has loses strength because now another group that wants to destroy you has more power.If two groups form a coalition, and gorup_A values "biscuits for all," and group_B values "cookies for all," and someone tries to start a fight between them based on this language difference, it would be tragic for them to fight. Because it should be obvious that what they want is the same thing, they're just using different language to talk about it. And if they team up, group_A won't be tempted to deny group_B cookies, because they deep-down value cookies for all, including group_B. It's baked into their decision making process.(And if they decide that what they want to spend all their time doing is argue over whether they should call their baked food product "cookies" or "biscuits," then what they actually value is arguing about pedantry, not "cookies for all.")But in a counter example, if group_A values "biscuits for all" and group_B values "all biscuits for group_B," then group_B will find it very available and easy to think of strategies which result in biscuits for group_B and not group_A. If someone is having trouble imagining this, that may be because it's difficult to imagine someone only wanting the cookies for themselves, so they assume the other group wouldn't defect, because "cookies for all? What's so controversial about that?" Except group_B fundamentally doesn't want group_A getting their biscuits, so any attempt at cooperation is going to be a mess, because group_A has to keep double-checking to make sure group_B is really cooperating, because it's just so intuitive to group_B not to that they'll have trouble avoiding it. And so giving group_B power is like giving someone power when you know they're later going to use it to hurt you and take your biscuits.And group_B will, because they value group_B having all the biscuits, and have a hard time imagining that anyone would actually want everyone to have all the biscuits, unless they're lying or virtue signalling or something. And they'll push and push because it'll seem like you're just faking....I find the way people respond to scissor statements ("don't bring that up, it's a scissor statement/divisive!") benefits only the status quo. And if the status quo benefits some group of people, then of course that group is going to eschew divisiveness. ...To bring it back to the Spanish Civil War, the communists were willing to ally themselves with big businesses, businesses who were also funding the fascists. They may have told themselves it was a means to an end, and for all I know (because my knowledge of the Spanish Civil War is limited only to a couple books,) the communists may have been planning to betray those big business interests, in the end. But in the mean time, they advanced the causes of those big business interests, and undermined the people who stood against everything the fascists fought for. It's difficult to say what would've happened if the anarchists had tried a gambit to force the hand of big business to pick a side (communist or fascist) or simply ignored the communists' demands. But big business interests were more supportive of Franco winning (because he was good for business), and their demands of the communists in exchange for money weakened the communists' position, and because the communists twisted the arms of the anarchists & the anarchists went along with it, this weakened their position, too. And in the end, the only groups that benefitted from that sacrifice were big business interests and Franco's fascists....whether the deal can plausibly be kept secret.That's a crapshoot, especially in the modern day. Creating situations where groups need to keep secrets in order to function is the kind of strategy Julian Assange used to cripple government efficiency. The correct tactic is to keep as few secrets from your allies as you can, because if you're actually allies, then you'll benefit from the shared information. The effectiveness or ineffectiveness of MAD as a strategy is not actually relevant to whether nuclear war is or is not a zero-sum game. That's purely a question of payoffs and preferences, not strategy.You're losing something real if you ally yourself with someone you're not value-aligned with, and you're not losing something real if you're allying yourself with someone you are value-aligned with, but mistakenly think is your enemy. The amount of power people like you with your value has loses strength because now another group that wants to destroy you has more power.The last sentence of this paragraph highlights the assumption: you are assuming, without argument, that the game is zero-sum. That gains in power for another group that wants to destroy you is necessarily worse for you.This assumption fails most dramatically in the case of three or more players. For instance, in your example of the Spanish civil war, it's entirely plausible that the anarchist-communist alliance was the anarchists' best bet - i.e. they honestly preferred the communists over the fascists, the fascists wanted to destroy them even more than the communists, and an attempt at kingmaking was only choice the anarchists actually had the power to make. In that world, fighting everyone would have seen them lose without any chance of gains at all.In general, the key feature of a two-player zero-sum game is that anything which is better for your opponent is necessarily worse for you, so there is no incentive to cooperate. But this cannot ever hold between all three players in a three-way game: if "better for player 1" implies both "worse for player 2" and "worse for player 3", then player 2 and player 3 are incentivized