🤝 I download new music to check if I like it, and if I do then go over to Qobuz and buy it DRM-free and download the flac file, which is my flac file. Happy to support the musicians.

+1! 🎶🚀

And Qobuz does as well.

https://github.com/BryanLunduke/DoesItAgeVerify

And what happens with it also depends on how people, including software developers, react to it.

😂 🤝

A reminder also to boycott, as much as possible, those thirteen major publishers – most or all of which are stealing from academia:

APRESS MEDIA, LLC; CENGAGE LEARNING, INC.; ELSEVIER INC.; HACHETTE BOOK GROUP, INC.; HARPERCOLLINS PUBLISHERS LLC; JOHN WILEY & SONS, INC.; MCGRAW HILL LLC; BEDFORD, FREEMAN & WORTH PUBLISHING GROUP, LLC D/B/A MACMILLAN LEARNING; MACMILLAN PUBLISHING GROUP, LLC; PEARSON EDUCATION, INC.; PENGUIN RANDOM HOUSE LLC; SIMON AND SCHUSTER, LLC; AND TAYLOR & FRANCIS GROUP LLC

How some Linux developers defeated (for now) the new OS age-verification laws. Long live those Linux developers, who “heavily criticized the mandates”, made public statements, and contacted the legislators.

Because other Linux developers, instead, immediately bent over backwards to start implementing changes towards accommodating those laws; for sure they didn’t heavily criticize the mandates, nor make public statements, nor contact the legislators.

Funny, seeing as EU itself is a loophole in democracy.

It was exactly as you said; a difference I didn’t know about. Also confirming that Kubuntu apparently installs them system-side, even if flatpak install ... is called without sudo, again as you inferred. I don’t know how I managed to install them user-side in one laptop, but now they mirror each other :)

For others interested, these two commands show the difference, as explained by another user in a cross-post:

flatpak --user list
flatpak --system list

Thank you!

It was as you said! Thank you so much for explaining the difference and showing the commands 🙏 I wonder how I managed to install user-side on one laptop, since the standard flatpak install ... (without sudo) seems to install system-side anyway. But now I’ve managed to have an identical setup :)

It goes deeper than that, though. Why is the person talking about this with a chatbot in the first place, rather than with some professional?

It shouldn’t be a chatbot what prevents suicide in the first place. Something has gone horribly wrong with society – and it has already been normalized too.

Thanks for sharing. I’d be happy if you posted updates on other stuff that works or that gives you problems after the OS change.

It is actually not so difficult to see this for yourself in a much simplified setting. One can easily build a “Small Language Model” that extracts correlations between only three consecutive words. On the web there’s plenty of short scripts that do this; here and here is one example. The output created by such a SLM can have remarkably long sentences with grammatical meaning (see the examples in the links above); this is remarkable since all it learned was correlations between triplets of words.

Now you can take a large amount of output from such a SLM, and use it to train a second, identical or even better SLM, then check the output generated by this second one. You’ll see that the new output is less coherent than the one from the first SLM. Give the output of the second SLM to a third, and you’ll see even less coherent text coming out. And so on.

They aren’t out of context, and you have just said the same thing. Data processing can help in removing noise, but it can’t help in creating information or extracting information that wasn’t there in the first place. In fact – again as you said – it can end up destroying part of the original information.

LLMs extract word correlations from textual data. Already in this process they are losing information, since they can’t extract correlations beyond a certain (yet large) length, and don’t extract correlations at shorter lengths. And in creating output they insert spurious correlations that replace (destroy) some of the original ones. This output will contain even less information than the original training data. So a new LLM trained with such an output will give back even less.

Yes it does. Indeed it is a mathematical theorem from Information Theory, called the data-processing inequality. Quoting from two good textbooks on Information Theory:

“No clever manipulation of the data can improve the inferences that can be made from the data” (Cover & Thomas, Elements of Information Theory §2.8).

“Data processing can only destroy information” (MacKay, Information Theory, Inference, and Learning Algorithms exercise 8.9).

You’ve read the stances of all different people. I agree with most and I’m a bit more conservative: I switch to a LTS (even-numbered) release only when its main non-LTS (odd-numbered) upgrade is out; and skip all non-LTS.

I agree. I’ll actually contact the national Consumer Policies department and ask if this is at all legal.

The fundamental problem is that age verification is bullshit. So let’s not normalize it. It must be fought, on all fronts, including the FOSS front.