gaigalas

8 ms·

121.
by gaigalas 4mo ago
Cool. That's it. What will people use to write for Jalapeño matters. Nvidia has multi-decade heritage. Apple spent almost a decade in MLX. Snapdragon failed partly here. OpenAI announced nothing regarding to that, so this big moat that
122.
by gaigalas 4mo ago
That's interesting, however, what you describe is philosophy in a coloquial framing (non-technical, purpose-driven, etc). AI companies are hiring academic philosophers, which is something else entirely. It's a discipline that de
123.
by gaigalas 4mo ago
A lot of companies that serve a single family of LLMs seem to prefer nvidia though. Why is that? It's not just good drivers, which is what moats them for games and ML. It's a multi-decade work of making chips that are nice to pro
124.
by gaigalas 4mo ago
> At the beginning of my career, I believed I’d be rewarded for the originality of my ideas. The truth is that you’re rewarded for identifying and solving problems efficiently. The "I'll be original and get directly rewarded&qu
125.
by gaigalas 4mo ago
But nvidia's moat is software support, isn't it?
126.
by gaigalas 4mo ago
I like the mantra: "prefer duplication over the wrong abstraction". Combined with another interesting idea "whatever I dislike is wrong", it makes me _always be right_, which is awesome. I can never lose a discussion abo
127.
by gaigalas 4mo ago
> a shared preoccupation with artificial intelligence, longevity, and the near future There are strong historical trends (going thousands of years) associating obsession with those three items (human mimicry, living forever, predicting t
128.
by gaigalas 4mo ago
> 40,0000+ years of oral Thanks, now I know exactly that it's you who was no idea what you're talking about.
129.
by gaigalas 4mo ago
It's not, you're being silly.
130.
by gaigalas 4mo ago
Possibly. You are always free to correct me with accurate information though. Your choice.
131.
by gaigalas 4mo ago
Animals, people, clay pots. It's textbook cave painting from that era, and no myth can be recovered from it. The only myths in cave paintings are the ones modern people project back when romanticizing them.
132.
by gaigalas 4mo ago
You are defending them from an attack that doesn't exist. I explicitly praised them, and claimed that we should understand their work. They're great developers. Also, what I did was make their myth-making fanbase uncomfortable. If
133.
by gaigalas 4mo ago
The oral tradition you're probably trying to refer to is the Rigveda, which is amazing. It's like a human recorder that kept a series of vocalizations intact for thousands of years through chanting, with error correction mechanism
134.
by gaigalas 4mo ago
Idolatry is not the same as mythologizing. And I never said there is a banality to it, just that there could be . We don't have enough to know.
135.
by gaigalas 4mo ago
It's an interesting hypothesis we can never prove. We don't even know how old names are.
136.
by gaigalas 4mo ago
I think it's the wrong lens for observing this conversation. You're looking for something that I might be attacking. I'm not doing what you think I am, that's why you can't pinpoint it. It's a simple observatio
137.
by gaigalas 4mo ago
There's no such thing as "human nature", that's just a way to justify something that can't be easily explained. I have nothing against it. The fact that I explained a mechanism (mythologizing diminishes one's r
138.
by gaigalas 4mo ago
Humanity has some 300.000 years of existing, and we can only trace back the prevalence of cult figures a few thousand years back. For all we know, it could be a temporary fluke and we'll snap back to something else. We could be beings
139.
by gaigalas 4mo ago
Honestly, two mythologized figures (Carmack and Bellard). They're good (like, quite good), but as soon as their names come up people start talking about some weird expectation of what they are supposed to think rather than the actual
140.
by gaigalas 4mo ago
> The first is to self host. You buy the machine, run open source models locally, and pay nothing per token after that. In the good ol' days, we bought machines not only to run stuff, but to experiment . I understand today experime
141.
by gaigalas 4mo ago
> Does open source matter anymore? Probably not, but maybe free software now matters more.
142.
by gaigalas 4mo ago
Man, Opus 4.8 is feeling a lot smarter in the last few interactions. Is Anthropic silently serving Fable as Opus just to stick it to the man?
143.
by gaigalas 4mo ago
It behaved proactively in one scenario . Perhaps, when it doesn't have tricks in its sleeve, it doesn't do that. The text is not an evaluation of a major trend in behavior (which could be true or false). Another way to frame it
144.
by gaigalas 4mo ago
We're about to win Genocide Bingo. Rule 1 of Genocide Bingo is: Don't win Genocide Bingo. https://www.youtube.com/watch?v=4kDPxbS6ofw
145.
by gaigalas 4mo ago
Let the battle for the harnesses give free tokens for all, until the next competition arena does the same. It seems that's the only way AI will remain acessible. This website is gorgeous, by the way. The mouse reveal on the background,
146.
Xkcd-mural – in-browser AI builds a mural with elements from strips you read(alganet.github.io)
2 points by gaigalas 5mo ago|1 comments
147.
by gaigalas 5mo ago
Here's an example mural: https://imgur.com/a/TPZLPxC Source code available here: https://github.com/alganet/xkcd-mural
148.
They're Made Out of Makes(gist.github.com)
5 points by gaigalas 5mo ago|0 comments
149.
by gaigalas 5mo ago
Lots of cheap chinese RAM is exactly what we need right now. RAM used to be cheap, it was a common excuse for not optimizing software 10 years ago ("this thing is cheap, don't worry about Electron!")
150.
by gaigalas 5mo ago
Perhaps all hobbyst developers interested in local AI combined is a smaller demand than AI companies hoarding parts. That would make demand decrease.

More ›