News for hackers. Stuff that matters.

xarope

Re:Scaling Git at OpenAI on Tuesday November 03, 2026 @02:40AM (#50349174)
They are using one of the oldest tricks in the book, rapid evolution as per Nature's handbook for how to constantly improve. By allowing the millions of monkeys^W AI agents total autonomy to "Throw Things at the Wall and See What Sticks"(TTWSWS, TM), this has lead to unprecedented and amazing progress towards AGI.

In doing so, they obviously need to keep track of the myriad of changes, hence the requirement to ensure Git can scale up.

(I'd like to think /s, but you know what, maybe not)

Re:My Favourite Laptop on Monday November 02, 2026 @09:56PM (#50347448)
my previous favorites were an X220 and a T480s. The T480s ran for 8 years, started losing bits of rubber from the screen area, to be replaced this year by a framework 13 pro, which will probably be my favorite lap for another 8 years, or maybe more (depending on how you accept the ship of theseus)
Re:F-Droid 2.0: A New Chapter for Android Freedom on Friday September 25, 2026 @03:43AM (#49841398)
Absolutely. I'll search for a GOOGLE app to install on a new phone, and it's 3rd on the list, with two completely unrelated apps showing up at the top, waiting to be accidentally clicked and installed, and probably then start popping up ads and other malware.
Re:Steam Frame starts at $1059 on Tuesday September 15, 2026 @12:57AM (#49707901)
Pistol whip, beat saber and audio trip are in my rotation. Also I expect you to die I/II/III or whatever the numbering is.
Re:Intellectual Fly Is Open on Monday September 07, 2026 @01:08AM (#49594115)
wasn't it thoreau who said "Not that the story need be long, but it will take a long while to make it short"

personally, I think there's a time and place for short versus long, just like there's a time and place for a 45mins TV episode versus a 2 hour marathon movie.

Re:Claude Fable 5.1 and Claude Mythos 5.1 on Wednesday September 02, 2026 @02:22AM (#49532440)
and when future LLMs are trained on this style, the prose (if I can call it that) becomes even worse?
Re:ArXiv has almost 600 submissions today and most are AI slop on Tuesday September 01, 2026 @03:37AM (#49519134)
Academic writing uses words with meaning in very domain specific ways; those in the domain understand it, and those outside find it confusing.

If an AI just randomly throws out domain specific words without that domain's specific meaning, it becomes very painful for everybody to read, and worse still adds a level of ambiguity that defeats the whole purpose of that academic writing in the first place.

Yes, this was a bit of a surprise to read when dealing with crypto: "This one lives in CIRCL's threshold RSA implementation... float64 has a 53-bit mantissa... the result is silently rounded before it is ever cast back to an integer. "

this is numerical methods 101, i.e. why 1+1 != 2

I'm a bit surprised that this is from cloudflare. Anybody from CF care to comment? Is this like prototype/super skunkworks code that somehow made it through?

are you allowed to video them to see whether they can actually do a good job?
-Phone notification- Your chinchilla rug had finished the wash cycle

I don't have a rug... oh wait...

I feel like this is the sort of thing William Gibson should write...
Re:Many Let's Encrypt renewals had errors today on Friday June 19, 2026 @01:37AM (#48595142)
That would a Microsoft'ese, "Some regions are encountering issues" => "The entire world is down, but our status page is working"
Re:Starfish by Peter Watts (1999) on Thursday June 11, 2026 @04:59AM (#48487967)
I've read blindsight and echopraxia, but not starfish, so thanks for the reminder.

If starfish is even despairing than blindsight and echopraxia, then this should be "fun"!

part of this I understand is survival. And I understand why you do this.

The other part of this is why it's so frustrating for me to find the right person. Everybody's resume looks perfect for the role, and I have to waste 30-45 mins digging into their actual experience. You have done yourself a great disservice by wasting that time sitting in an interview you were not qualified for, and worse still I always feel there are other, more qualified people, who I have missed/passed over their resume since it wasn't AI tuned.

Sincerely, I don't know how t make this better.

Re:The Eternal Sloptember on Monday May 25, 2026 @02:16AM (#48263953)
that's a very dangerous analogy, because you would be considered the domain expert and you are just asking for synonyms for something you already know but may not remember off-hand.

now, what if you asked for the synonym for "provides" in a language that has gender differences (e.g. spanish/portuguese) as well as societal nuances (e.g. japanese) and it gives you "confers", how would you now know that's correct?

ah, so you say you tell it to take into consideration gender differences, as well as societal nuances. What are those, if you were not already familiar with the language?