to cooperate against player 1. Three player games always incentivize cooperation between at least some players (except in the trivial case where there's no interaction at all between some of the players). Likewise in games with more than three players. Two-player games are a weird special case.That all remains true even if all three+ players hate each other and want to destroy each other.But in a counter example, if group_A values "biscuits for all" and group_B values "all biscuits for group_B," then group_B will find it very available and easy to think of strategies which result in biscuits for group_B and not group_A. If someone is having trouble imagining this, that may be because it's difficult to imagine someone only wanting the cookies for themselves, so they assume the other group wouldn't defect, because "cookies for all? What's so controversial about that?" Except group_B fundamentally doesn't want group_A getting their biscuits, so any attempt at cooperation is going to be a mess, because group_A has to keep double-checking to make sure group_B is really cooperating, because it's just so intuitive to group_B not to that they'll have trouble avoiding it. And so giving group_B power is like giving someone power when you know they're later going to use it to hurt you and take your biscuits.Note that, in this example, you aren't even trying to argue that there's no potential for mutual gains. Your actual argument is not that the game is zero-sum, but rather that there is overhead to enforcing a deal.It's important to flag this, because it's exactly the sort of reasoning which is prone to motivated stopping. Overhead and lack of trust are exactly the problems which can be circumvented by clever mechanism design or clever strategies, but the mechanisms/strategies are often nonobvious.That gains in power for another group that wants to destroy you is necessarily worse for you.Yes. In many real-life scenarios, this is true. In small games where the rules are blatant, it's easier to tell if someone is breaking an agreement or trying to subvert you, so model games aren't necessarily indicative of real-world conditions. For a real life example, look at the US's decision to fund religious groups to fight communists in the middle east. If someone wants to destroy you, during the alliance they'll work secretly to subvert you, and after the alliance is over, they'll use whatever new powers they have gained to try to destroy you.People make compromises that sacrifice things intrinsic to their stated beliefs when they believe it is inevitable they'll lose — by making the "best bet" they were revealing that they weren't trying to win, that they've utterly given up on winning. The point of anarchy is that there is no king. For an anarchist to be a kingmaker is for an anarchist to give up on anarchy.And from a moral standpoint, what about the situation where someone is asked to work with a rapist, pedophile, or serial killer? We're talking about heinous beliefs/actions here, things that would make someone a monster, not mundane "this person uses ruby and I use python," disagreements. What if working with a {rapist,pedo,serial killer} means they live to injure and kill another day? If that's the outcome, by working with them you're enabling that outcome by enabling them.The last sentence of this paragraph highlights the assumption: you are assuming, without argument, that the game is zero-sum.On the contrary, it highlights no such thing.**You may argue that this is the case with regard to that 'assumption' - but you have not proved it.This need not be the case, for the argument to be correct:That gains in power for another group that wants to destroy you is necessarily worse for you.yes - and this is so even if the game isn't zero sum. This is a nice story, and nicely captures the internal dissonance I feel about cooperating with people who disagree with me about my "pet issue", though like many good stories it's a little simpler and more extreme than what I actually feel.I like it, but 95% seems surprisingly high. Surely there are plenty of other people out there with a similar psychological makeup to Red and Judge, or to the protagonist (who can at least be convinced, with a sufficient threat, to listen before punching.) But I shouldn't fight the hypothetical too much...If you also consider the indirect deaths due to the collapse of civilization, I would say that 95% lies within the realm of reason. You don’t need anywhere close to 95% of the population to be fully affected by the scissor to bring about 95% destruction.Without the statement, it seems unlikely there'd be only 2 answers. So, why not fight the hypothetical? If someone asks 'in a world where 2+2=3, how does math work' I have no answer, without a map of this strange world.I imagine that the protagonist can be more easily convinced because of the state of the new world; the scissor statement may not as much of an issue in the post-apocalyptic world where there are more important things.I feel like this story describes very well the compromises that certain religious individuals make, or don't make, regarding abortion. I liked this story enough to still remember it, separately from the original Sort By Controversial story. Trade across moral divide is a useful concept to have handles for.The claim that scissor statements are dangerous is itself a scissor statement: I think it's obviously false, and will fight you over it. Social interaction is not that brittle. It is important to notice the key ruptures between people's values/beliefs. Disagreements do matter, in ways that sometimes rightly prevent cooperation.
