FooBarWidget

That objection is completely orthogonal to my point. The people who want open weights models are not so concerned about mass data ingestion. You are, but I guess you also don't want any good models to exist (to be clear, which I think is a valid opinion, even if it's one I don't fully agree with) since nobody has figured out how to make a good AI model without trampling on other people's feelings regarding data usage, or violating licenses or copyright.

To the people who demanded completely open source licensed software, or that complained about "they shove Wayland/systemd down our throats", I've said many times that this is a supply/demand problem. Open source licensed software exists because there is a lot of excess software production capacity, e.g. there are many volunteers. Non-systemd distros are rare because such production capacity is low, e.g., there are few volunteers willing to do this work. To those people I've said: stop complaining about others not doing it for you, if you so want your non-systemd distro or FOSS-licensed whatever, you gotta do it yourself and contribute to the community, or fund an effort.

I got super angry reactions.

Now, the same thing is happening with open weights models, but in on a much more extreme scale. They exist because of excess capacity for whatever reason, and they will cease to exist when this excess capacity reduces. There is no use complaining about something not being open weights. What is productive is funding an effort.

Unfortunately, most complainers want others to do things for them, for free.

Or the answer is even easier. There is public data consisting of chat transcripts with ChatGPT, Claude and DeepSeek, and all AI providers train on that data. You don't need distillation for the text "I am ChaGPT/Claude/DeepSeek" to often appear in your trraining data.

I don't get it even after reading the guide. How do you deal with backups when using YubiKeys? The private key is stored on the YubiKey with no ability to export it, ok... then what do you do when you lose the YubiKey?

Are you supposed to have multiple YubiKeys (multiple private keys) and then ensure that all your systems accept the matching multiple pubkeys?

The guide's Backup section is... super vague. It says exporting keys, of which some are private keys, which I thought shouldn't be possible?? And then it goes off about partitioning and filesystems?

What is "wrong" is that they have a culture of being conversative when it comes to 0-1 innovation. Everything is set up to reward concrete results. With 0-1 innovation you have no idea whether it's viable, so few people have the risk appetite and would rather pursue proven venues.

But westerners also underestimate the importance of 1-10 innovation. Scaling manufacturing and iterating on existing concepts are not at all trivial, and are also very valuable.

0-1 and 1-10 are complementary. Unfortunately, many people deliberately turn this into a pissing contest by accusing the Chinese of "only being able to steal" while ignoring how their own country is unable to produce basic goods like masks on a large a cheap scale, and can't fix a brige for less than xx million.

It doesn't, at all. Unlike the US (and to a lesser extend, Europe), Chinese do not have a majority of their wealth invested in stocks.

Besides, the Chinese st9ck market has been atrocious since.... pretty much forever. You'd think they still live in mud pools if you think the stock market is representative of overall development.

Have you looked at China from 15 years ago? It was also said to be in a depression and youth unemployment was also high. Just look at articles from back then.

What people don't seem to understand is that China is developing in stages. Every 10 years there is a new stage, a new revolution, with different winners and losers. Maxxing out on the previous growth engine was already forseen 10 years ago. That's why they bit the bullet and decided to pay the price to transition to the next growth engine. All transitions are painful.

Besides, these are Chinese we're talking about. They'd even complain about the economy being bad even when it's growing 10% and they drive away in 2 luxury cars. When they say they're "barely surviving" they mean their savings account isn't growing.

When you look at the bad economy in China, are you looking at the costs, or are you looking at the development goals? Development goals have costs, not paying those costs to reach goals has even bigger costs pater on. The Made in China 2025 goals have already been largely achieved.

Deepseek never subsidized their plans. At some point they dropped pricing, then after a while they raised them, but that's because they had capacity issues, not because of dropping subsidies. They don't want your money, they'd rather have fewer users.

OpenCode is building their own inference and they've independently stated that DeepSeek's old (cheaper) pricing is achievable without subsidies.

Ok fine. But again: any repeatable way to prove your thesis or disprove the trashing-code-pages thesis? And how to do this when the system is freezing up?

So you're saying processes are stuck for minutes on end at trying to get memory from the kernel? I find this implausible because relatively simple processes such as ssh and bash shouldn't be getting memory from the kernel at a continuous basis. Unless they're suddenly processing a lot of new data (like me trying to cat /dev/urandom), they should be spending most of the time working with the memory they already have. Especially when all I'm doing is pressing keystrokes in ssh+bash, not even hitting enter. Note that malloc doesn't ask memory from the kernel at every call. Glibc malloc is (in)famously stubborn with caching memory it has gotten from the kernel, hesitant to release it back to the kernel.

Any repeatable way to prove your thesis or disprove the trashing-code-pages thesis?

Then why are AI plans still so super expensive, and AI spending going through the roof, while all the subsidies are ending?

It's the explanation that I've read when scouring forums for reasons why it happens. I'd love to hear alternative explanations and how to measure/verify them. But the measurement tools also become unresponsive so I can't check anything.

But frankly, as a user I don't really care what the real technical reason is. I just don't think the system should become unresponsive for minutes on end during memory pressure. Just OOM-kill earlier, when it's supposed to.

> The active code pages are specifically excluded from being discarded during the reclaim process.

This is not true, or the system wouldn't grind to a halt due to having to reload code pages (even those that are recently in use) from disk over and over.

For example, key presses and output streaming in SSH become unresponsive. The code pages for handling key presses and output streaming are obviously in use — I just used them!

I wish this is true but evidence is against you. Just try testing memory pressure on a no swap system. You will see that the system grinds to a half for a few minutes before you even get to an OOM event.

While I've seen systems grinding to a halt due to evicting code pages, the idea that disabling swap distributes load between code pages and data pages makes no sense to me. Data pages are also disk backed, so they could also be evicted at any time. Why would data pages be written to swap if they came fron the disk in the first place?

So that only leaves anonymous data pages, i.e., regular data in memory, that could be swapped. Ok but you're not easing the load on code pages, those still get evicted during memory pressure whether you have swap or not.