Re:Migrating from Go to Rust on Monday May 25, 2026 @01:48AM (#48263790)
if you are hitting pauses due to GC issues, you should into putting appropriate data structures into a memory arena, here's a reasonable read:

https://uptrace.dev/blog/golang-memory-arena

These are all tools. Java used to have this all the time, and we (ex-java programmer) had ways around this until the JVM improved.

something about the current trend of AI vulnerabilities reminds me of the old Asimov Robot series and the three laws of robotics:

https://en.wikipedia.org/wiki/Three_Laws_of_Robotics

In healthcare, HIPAA/GDPR equivalent would block this. Let's be realistic in our discussion; this is not the same as google buying up a library worth of books, scanning and destroying them
the consensus seems to be that the foam itself is the spring (hence the successful adidas evo sl and dynafish xiaonian), and the carbon plate/rod/whatever is more to control/manage that "spring".
Re:Cybersecurity Looks Like Proof of Work Now on Thursday April 16, 2026 @02:24AM (#47789368)
I can see the dichotomy forming in the "post AI" world;

1) massive companies spending millions of tokens to write+secure their software

2) in the shadows, "elite" software contractors writing bespoke software to fulfill needs for those who can't afford the millions, or fix cracks in (1)

(Oh wait, I think this is what is happening now, anyway, minus the millions of tokens)

Re:DIY Soft Drinks on Monday April 13, 2026 @04:40AM (#47749399)
can't several small hometown grocers get together to be able to make that volume/quantity? Surely that would be better than just buying from costco?
Re:A Perfectable Programming Language on Sunday April 12, 2026 @11:33PM (#47747265)
interesting the ones they chose to name; I would have probably started with 6502/68000/68020/z80 assembly, fortran, cobol, basic, c, ada, simula 67, sh, zsh, bash, napier 88, tcl, perl, rexx, before hitting the next generation of python, c++, etc.
Re:Ask HN: Any Interesting Niche Hobbies? on Thursday April 09, 2026 @08:57AM (#47703129)
somehow mixed grip makes me twist slightly which caused me back problems; after a few tweaks to the lower back, I decided that mixed wasn't worth it (I don't compete anymore). Yes, you are right, I do use the hook grip, since I also do some weightlifting.
Re:Ask HN: Any Interesting Niche Hobbies? on Thursday April 09, 2026 @03:10AM (#47700287)
my clothes haven't changed in size, but I fill them out better, is probably best to describe it!
Re:Ask HN: Any Interesting Niche Hobbies? on Thursday April 09, 2026 @03:08AM (#47700271)
ditto. I don't have great numbers, my bench is terrible, not even 1.5x BW, but I did manage to get my DL to ~2.5x in a competition, then tweaked my back and now only double overhand, which means my grip tends to go when I get to ~2x BW (my gym doesn't allow chalk, and I don't use straps, so it is what it is).
Re:Ah, peptides. where to begin? on Tuesday April 07, 2026 @04:14AM (#47672134)
It's interesting you mention this (anhedonia), since the guardian just published this article: https://www.theguardian.com/science/2026/apr/06/is-retatruti...

How did you actually feel? Disinterested in stuff, ennui, or other?

wasn't one of the controversies, that the simulation didn't account for the fact that Red's boats couldn't actually/launch carry the cruise missiles that were used to sink these ships?

be that as it may, the lesson still stands

Absolutely. Let's say I have a problem with gRPC and traced it to code generated using the gRPC compiler. I can reproduce it, highlight it and I'm pretty sure the gRPC team would address the issue.

Replace gRPC compiler with LLM. Can you reproduce? (probably not 100%). Can anybody fix it short of throwing more english phrases like "DO NOT", "NEVER", "Under No Circumstances"?

Probably not.

Re:How the Turner Twins Are Mythbusting Modern Gear on Friday March 20, 2026 @01:34AM (#47450864)
I've done a lot of long hikes (200+km in the sahara, 6000+m mountains in kazakstan), and 2kg extra means a lot, like the difference between carrying extra fuel/food versus just clothing.

Anyway, you can try it yourself, wear a 2kg wax cotton jacket versus a 500gm technical jacket and see how you feel after a day's hiking.

Re:Show HN: RAG knowledge base poisoning lab, 100% local on Thursday March 12, 2026 @11:37PM (#47360404)
keen eye. 4 days old account, verbose comments.

Sigh.

As far as I know, the problem is still how to segment data flow from control plane for LLMs. Isn't that why we still can prompt inject/jail break these things?

Older »