HN user

fc417fc802

4,830 karma
Posts0
Comments4,007
View on HN
No posts found.

Yes the fact that the scientific literature - created largely on the back of the tax payer - isn't open to all free of charge by force of law is a travesty. A cartel should not get to charge for access to the bulk of human knowledge. That is indeed a far more important issue than whether or not Moonshot violated the Anthropic ToS, possibly committing mass fraud in the course of doing so.

I mean honestly if they did that why should I care? I'm happy to see copyright violated in a manner that leads to the creation of new technology. IP law exists strictly for the benefit of society and by all appearances AI is an incredibly powerful tool.

Also while I'm at it libgen is a gift to humanity. Information wants to be free. Spreading and preserving knowledge is generally one of the most wholesome activities anyone can undertake as far as I'm concerned.

> Complex societies exist and work.

They certainly exist. Whether they work is rather the question.

I'd take the computing device and global network you used to post that - both of which require at minimum continent spanning efforts for both R&D and manufacturing - as a decisive answer.

You make many interesting points but I think such hyperbole detracts significantly. Modern society might not represent the optimum but it clearly works extremely well.

Qwen 3.8 10 hours ago

Sure but at that point effectively you're taking about stuff related to national security which is essentially an edge case and inevitably tightly regulated with all sorts of hoops to jump through.

It very much is the case. Your own quote contradicts you. A smart tv is definitely a computer (one of the selling points is literally that it runs arbitrary apps) and LG is declaring that this "is not an intended use for LG smart TVs" and banning it.

There is certainly a conversation to be had regarding consumer protection laws, the nature of an appliance, and the trade-offs thereof but this is not that conversation and I don't think you engaged in good faith with the comment you replied to.

Obviously. But that is missing the point of the conversation. It was suggested that open models offer a security advantage which it seems is not the case.

While using floating point? Not happening. You'd have to switch to fixed point, not just for the models themselves but also _all_ the training code (ie backprop).

Even then you'd still need to account for order of events when an entire cluster of GPUs is involved. Also don't forget to account for any synthetic data sources. Or even non-synthetic for that matter - does your pipeline do any image resizing on the fly? Better make sure that's fully deterministic between machines (it almost certainly won't be).

It's theoretically possible but I don't expect it to materialize any time soon.

That's understating things. The efforts to violate the ToS involve what amounts to large scale organized fraud. But I'm not convinced that should be considered an "attack" rather than merely "piracy" and I certainly don't feel like there's any ethical issue. It's nothing more than an attempt to politicize competition.

Qwen 3.8 1 day ago

it does show that there are other strategies that might occur to China that don't reflect the knee-jerk reflex to restriction and control that Anglosphere governments have around tech.

I'm not sure why you think that is necessarily reflexive nor why you think China differs. The current cost benefit is simply different for them than it is for the US. I expect the US might do the same if the tables were turned.

Disregarding historical examples, this is the country that implemented the "great firewall". That alone leaves me perplexed at your position here. Clearly they are happy to restrict and control when it suits them.

Qwen 3.8 2 days ago

This doesn't pass the sniff test.

Consider that the US was intimately involved with the start of the FOSS movement. Compilers and commercial operating systems are hugely valuable.

Also consider that history the world over is littered with counterexamples. Historical China itself is rife with examples of technological secrecy and exclusivity.

No competent government will voluntarily release (or allow to be released by its subjects) technology that would give them a meaningful advantage on the world stage.

Qwen 3.8 2 days ago

the "open" part of this and K3, etc. is of limited use depending on how/where you can actually use it yourself.

... from servers owned and operated in the US by US providers? Go check out the list of providers on ex openrouter for ex the latest deepseek model. Some of them even fall under the zero data retention agreement.

I don't think that explains it. CS has been an extremely popular degree for at least the past 20 years. Not to mention coding bootcamps and the like.

The obvious notable event circa 2024 was LLMs becoming reasonably useful for code assistance (IMO and IIRC OFC).

It doesn't matter. Assume it's hyperbolic and the point still stands.

Of course all the other models from everywhere else in the world will also refuse to engage on various topics so the point is hardly limited to china.

Why would anybody think it's meant to apply outside of the EU?

Because it's clearly worded in such a manner, similar to US financial laws. The key difference is that the EU so far lacks the leverage to throw its weight around outside its own territory to the extent that the US does. (Also presumably politicians won't be willing to burn bridges over PII handling violations to the same extent that they do over financial crimes.)

.. which is very different from ...

And yet still an attempt at extraterritorial overreach. Regardless, I imagine that the rest of us who don't do business in the EU will continue to disregard its very existence. (Except in principle when we write negative comments about it on the internet that don't in practice matter whatsoever, such as this one that you're reading right now.)

Not at all. "Sure it finally mostly works at this point. Mostly." I explicitly acknowledged that after 10+ years of needless drama something like 95% (handwave) of the extended feature list is working.

And so much of it really was needless drama. Go read through many of the wayland protocol extension issue tracker threads. The amount of intentional heel dragging, willful ignorance, and purity spiraling is off the charts (IMO obviously).

To be clear I like and use wayland in its current state. There are plenty of valid criticisms of X11. IMO a redo was not a bad idea per se and it eventually turned out well enough (provided your app doesn't depend on the 5% that suffers an arbitrary rejection).

It's also an 80% solution, the same thing so many rust "rewrites" are notorious for. Sure it finally mostly works at this point. Mostly. But it took a decade of kicking and screaming to get here and there are still edge cases that don't work, many clearly viewed as wontfix simply because there isn't enough popular support to force the hand of the purists. Hopefully llms make a community bootleg wayland protocol effort viable.

We should just go with numbers in general in order to play it safe. The new guidance from legal is absolutely no numbers on invoices for liability reasons. Similar to the removal of actual math from math class we're going to let the experts figure out how to implement it.