News for hackers. Stuff that matters.

igravious

"Top all-time" should be called "Most mentioned" and it should display the mentions stat followed by the recommends stat

"Most recommended" can stay as it is

and then a new "Top all-time" should be a weighted sum of recommendations plus mentions with recommendations getting a higher weight (obviously) … l33t c0d1ng sk1llz for deducting negative mentions (as in "awful book", "do not recommend" etc)!

Re:Show HN: A working Nintendo in ~1GB of Qwen weights on Friday October 30, 2026 @01:52PM (#50307637)
Sometimes projects have no point other than to show what can be done – and the pure thrill of doing is an end in itself … but I agree with you in the "so what" sense here … I feel like with this project there is meant to be a point (other than play for the sake of play and art for the sake of art and coding for the sake of coding) but I'm not seeing it either?
Re:US tech stocks see sharp falls over AI cost fears on Friday October 30, 2026 @01:43PM (#50307486)
EUV, or EUVL, no?
Re:Revolut down as users report app not working on Tuesday October 20, 2026 @09:24AM (#50166039)
it's a strawman that the only two available options are the mutually exclusive "grin and eat shit" one or alternatively a heated argument

i stand by what i said here down-votes be damned; judging by our brief interaction and your telling anecdote you come across as a smug disagreeable know-it-all and in their position i wouldn't have hired you -- there is after all the option to hire someone who both knows their stuff and doesn't come across the way you do

Oh Revolut's down are they? Let me tell you about an interview I did with them. Blah blah blah. Yeah, no thanks.

Re:Revolut down as users report app not working on Tuesday October 20, 2026 @08:03AM (#50165026)
oh!

who or what is mrkeen?

Re:Revolut down as users report app not working on Tuesday October 20, 2026 @08:00AM (#50164990)
even if you were technically correct why would a company hire someone who demonstrates that they are unable to disagree in such a way that a heated argument does not ensue? i'd say they dodged a bullet.
Re:DeepSeek v4.1 Flash on Thursday September 10, 2026 @12:59PM (#49646897)
https://news.ycombinator.com/from?site=twitter.com

There have been 34 Twitter/X link submissions in the past day, ~that's 12,000 submissions a year.

If your reason is that you have to be logged in to use it properly then I'd nearly agree with you. If it's for any other reason, how about no?

Re:Terrance Tao explains 6 essential mathematical concepts on Tuesday September 01, 2026 @04:56AM (#49519622)
Exactly; couldn't have put it better myself.
Re:Show HN: The load-bearing vocabulary of Claude on Friday August 28, 2026 @02:30AM (#49475057)
My Claude too, have no fear. You are not the only one.
Intriguing. Who is it? What is it? any ideas?

https://openrouter.ai/stealth/ox-alpha

Re:Say It Four Times (In Your System Prompt) on Sunday August 23, 2026 @04:07PM (#49412095)
Now fill the context in batches of 10% and see where its obedience drops off
I like Grok 4.6 the model, I like it for a number of reasons. I like Grok Build too.

However, I've pointed Grok 4.6 at a fairly complex codebase and asked it to review/audit it for issues and it's come back with a whole laundry list of issues. I've passed that list to Kimi and Claude and they both were like "a couple of good catches but some of those are not issues at all". Grok 4.6 is noticeably weaker that Claude Fable 5, Claude Opus 5, Claude Opus 4.8, Kimi K3, GLM 5.3, … your suggestion to use Grok 4.6 instead of a recent Claude doesn't pass empirical scrutiny.

ah, i see -- you're one of those people

> There are no other models out there for coding.

that wasn't your claim.

your claim was:

> There are so many other better models right now.

wrt coding that is untrue.

not for coding.
Also related and recent: https://news.ycombinator.com/item?id=49265305

"Claudish to English" (https://github.com/gvzdv/claudish-to-english)

4 points | bryan0 |10 days ago | 2 comments

Re:Clean up Claude 5's token vomit with a separate LLM on Thursday August 20, 2026 @05:48PM (#49380725)
Sam? Is that you?
Re:Show HN: Huzzah – a novel approach to coding with AI on Thursday August 20, 2026 @05:06PM (#49380276)
Clever! :)

