kadoban

It's not like this is a new problem. Turing had a definition most of a century ago, he wasn't the first and certainly wasn't the last. I don't think we need new terms necessarily, and I doubt we're all going to agree on a definition tomorrow.

Different quants, but I've been having success with: https://unsloth.ai/docs/models/qwen3.6 I do llama.cpp

It's an open question in mathematics/CS, that's all we know. If you want to know more, get to mathing :)

They make little ~rubber caps to cover the charging port when you're not using it. That, and make sure it's dry when you charge it to rule out galvanic nonsense.

Once you start getting shooty up in space, I think you fairly quickly get into https://en.wikipedia.org/wiki/Kessler_syndrome cascades of collisions and debris anyway, so it probably wouldn't matter.

in fact there is already no common international agreement about where a nation's boundary extends to in the z-axis. The observed limit is "however high that country can detect and shoot something down"

That's the thing though, if you use that definition then 99.9+% of them are fair game, because _only_ ones around equator at a specific height are at a fixed point. Otherwise they migrate a lot and what country they're over would depend on what second you look.

Coders like to code. It's in our nature. Even those who will get no benefit from it still _very_ often code their own version of things for various reasons, often bad reasons.

Figuring out how to prompt and test is educational, and it's potentially a useful test of a model/harness.

You're being sarcastic but you're not far wrong. Scent is a big part of attraction and sex. It's somewhat subtle, but important.

Artifically capturing that sounds...doomed though.

I think there's some setting to restrict the number of them, or maybe turn them off. Doesn't happen for me ~ever and it's not my $$ (work) so I haven't really looked at it much.

Not every possible thing has real situations where it's better. There are legitimately terrible solutions to problems, that are dominated by better ones in every situation.

Like this, for example.

It's possible, it would just be _very_ dumb. It'd also turn into just...shooting them all down, restrictions of who anything is over wouldn't survive.

It could only ever increase by the surface area of the satellite lens things. It'll be a rounding error compared to the surface area of Earth (or compared to the effect of greenhouse gasses over the time period you're setting up these things).

Is it ever, ever going to be cheaper or easier or more effective than just launching a missile? Seems...unlikely just on the face of it.

That would be fairly nonsense. There's _very_ few orbits that can stick over a particular place, and it's only places on the equator. We'd be giving up 99+% of useful orbits.

Thanks! Nice findings.

Is the output, that others can use, the research and findings, or is there a tool or something that came out of this that I can plug in to one of the common harnesses?

I guess that brief note in the twit is that it'll be part of that harness you're opening up in the future?

It could still be the UI for this, if they wanted to, and they just implement the guts behind it with that in mind. It's not everything, but the UI is still something you can ~skip.

Is it really false?

Even your own analogy for it is buying a lottery ticket.

Does buying a lottery ticket "lead to success"? Only if your picture of what that looks like focuses on only the winners.

Is that true? I don't know of really any medicine that has side effects 25 years down the line. Would we even know? We don't test new meds that long before release.

Isn't it more because meds are cheaper to test on animals and liability is much lower?

Thanks for the info, I'll track down that interview. That is an interesting strategy that I didn't consider.

It was an honest question.

I assume you're talking about "slow and expensive" like sending tens of thousand of troops to occupy at _least_ the ~entire shore around the narrow part of the Strait, indefinitely?

Yeah, shocking there's no political will for that.

Not OP but one thing you can do is boil or lightly-pan-fry (don't use any spices, or at least check that they're dog-safe, _do not_ use garlic) chicken thigh or chicken breast, cut it into small cubes and freeze it in sandwhich bags.

Dogs love it because it's just little meat cubes. Thaw it before you give them.

You can do the same thing with hotdogs.

I haven't done this model yet, but comfyui will definitely support it and I've found it a nice interface once you get used to it. Copy/paste a workflow to start with if you're lost.

AI is actually useful, in the strict sense. It can do things. If you consider its eventual effects good or bad for humanity is fair either way, but it's still different from NFTs or the "metaverse" nonsense. It's a categorically different thing.

If Iran was in more pain than the USA, they wouldn't be getting $300 billion, sanctions relief and ongoing fees for the Strait. That's how wars work.

They're getting favorable terms because the USA can't win this war as it was started.

Stupid question: does the computer or whatever the monitor is hooked up to need to know to do something special to then show those colors, or it's just normal rgb color levels and in a less-good-color-space monitor those would have been shifted to less accurate colors?

What does sorting have to do with violating p != np?

The common bound on sorting is you can't do better than O(n lg n) worst-case, but that is strictly only for comparison sorts anyway.

Really? Why not. From the benchmarks at least it's a pretty decent small model.

I mean, Qwen 3.6 kicks ass. I don't know who these people are, but if their first outing is "not quite as good as Qwen 3.6", that's not a bad start by any means.

30B vs 35B isn't nothing either.

If it ends up just being some tweaks to someone else's weights, then meh.

What tool do you use to drive things for you, out of curiosity?