News for hackers. Stuff that matters.

almostgotcaught

> For whom?

what is the name for this kind of pointless, lazy, selective, quoting that willfully misconstrues what's being quoted? the answer to this question is incredibly clear: for the developer that created this tool. if that makes you unhappy enough to malign them then maybe you should just not use it?

There are so many people on here that are outsiders commenting way out of their depth:

> Google Gemini already uses their own TPU chips

Google has been using TPUs in prod for like a decade.

> soon

When people say things like this I always wonder if they really think they're smarter than all of the people at Nvidia lolol

Re:Memory Subsystem Optimizations on Friday January 02, 2026 @01:11AM (#46461928)
> neckbeard

Hmmm didn't realize proper grammar and syntax was neckbeard territory. TIL!

Re:Memory Subsystem Optimizations on Friday January 02, 2026 @12:19AM (#46461677)
> The use tracks with the Latin or English definitions

No it doesn't

> sequitur noun : the conclusion of an inference : consequence

https://www.merriam-webster.com/dictionary/sequitur

> "malapropism" is a five dollar word

It is of course but I spent my $5 wisely because my use is syntactically and semantically correct.

Re:Memory Subsystem Optimizations on Thursday January 01, 2026 @09:19PM (#46460655)
> sequitur

I love the malapropisms on hn because it always reeks of "I'm trying so hard to sound smart" lol. FYI non-sequitur doesn't mean "non-sequential" it means "illogical" (and thus sequitur doesn't mean "in sequence"). Also both of words these are nouns not adjectives.

Re:A Vulnerability in Libsodium on Wednesday December 31, 2025 @12:15PM (#46446065)
Lol if you say so. I contribute to an OSS project with thousands of industry users and we break downstreams all the time - we literally have no stability guarantee. In the 2 years I've been a contributor I've seen exactly once when someone got upset about a breakage.
Re:A Vulnerability in Libsodium on Wednesday December 31, 2025 @09:01AM (#46444102)
> and you can basically forget about ever meaningfully refactoring that codebase.

Ummm why? Breaking changes aren't the end of the world? Deprecate and communicate clearly and people are usually fine with them (if it's meaningful progress instead of churn).

Re:CEOs are hugely expensive. Why not automate them? on Monday December 29, 2025 @04:54PM (#46426226)
> pretty reasonable assumption.

Sure if you have no clue about the frontier-lab landscape then it's a pretty reasonable assumption.

Is there like a list of canned responses that low-effort commenters just rotate through? Here lemme try:

This would've been 39,000X better if written in mojo.

Re:List of domains censored by German ISPs on Monday December 29, 2025 @02:46PM (#46424659)
> German government is just as draconian as the US

this is called "disinformation"

Re:Feynman's Hughes Lectures: 950 pages of notes on Monday December 29, 2025 @12:45PM (#46423166)
Greiner's QFT book is by far the best I've ever seen.
Re:CEOs are hugely expensive. Why not automate them? on Monday December 29, 2025 @08:43AM (#46420674)
> xAI is now firmly in the top 4 of AI companies worldwide

Lolol literally no one thinks this

How do people get away with this kind of dishonesty today? It's shameful.
Re:Rob Pike Goes Nuclear over GenAI on Friday December 26, 2025 @01:36PM (#46394806)
> tech workers to be quite intellectually lazy and revisionist

i have yet to meet a single tech worker that isn't so

Re:CUDA Tile Open Sourced on Thursday December 25, 2025 @10:20PM (#46388911)
it's all over this thread (and every single other hn thread about GPU/ML compilers) - people quoting random buzzword/clickbait takes.
Re:CUDA Tile Open Sourced on Thursday December 25, 2025 @10:19PM (#46388905)
> Google leading XLA & IREE

IREE hasn't been at G for >2 years.

Re:FPGAs Need a New Future on Tuesday December 23, 2025 @10:38AM (#46366166)
> we need a tiled layout

You are presuming an existing tile IP - if you're already in possession of such an IP then the place and route is already coarse grained. There are lots of papers on this.