World population is ~2^33, so 33 independent scissor statements would set you frothing in total war of everyone against everyone. Except people are able to fluidly navigate much, much higher levels of difference and complexity than that. Every topic and subculture has fractal disagreements, each battle fiercely fought, and we're basically fine. Is it productive to automatically collaborate on a project with someone who disagrees with your fundamental premises? How should astronomy and astrology best coexist, especially when one of the two is badly out-numbered?
Vigorous, open-ended epistemic and moral competition is hard. Neutrality and collaboration can be useful, but are always context-sensitive and provisional. They are ongoing negotiations, weighing all the different consequences and strategies. A fighting couple can't skip past all the messy heated asymmetric conflicts with some rigid absolutes about civil discourse.
Vigorous, open-ended epistemic and moral competition is hard. Neutrality and collaboration can be useful, but are always context-sensitive and provisional. They are ongoing negotiations, weighing all the different consequences and strategies. A fighting couple can't skip past all the messy heated asymmetric conflicts with some rigid absolutes about civil discourse.I agree with this. The intended message is not that cooperation is always the right choice, but that monstrous morals alone should not be enough to rule out cooperation. Fighting is still sometimes the best choice.What are your thoughts on Three Worlds Collide?This is a beautifully written story. One criticism is that it seems to have a Moral that assumes the Blue cryonicist acts as he does in the story... an entertaining story, but limits the applicability of the Moral. In particular, signing up for cryonics and going out in a blaze of glory are really quite opposite personality traits if you think about it.Just shows how effective the disagreement is at getting people to care deeply about it I guessSocially oblivious question: Is this part of that genre of "posts that are secretly about a specific person or employer"? (aka "subtweeting")Replace blue and green with protestant and catholic, 95% with 60% and what you get is the Thirty Years' War and the beginning of the modern world order.Up to 60% "in some areas of Germany", Wikipedia says. Considering Europe as a whole, it says 8 million out of about 75 million. But yes, the Thirty Years War ended when the parties finally got it through their heads that neither side would ever win. That beginning of the modern world order was the great agreement to disagree that was the Peace of Westphalia, and even that involved conferences at two different places because the parties couldn't bear to all meet each other together.
Thirty years. One generation. Maybe no-one ever changed their minds, it's just that the ones who grew up with it and then came into power realised that none of it had ever mattered.
Should a country where cryonic preservation is routine try to take over one where it is forbidden?
[-][anonymous]3y40Should a country where cryonic preservation is routine try to take over one where it is forbidden?Or a country where anti-aging medicine delivered in international aid is being stolen and wasted to prevent out-groups from receiving treatment?It's a moderately interesting question though only because our current moral frameworks privilege "do nothing and let something bad happen" over "do something and cause something bad but less bad to happen". It's just the Trolley problem restated. The solve I have for the trolley problem is viewing the agent in front of the lever as a robotic control system. Every timestep, the control system must output a control packet, on a CAN or rs-485 bus. There is nothing special or privileged between a packet that says "keep the actuators in their current position" and "move to flip the lever". Therefore the trolley problem vanishes from a moral sense. From a legal sense, a court of law might try to blame the robot, however.Should a country where cryonic preservation is routine try to take over one where it is forbidden?Aside from 'yes' or 'no' or 'it depends on X', there are also other actions that can be taken.I think it's not just that the old generation has died out. It's also that the conflict theorists shut up for a while after such a bloodshed and gave the people like Hugo Grotius a window of opportunity to create the international law.
Similar thing, by the way, happened in Europe after WWII. I've written about it here. I wonder whether this opening of the window of opportunity after a major catastrophe is a common occurrence. If so, working on possible courses of action in advance, so that they can be quickly applied once a catastrophe is over, may be a usful strategy.
