sosodev

Acceptable? No, but some people realize too late that they want an alternative mode of existence. As somebody who had abusive parents, I would rather they had pursued their desires.

Isn’t the Waymo version used for some small portion of the dataset to cover edge cases that they have no real data for?

This blog is suggesting that we could train robots using only a world model.

> Do you have a basis for this idea, or it's a gut-feel?

My basis is that I’m a real human and these videos look far more like bad video games than the real life I’ve experienced.

Cool demos, but does anybody else find it troubling that companies keep claiming this tech can train robots? It’s so obviously not representative of the real world.

Yeah, what people want is games that aren’t slop. Defining that is probably impossible. We’ve always had shovelware, but the volume now is unsustainable and only going to get worse.

I’ve been agentic coding at home with like 15 TPS on average. I don’t mind it at all. I can give my agent a big task or a list of little ones and just… do something else. Why do I need to watch it work?

It’s something I’ve thought about a lot recently.

I decided not to build it because I think open, decentralized, and encrypted file sharing with good search is something of a Pandora’s box.

So I’m supposed to maintain a native app for each platform and try to be sure they’re aligned in every way except visuals? Sounds like a recipe for bad UX if anybody uses more than one of them.

But there’s still no way to get consistent results cross-platform without a web view, right? I guess Tauri isn’t even consistent because it relies on the system’s browser and they do differ to some degree.

In my experience, Qwen3.8-27b is so much more capable than its MoE predecessors. I find it odd that you think it produces the same results.

Yeah, I know we’re not supposed to say it, but HN is dying. The amount of slop that is getting praise on the site lately is insane.

Not ASCII and a lot of art looks off and has broken animations.

I’m not one to say that all AI art is slop, but there’s so little effort here. Why exactly are so many praising this?

Tracing definitely develops skills. I know many artists have traced their favorite works to get a sense of how it was put together. The trace doesn’t have any artistic value though.

Oh, I totally agree that it’s a valid reason. I just don’t find the authors gripe very meaningful. I say that as somebody who switched to Proton for political reasons too haha.

There are a lot of bad arguments in this. My biggest problem is that he wants to claim that he knows the truth (AIs do not have rights, feelings, or consciousness), but all of his arguments point to something else (we have no clue).

It's in the training data? Training it to say "I'm just a LLM, I have no feelings" is the same bias.

Anthropomorphization? Completely disregarding the possibility of consciousness is no better.

Consciousness is very likely biological? We only have evidence of biological life due to our circumstances, but observation is not the same as truth. Every belief can be invalidated. That's the foundation of science!

Maybe Proton’s copy needs some work, but another page on their website very clearly explains that emails aren’t encrypted when an email leave the ecosystem. Because, as the author points out, they can’t be.

Also, I hate Trump too, but a broken clock is right twice a day. It’s possible to agree with some of the things he says while not being a “Trump man”. Why should we be surprised that the company founded on anti big-tech principles would have a founder who is against big-tech?

I think too that the only way to escape AI-written code in software products is to have written every line yourself. Which is obviously impossible because you’re not going to bootstrap the operating system.