> (like the GPU does)

What exactly does the GPU do? Yes there are tiles but it's up to you to now tile your workload. You understand this is the exact same problem you're bemoaning re place and route - you need to figure out how to shuffle individual bits efficiently through an existing fabric (roughly it's the same thing as routing 32 wires at a time).

> we need for this to happen is for the companies making the FPGAs to open up the board layout file spec

What exactly is this going to do for you if you're placing tiles? Also you can already recover this by exhaustively enumerating all a->b paths (yes people really do this).

To anyone else that thinks they just absolutely are certain of the silver bullet for digital design: download a copy of Vivado and report back what you discover!

Re:Mathematicians don't care about foundations on Sunday December 21, 2025 @02:06AM (#46342855)
> The time it needs to run is irrelevant for its correctness. And so they can stack and stack and stack

This is a very naive take - the very direct translation of what you're saying doesn't happen does in happen in analysis all the time: there are many inequalities which can be "stacked" to prove a bound on something but their factors are too large so you cannot just stack them if you need a fixed bound for your proof to go through. Unsurprisingly this is exactly how actual runtime analysis also works (it's unsurprising because they're both literally math).

Re:1.5 TB of VRAM on Mac Studio – RDMA over Thunderbolt 5 on Thursday December 18, 2025 @08:38PM (#46321326)
I don't know what you're bemused by - there's no mystery here - you can read the release notes where it literally says this was added to support MLX:

https://developer.apple.com/documentation/macos-release-note...

Which I'm sure you saw in literally yesterday's thread about the exact same thing.

Re:Developers can now submit apps to ChatGPT on Thursday December 18, 2025 @05:54AM (#46311171)
You can always depend on "brilliant" hn users to contribute the most braindead business hot-takes (not you but the person you're responding to).
Re:OBS Studio Gets a New Renderer on Wednesday December 17, 2025 @10:49PM (#46308738)
No it's not - the compiler for MSL is of course C++ because it's LLVM but the runtime is absolutely written in objc (there weren't even C++ bindings until recently).
> I am now confident that within 5-10 years (most/all?) junior & mid and many senior dev positions are going to drop out enormously.

yes because this is what we do all day every day (port existing libraries from one language to another)....

like do y'all hear yourselves or what?

Agree with the first part

> he was basically an IC

Disagree with this part - ICs have to write code. He literally did nothing except meetings and WP posts.

Re:Pricing Changes for GitHub Actions on Tuesday December 16, 2025 @02:42PM (#46293387)
> Perfectly good excuse to make society worse for people

What an incredibly silly accusation to make of a company/service that streams movies and television. Like you understand it is possible to dilute the concept of civic responsibility right?

Re:Compiler Engineering in Practice on Monday December 15, 2025 @02:10AM (#46271302)
^ Found a person who hasn't worked on high-throughout compilers!
Re:Compiler Engineering in Practice on Monday December 15, 2025 @12:13AM (#46270706)
Ladies and gentlemen dunning-kruger contender #2
Re:Compiler Engineering in Practice on Sunday December 14, 2025 @03:51PM (#46266768)
This has gotta be one of the most dunning-kruger comments on hn
> You say that like it’s even remotely feasible at the frontier of mathematics and not a monumental group effort to turn even established proofs into such.

people on hn love making these kinds of declarative statements (the one you responded to, not yours itself) - "for X just do Y" as a kind of dunk on the implied author they're responding to (as if anyone asked them to begin with). they absolutely always grossly exaggerate/underestimate/misrepresent the relevance/value/efficacy of Y for X. usually these declarative statements briskly follow some other post on the frontpage. i work on GPU/AI/compilers and the number of times i'm compelled to say to people on here "do you have any idea how painful/pointless/unnecessary it is to use Y for X?" is embarrassing (for hn).

i really don't get even get it - no one can see your number of "likes". twitter i get - fb i get - etc but what are even the incentives for making shit up on here.

"not bad, not bad at all"

Older »