Curated. I don't know whether or not LessWrong needs this story–I hope that it doesn't–but I could see that increasingly this is a story that the world benefits from existing. Short, evocative, and with a good point. Curated for your consideration.This largely captures my views about myself and choosing to follow a generally civil life -- accepting that I am not the moral authority, judge and jury even when I find my own moral senses insulted by various actions from others.I think for me though it's about not even making the choice between blue or green explicitly -- perhaps creating an internal ambiguity that I may well be a monster (when I decide to say eff it all for following civil conventions and laws) rather than the moral person I claim (make the appearance to be) by limiting my actions and let social rules govern various outcomes.If you follow a law that is grossly unjust because it's a law, or follow a social convention that is grossly unjust because it is a social convention, you would be actively choosing to contribute to that injustice. Sticking out your neck & going against the grain based on your best judgment is (I thought) a kind of rationalist virtue. Sticking out your neck is only a virtue if it ends up giving you greater expected utility than following the social norm. Sticking out your neck because you like the idea of yourself as some sort of justice warrior and ruining your entire life for it is the non-rationalist loser's choice.
The point of Johns story is that both Red and Judge are better off working together than they would be if they fought, even though they strongly disagree on the scissor statement. Fighting would in effect be defecting even when the payoff from defection is lower than the payoff from cooperation. This is basically how all of society operates on a daily basis. It's virtually impossible to only cooperate with people who share your exact values unless you choose to live in poverty working for some sort of cult or ineffective commune.
What makes Judge and Red special is that they have a very advanced ability to favor cooperation even when they have a strong emotional gut reaction to defect. And their ability is much greater than that of the general populace who could get along with people just fine over minor disagreements, but couldn't handle disagreeing over the scissor statement.
I think you're confusing rationality for plain self-interest. If you have something to protect, then it may be reasonable to sacrifice personal comfort or even your life to protect it. Also, you comment implies that the only reason you'd fight for something other than yourself is out of "liking the idea of yourself as some sort of social justice warrior," as opposed to caring about something and believing you can win by applying some strategy. And saying you'd "ruin your life" implies a set of values by which a life would count as "ruined."Seems like you're rejecting the idea that a "grossly unjust" law could be a scissors statement?I think it can be both, but I don't have the sense that something being a scissors statement means that one should automatically ignore the scissors statement and strike solely at the person making the statement. Scissors statement or not, if a law is grossly unjust, then resist it.Scissor statements reveal pre-existing differences in values, they don't create them. There really are people out there who have values that result in them doing terrible things. It's one thing when you make that choice for yourself. This is about a disagreement so heinous that you can't countenance others living according to a different belief than your own. I read JMH as arguing for a humility that sometimes looks like deferring to the social norm, so that you don't risk forcing your own (possibly wrong) view on others. I suspect they'd still want to live their life according to their best (flawed) judgment... just with an ever-present awareness that they are almost certainly wrong about some of it, and possibly wrong in monstrous ways.This is about a disagreement so heinous that you can't countenance others living according to a different belief than your own.Beliefs and values aren't just clothes we wear -- we act on them, and live by them. (And don't confuse me for talking about what people say their values are, vs what they act on. Someone can say they value "liberation for all," for example, but in practice they behave in accordance with the value "might makes right." Even if someone feels bad about it, if that's what they're acting out, over and over again, then that's their revealed preference. In my model, what people do in practice & their intent are what is worth tracking.) So it's reasonable to assume that if someone has a particularly heinous belief, and particularly heinous values, that they act on those beliefs and values.I read JMH as arguing for a humility that sometimes looks like deferring to the social normWhy should that particular humility be privileged? In choosing to privilege deference to a social norm or humility over $heinous_thing, one is saying that a {sense of humility|social norm} is more important than the $heinous_thing, and that is a value judgment.I suspect they'd still want to live their life according to their best (flawed) judgment... just with an ever-present awareness that they are almost certainly wrong about some of it, and possibly wrong in monstrous