cafebabbe

5 ms·

31.
by cafebabbe 6mo ago
Try multithreaded code, it's wild (how wrong it is). I wonder how/where next models will learn thread safety.
32.
by cafebabbe 6mo ago
Humans are notoriously bad at game theory.
33.
by cafebabbe 6mo ago
Not by everyone, so that would be better than nothing.
34.
by cafebabbe 6mo ago
They are even playing the "come back, i've changed" card recently :D
35.
by cafebabbe 6mo ago
Aaah, Ubuntu Mono, i found you, you little rascal
36.
by cafebabbe 6mo ago
I like the declarative nature of it. It makes it so easy to debug anything, with "simple" introspection tools. I feel that many horrors will be created when we introduce control flow to CSS. And this project kinda show how far you
37.
by cafebabbe 6mo ago
Seriously? Your takeaway from this is bad armor bonus computations?
38.
by cafebabbe 7mo ago
Only with chrome devtools :)
39.
by cafebabbe 7mo ago
Socrates was proven dead wrong by neurobiology. LLMs making you dumber is far from being "disproven" by science. Quite the opposite https://arxiv.org/abs/2506.08872
40.
by cafebabbe 7mo ago
AI is useful when paired with an experienced programmer. Experienced through old-school (pre-LLM) practice. I don't clearly see a good endgame for this.
41.
by cafebabbe 8mo ago
Question is, did they even realize they added a network-aware rendering stack...
42.
by cafebabbe 8mo ago
Of course, we are stripping the earth bare to build word-guessers GPUs in orbit, but aliens are definitely the problem.
43.
by cafebabbe 9mo ago
Any particular example you could provide to support this wild claim ?
44.
by cafebabbe 11mo ago
Seems like the whole world forgot what this job was really about :/
45.
by cafebabbe 1y ago
Ah so this is where the current GDP growth comes from.
46.
by cafebabbe 1y ago
I'm a big AI supporter. I'm just waiting for the slop to be so metastasized that our terminally ill "social networks" finally die, alongside with it. Of course i'll be proven terribly wrong. But hey. hope.
47.
by cafebabbe 2y ago
Code is law, and it's becoming more & more like this every day, so it's kind of weird that people insist it's feasible to be able to discuss technology without any of the its consequence & implications. Plus, when i t
48.
by cafebabbe 2y ago
I don't see what's preventing you to discuss "ideas" on Bluesky, or Twitter. I don't discuss politics either, but i had good interactions on both of these platforms.
49.
by cafebabbe 2y ago
Adtech used AI to make money before it was cool. You wouldn't believe how many ML models are required to build those shitty little pieces of HTML.
50.
by cafebabbe 3y ago
"gross", lol. Are you 4 year old ? Also, having some hair (and decent hygiene ofc) makes sweat more bearable imho.
51.
by cafebabbe 3y ago
Survivorship bias is always interesting to observe.
52.
by cafebabbe 3y ago
Now let's have look at similarly-worded ads for the 2000s dotcoms
53.
by cafebabbe 6y ago
It's basically an IRL version of the Total Perspective Vortex.
54.
by cafebabbe 6y ago
Looks cool. But I think the people in the comments need to lower their expectations a lot, theme will be mario-only and the park is rather small.
55.
by cafebabbe 6y ago
Most adtech players are legit and don't rely on such practices, illegal in many countries. Of course some smaller one will still try to do it, don't get me wrong.
56.
by cafebabbe 6y ago
*Shortened phase 3 trials.
57.
by cafebabbe 6y ago
A good beer.
58.
by cafebabbe 6y ago
Right. The next emergency is clearly earth's magnetic field.
59.
by cafebabbe 6y ago
There isn't a way out to this mess that doesn't involve rethinking wealth redistribution.
60.
by cafebabbe 6y ago
Cool story, like a mini american dream! Some drivers have differing views on the topic though :(

More ›