Hacker Newsnew | past | comments | ask | show | jobs | submit | do_not_redeem's commentslogin

> Pure-Go tree-sitter runtime — no CGo, no C toolchain, WASM-ready.

No you didn't. The readme is obvious LLM slop. Em-dash, rule of three, "not x, y". Why should anyone spend effort reading something you couldn't be bothered to write? Why did you post it to HN from a burner account?


Zig supports this. If you compile with -fsingle-threaded, operations on mutexes turn into nops, atomics become simple loads/stores, etc.


Glibc has a bunch of tests throughout the codebase where it checks if there have been any threads started besides the main one. I don’t really know how effective they are from a performance perspective. (In principle, turning fgetc into getc_unlocked, for instance, could be quite beneficial.) Microsoft used to have a single-threaded C runtime, but it was done away with some time ago, I’m guessing because they started putting things into the platform that would start and manage random threads outside the programmer’s control.


Okay, let's say the cache is cold and you're on an old clunky spinning rust 5400 RPM hard drive. Do the math. How long will it take, worst case, for the platter to spin to where calc.exe is stored?


For a 5400 RPM drive, worst-case rotational latency is one full rotation: 5400/60 = 90 rev/sec, so ~11ms. Average is half that (~5.5ms). If you also need to seek (yes, we'll definitely need to move on both axes in the worst case scenario requested, likely all the time), 2006-era datasheets show average seek around 11-12ms, with full-stroke seeks around 21-22ms. So worst case total access: ~33ms.

Seagate Momentus 5400.3 manual (2005): https://www.seagate.com/support/disc/manuals/ata/100398876a....

Hitachi Travelstar 5K120 (2006):http://www.ggsdata.se/PC/Bilder/hd/5K120.pdf

WD Scorpio (October 2007): https://theretroweb.com/storage/documentation/2879-001121-a1...


I won't spoil the article, but if you read to the end you'll find out what this "FMZ" project is called.


Can you spoil it for me, because I read it to the end and saw no mention of such a project. Unless you are referring to the DIY approach the article suggests.


Somewhere else in the comment thread Zim (zimfw) was mentioned which after reading their website sounds pretty much like that.


It was an offhand 20 words out of a 5000 word article and a very pertinent example. I guess the question is, why were you not bothered by the negative opinion about instant coffee or particleboard or malls or Tinder or McNuggets, but hearing anything negative about LLMs is worth singling out to complain about?


The article boils down (pun intended) to, “you already accept all these things which I am glossing as basically instant mashed potatoes, so you should also accept LLMs.”

That’s why they not-so-subtly start calling them IMPs when they introduce the “abstracted version.”

It’s not merely an example. It’s the thesis of the article.

EDIT: Out of perversity, I skimmed the comments. The audience of Astral Codex Ten seems to share this interpretation, for whatever that’s worth.


I read the post as pretty clearly anti-LLM (and anti-instant-mashed-potatoes).


The site is open source and the commits are still there. No need to be so dramatic.

https://github.com/ziglang/www.ziglang.org/commit/c8d046b288...


Oh, thanks, I thought watwut meant archive.org. Is this diff also linkable on codeberg?


To be fair, Windows 8 only came out in 2012, so they haven't had that much time to finish the settings migration. But they're making good progress. If they keep up this pace of moving 2 settings per month, they should be able to finish by 2053.


Given how bad Project Reunion went, that is being too positive.


I wonder what tools the Zig team has to deal with trolls like this.

Is the zig name or logo trademarked? What about the mascot he's using as his github picture?

They're violating the terms of the MIT license as mentioned in the article, so maybe Zigtools has legal standing.

As for lying about no AI, being an asshole isn't illegal, so no angle there.

Any other ideas I missed?


Lying potentially opens up fraud angles if they are soliciting or receiving something of value. Maybe false advertising even they are giving it away for free. A lot of this will depend on who has jurisdiction


Congratulations on the move!

> Thank you to the Forgejo contributors who helped us with our issues switching to the platform, as well as the Codeberg folks who worked with us on the migration

I'd love to see a writeup about these problems/solutions at some point.


Although it's not a writeup, most of the problems can be traced through this "moving-to-forgejo" meta-issue: https://codeberg.org/forgejo-contrib/moving-to-forgejo/issue...


I think gorhill is far more trustworthy than a whole new browser based on crypto.


It's not based on cryptocurrency, there are just extra features that use it. Unstoppable domains is an optional feature. You don't need to visit them, but it gives value to people by letting them actually own their domain instead of leasing it from ICANN. Viewing ads to earn BAT is an optional feature. As I mentioned ad blocking is built in so you can have it show no ads if you want.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: