tucnak

33 ms·

511.
by tucnak 2y ago
I guess it's fine for what it enables, Tree[T] of some kind basically. But we had that with codegen already.. People who really needed trees, could do that forever. However, the issue is the "unwashed"—like they said in adjac
512.
by tucnak 2y ago
And they were right. The generics in Go have turned out dogshite... I try to avoid them at all costs because besides simple data structures like Trees and so on, they're just more trouble. Go was not meant to have generics.
513.
by tucnak 2y ago
For a community supposedly "committed to providing a friendly, safe and welcoming environment for all, regardless of level of experience"[1] Rust activists perhaps unsurprisingly lack in tolerance, empathy, and awareness. The comm
514.
by tucnak 2y ago
> made by morons for morons The creators of Go have also built the foundation of everything you take for granted. The real "harsh truth" here, of course, is that the Go team exhibits engineering genius, taste, and particularity
515.
by tucnak 2y ago
They're showcasing 3d sprite generation capability; it's pretty big for game-dev, if it also generalises well.
516.
by tucnak 2y ago
The Rust people pursue "solidarity" as a virtue. They don't understand that factions is a way of life, so any sufficiently impactful technology will be "fractured" to some extent. This is a good thing. Unitarity, so
517.
by tucnak 2y ago
In the context of OP, the corporate structure of SpaceX et al. is about as far from interesting as it could possibly get. The obsession some posters have with rich people (and Musk in particular) is really unhealthy!
518.
by tucnak 2y ago
qubes is x86-only that's why people hate it
519.
by tucnak 2y ago
https://en.wikipedia.org/wiki/Misinformation_in_the_Israel%E...
520.
by tucnak 2y ago
The ship is owned and operated by russians.
521.
by tucnak 2y ago
> then they don't need to make joins between sqlite dbs. The extension could also provide custom index access methods (considering that SQLite only has a handful of column types in the first place.) That would allow you to incorpora
522.
by tucnak 2y ago
@dang may I ask if this submission and https://news.ycombinator.com/item?id=42174404 is receiving many downvotes, or just not seeing attention? Just bad luck, no suspicious downvotes at play? A wild guess... Sorry to bother
523.
by tucnak 2y ago
The o1 connection is made through "Evaluation of openai o1: Opportunities and challenges of AGI"[63]—a paper mill product with 50 or so authors. They created that 280-page monstrosity in less than two weeks of the o1 release. Did
524.
by tucnak 2y ago
I mean, most "servile" people don't think of themselves that way; usually, they only do—if some third party actor managed to impose their judgement on them, fuelled by the actor's own resentment, or personal gain. For in
525.
by tucnak 2y ago
Grammar sampling has been around for months, and remained largely unexplored. Don't fall into the common trap of thinking in a fixed language, rather think about a superset of possible languages (grammars) and how they could evolve fro
526.
by tucnak 2y ago
To add, nothing is wrong with servile mentality! Not everyone wants to be the big-time leader.
527.
by tucnak 2y ago
You should make this a tag-line for your business so that the prospective clients are aware of your delusions.
528.
by tucnak 2y ago
> if you aren't thinking about Amdahl's law you're just a hacker and not actually an engineer. This is really funny stuff, thank you!
529.
by tucnak 2y ago
The "naive", all-or-nothing view on LLM technology is, frankly, more tiring than the hype.
530.
by tucnak 2y ago
To be fair, top-of-the-line proprietary models suck at i18n. Try gemma-27b it's much better
531.
by tucnak 2y ago
"Pelican on a bicycle" is really fun, honestly. I hope it doesn't go the way of the unicorn! :-D
532.
by tucnak 2y ago
Hey, Simon! Have you ever considered to host private evals? I think, with the weight of the community behind you, you could easily accumulate a bunch of really high-quality, "curated" data, if you will. That is to say, people woul
533.
by tucnak 2y ago
So far, non-English inputs have been most telling: I deal with Ukrainian datasets mostly, and what we see is OpenAI models, the Chinese models, of course, and Llama's, to admittedly, lesser extent—all degrading disproportionately compa
534.
by tucnak 2y ago
They say a lot of things, like that their base models weren't instruction-tuned, however people have confirmed that it's impossible to find instruction that it wouldn't follow, and the output would indicate that exactly. The
535.
by tucnak 2y ago
> All the big guys are hiring domain experts - serious brains, phd level, in some cases I don't know where this myth had originated, and perhaps it was true at least at some point, but you just have to consider that all the recent m
536.
by tucnak 2y ago
The issue with some recent models is that they're basically overfitting on public evals, and it's not clear who's the biggest offender—OpenAI, or the Chinese? And regardless, "Mandelbrot in plaintext" is a bad choic
537.
by tucnak 2y ago
https://gaultier.github.io/blog/perhaps_rust_needs_defer.htm...
538.
by tucnak 2y ago
Thank you
539.
by tucnak 2y ago
I'm sorry, I don't understand either. You use these hundred-dollar words that sound smart, but their meaning escapes me completely.
540.
by tucnak 2y ago
Only somebody who didn't live in russia during the 90s would suggest it were a "functioning democracy." The regime in russia is not about putin, it's the spooks meet apparatchik's regime, and have been such for deca

More ›