Difficult! :(

Re:Why does Opus 5 feel worse to work with? on Friday August 14, 2026 @11:20PM (#49307277)
create a file called (for instance) `.ai-instructions.md` and symlink both `CLAUDE.md` and `AGENTS.md` to it
Amen brother, at this point I just copy and paste Claude's (Opus 5, Opus 4.8 -- doesn't matter which) summaries over to the window Kimi is in and:

   this is from claude, turn it into English for me would you?
   """
   [claude's tortuous prose]
   """
No amount of asking it to answer me in a straight-forward manner, to be succinct, to not use phrases like "honest caveat", "crux", "load-bearing", "blocker", etc ever sticks for more than a few turns … coupled with the fact that it can ignore instructions and do its own thing and then what I can only describe as lie about it using Claude can be an exercise in frustration. Kimi and GLM talk to me like a human, Luna/Terra/Sol are much better in that respect also, and Grok is marvelously structured and bullet-pointy in its explanations but unfortunately it is not as strong …
hats off -- fantastic work -- worked out the box

   Operating System: TUXEDO OS 
   KDE Plasma Version: 6.6.5
   KDE Frameworks Version: 6.24.0
   Qt Version: 6.10.2
   Kernel Version: 6.17.0-124040-tuxedo (64-bit)
   Graphics Platform: Wayland
   Processors: 24 × AMD Ryzen 9 3900 12-Core Processor
   Memory: 32 GiB of RAM (31.2 GiB usable)
   Graphics Processor: NVIDIA GeForce RTX 2070/PCIe/SSE2 (8 GiB)
   Manufacturer: PC Specialist LTD
   Product Name: NH5xAx
My DeepSeek V4 can see!
Re:DeepSeek Harness on Thursday August 13, 2026 @03:56PM (#49291081)
DeepSeek Harness -- your common-or-garden coding harness https://venturebeat.com/technology/deepseek-harness-launches...
Re:Gemini 3.7 Flash on Thursday August 13, 2026 @02:37PM (#49290172)
why Grok not through `Grok Build` ?
Re:I'm done coding with AI [video] on Thursday August 13, 2026 @02:32PM (#49290112)
feel for the guy, but times change and we gotta change with them, such is the nature of things

i've been thinking about this upending process a bit (as i'm sure we all have) and the nearest historical parallel to the backlash against AI slop is the arts & crafts movement's rejection of mechanised production, from Wikipedia “Initiated in reaction against the perceived impoverishment of the decorative arts and the conditions in which they were produced” https://en.wikipedia.org/wiki/Arts_and_Crafts_movement

Re:Principia Mathematica is modern and insightful on Thursday August 13, 2026 @03:25AM (#49282732)
The Begriffschrift has in no way been consigned to the rubbish heap of history. What gave you that impression? It is seminal. That it had one unresolved paradox in its set-theoretic foundations does not scupper the philosophical insights, nor the creative notation, nor the more-or-less novel approach of conjoining mathematical functions and logic to give us predicate logic (apologies for this brutally simplified sketch)

i like to think of Frege and the Begriffschrift like this

Boole: logic + algebra = algebraic logic

Frege: logic + functions = predicate logic

ergo, if Boole is rightly deified then so should Frege regardless of minor infelicities (which prompted type theory anyhow) -- again, apologies if this is totally misleading

Re:Principia Mathematica is modern and insightful on Thursday August 13, 2026 @03:19AM (#49282693)
It is not. The foundation of math is contested -- but afaik it is widely held that HoTT is the, erm, hottest contender to the throne https://en.wikipedia.org/wiki/Homotopy_type_theory
Re:Principia Mathematica is modern and insightful on Thursday August 13, 2026 @03:03AM (#49282606)
Logicomix is novel, and done well, but flawed … it's deficiencies lie in what it leaves out which may come across as an unfair charge but in this case the charge is warranted. There is a more historically correct and less orthodox work waiting in the wings for whosoever should attempt it.
Re:LLMs reward expertise on Wednesday August 12, 2026 @04:28PM (#49278118)
Exceptional bovines.
Re:We just raised $400M in Series C on Wednesday August 12, 2026 @01:33PM (#49275955)
I have genuinely never heard of Lovable and I thought I was relatively plugged in to the tech zeitgeist :/

Maybe I'm too much of a CLI guy (always preferred Vim & co. to Rubymine/VS Code/etc)

Older »