ways.If you think your judgment is wrong, you always have the option to learn more and get better judgment. Being so afraid of being wrong that a person will refuse to act is a kind of trap, and I don't think people are acting that way in the rest of their lives. If you're wiring an electrical system for your house, and you have an ever-present awareness that you're almost certainly wrong about some of it, you're not going to keep doing what you're doing. You'll crack open a text book, because dying of electrocution or setting your house on fire is an especially bad outcome, and one you sincerely care about not happening to you. Likewise, if you care about some moral value, if it feels real to you, then you'll act on it.I read JMH as arguing for a humility In order to argue for humility, one would have to give the "scissors statement".your own (possibly wrong) view on others. Have you read the fictional sequences?+12112222The rough table above describes a system in which "+" denotes a maximum operator. Applied to (x, x) it returns x, applied to (x, x+1) or (x+1, x) it returns x+1. An operator can be redefined, though a question concerning an unknown system "How does a system where "2+2=3" work?" is open ended, and doesn't contain enough information for a unique solution. It does have enough information for an arbitrary solution like:"+" takes two inputs and returns their sum (in conventional terms) minus one. This has a number of properties:a+b = b+aa+b+c = a+(c+b) = b+(a+c) = b+(c+a) = c+(a+b) = c+(b+a)That is, order independence. What it 'lacks' relative to the normal definition is distributional invariance, i.e. 1+1+1+1 is different from 2+2 because a more distributed quantity loses more in agglomeration. Though it does have a different 'equivalence' - that is, for a given relationship that can be described in "the original system" a new one can be found.1+2+2+1 = 2+2or even (not with integers*)(*nevermind it is with integers, at least in this case)x+x+x+x = 2+2can be solved because in original terms that's 4x-4 = 2+2, which is 4x-4=4, 4x=8, x=2. 2+2+2+2 (new system) = 2+2 (new system)Even the question itself becomes meaningless wouldn't it?The question (w.r.t. 2+2=3) is however meaningless in the sense it doesn't describe a system ready at hand which exists, though it could be constructed.It's an understanding that working together is better for both Judge and Red's individual utility functions than is fighting against each other. Call if moral relativism if you want, but it's more accurate to call it a basic level of logical thinking. Rational moral absolutists can agree that it makes no sense for Judge and Red to fight and leave each other either dead or severely injured rather than work together and be significantly better off.Don't bother trying to escape from here. Nobody ever has, and nobody ever will! | 2024-11-08T11:35:34 | en | train |
42,071,200 | ReadCarlBarks | 2024-11-06T23:28:58 | Help Mozilla improve its alt text generation model | null | https://blog.mozilla.org/en/mozilla/ai/help-us-improve-our-alt-text-generation-model/ | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,201 | teleforce | 2024-11-06T23:29:03 | ECG Analysis Platform Using Deep Neural Networks in Routine Clinical Practice | null | https://www.ahajournals.org/doi/10.1161/JAHA.122.026196 | 1 | 1 | [
42071265
] | null | null | null | null | null | null | null | null | null | train |
42,071,207 | PaulHoule | 2024-11-06T23:29:40 | As global warming rises, offshore hydrogen hubs could carry renewable energy | null | https://techxplore.com/news/2024-10-global-offshore-hydrogen-hubs-renewable.html | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,217 | sumarum1969 | 2024-11-06T23:30:21 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,071,228 | bookofjoe | 2024-11-06T23:32:15 | Ping Rate Clock | null | https://github.com/turingbirds/ping-clock | 2 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,260 | aard | 2024-11-06T23:34:51 | Code Reviews, Not Code Approvals | null | https://rethinkingsoftware.substack.com/p/code-reviews-not-code-approvals | 3 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,279 | toomuchtodo | 2024-11-06T23:36:58 | A new stronger Ozempic is coming. Here's what to know | null | https://qz.com/novo-nordisk-cagrisema-ozempic-1851691005 | 1 | 1 | [
42071289
] | null | null | null | null | null | null | null | null | null | train |
42,071,298 | null | 2024-11-06T23:38:12 | null | null | null | null | null | null | [
"true"
] | null | null | null | null | null | null | null | null | train |
42,071,310 | robbiet480 | 2024-11-06T23:38:58 | Australia proposes ban on social media for those under 16 | null | https://www.reuters.com/technology/cybersecurity/australia-proposes-ban-social-media-those-under-16-2024-11-06/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,321 | chany2 | 2024-11-06T23:39:41 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,071,326 | miles | 2024-11-06T23:39:56 | Windows 10 given an extra year of supported life, for $30 | null | https://www.theregister.com/2024/10/31/microsoft_windows_10_support/ | 2 | 1 | [
42071349
] | null | null | missing_parsing | Windows 10 given an extra year of supported life, for $30 | 2024-10-31T21:58:08Z | Iain Thomson |
Microsoft has thrown a lifeline to Windows 10 users ahead of the OS going end-of-life, by offering an extra year of patches for $30.
Support for Windows 10 ends in October 2025 and Redmond is pushing people to upgrade to Windows 11, with mixed success to date – as of last month, Windows 10 had 62.75 percent of Redmond's OS market share, compared to 33.42 percent for the newer version ago.
Perhaps that's why the software behemoth has decided to offer Extended Security Updates – previously only available for business, education, and government users – to anyone who wants them.
"For the first time ever, we're introducing an ESU program for personal use as well," wrote Yusuf Mehdi, consumer chief marketing officer at Microsoft. "The ESU program for consumers will be a one-year option available for $30. Program enrollment will be available closer to the end of support in 2025."
This will be a boon to those who don't care to upgrade or who can't because their PCs aren't capable of running Windows 11.
Enterprise users can pay $61 per device for an extra year of support, but that doubles the next year to $122, and again to $244 in year three. Users in the education sector have it much easier – they pay $1 per license for the first year, then $2, and then $4 per Windows 10 machine.
One-year countdown to 'biggest Ctrl-Alt-Delete in history' as Windows 10 approaches end of support
After 3 years, Windows 11 has more than half Windows 10's market share
AI to power the corporate Windows 11 refresh? Nobody's buying that
Microsoft releases Windows 11 Insider Preview, attempts to defend labyrinth of hardware requirements
Windows 11 is one of Microsoft's most poorly performing operating systems, in part due to the powerful hardware it requires. Chipmakers and PC players expect the need for upgrades to bring a payday, but that hasn't happened yet.
Part of the problem, as The Register readers have noted on our forums, is that Windows 11 isn't a significant improvement over its predecessor. While Redmond repeatedly touts the benefits of Copilot and AI, it doesn't seem to be an incentive for many people to rip and replace their hardware to take care of it.
Microsoft also risks driving users to non-Windows machines. With Apple's market share steadily growing in the US – and the iPhone's popularity – many may consider making the switch.
Or perhaps 2025 will be the year of Linux on the desktop. ®
| 2024-11-08T17:27:03 | null | train |
42,071,330 | oidar | 2024-11-06T23:40:06 | Ableton's Push 3 running Doom in standalone mode [video] | null | https://www.youtube.com/watch?v=Mk8t_wufi4k | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,377 | jinqueeny | 2024-11-06T23:43:36 | What can you do with tiny (1B/3B) LLMs in a local RAG system? | null | https://nexa.ai/blogs/small-llm-local-rag-practical-guide | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,383 | nextone87 | 2024-11-06T23:44:07 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,071,398 | mkaic | 2024-11-06T23:46:19 | The Best Things in Life Are Model-Free (2018) | null | https://archives.argmin.net/2018/04/19/pid/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,411 | ngninja | 2024-11-06T23:48:06 | null | null | null | 1 | null | null | null | true | null | null | null | null | null | null | null | train |
42,071,429 | danielam | 2024-11-06T23:49:46 | Egyptian priestess' burial chamber unearthed after 4k years | null | https://nypost.com/2024/11/06/science/egyptian-priestess-mummy-unearthed-from-extraordinary-tomb-after-4000-years/ | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
42,071,478 | iddan | 2024-11-06T23:54:30 | What if Apple made a Search Engine [video] | null | https://www.youtube.com/watch?v=J1cQtjS7AAE | 1 | 0 | null | null | null | null | null | null | null | null | null | null | train |
Subsets and Splits