What makes these consumer-oriented models different is that that rather than being trained on raw data, they are trained on synthetic data from pre-existing models. That’s what the “Qwen” or “Llama” parts mean in the name. The 7B model is trained on synthetic data produced by Qwen, so it is effectively a compressed version of Qen. However, neither Qwen nor Llama can “reason,” they do not have an internal monologue.

You got that backwards. They’re other models - qwen or llama - fine-tuned on synthetic data generated by Deepseek-R1. Specifically, reasoning data, so that they can learn some of its reasoning ability.

But the base model - and so the base capability there - is that of the corresponding qwen or llama model. Calling them “Deepseek-R1-something” doesn’t change what they fundamentally are, it’s just marketing.

There are already other providers like Deepinfra offering DeepSeek. So while the the average person (like me) couldn’t run it themselves, they do have alternative options.

A server grade CPU with a lot of RAM and memory bandwidth would work reasonable well, and cost “only” ~$10k rather than 100k+…

To be fair, most people can’t actually self-host Deepseek, but there already are other providers offering API access to it.

It’s almost sure to be the case, but nobody has managed to prove it yet.

Simply being infinite and non-repeating doesn’t guarantee that all finite sequences will appear. For example, you could have an infinite non-repeating number that doesn’t have any 9s in it. But, as far as numbers go, exceptions like that are very rare, and in almost all (infinite, non-repeating) numbers you’ll have all finite sequences appearing.

I’m confused, isn’t Fedora atomic immutable? Shouldn’t that make it stateless automatically?

Well, he didn’t even buy the original (I guess it has spoiled by then), but a DIY replica and a certificate.

Wary reader, learn from my cautionary tale

I’m not sure what to learn exactly. I don’t get what went wrong or why, just that the files hit deleted somehow…

I guess technically that makes them “not in Ukraine”, but it is the same war in the end. At least for me that’s the important part, not where exactly on the front line they are.

Well, NK and Russia have a defense treaty which obliges NK to sent military assistance to Kursk. So if they aren’t, they’re breaking their obligations.

I’m sure removing these maintainers would be of great help to the Ukrainian war effort…

More seriously: We need to help Ukraine more. But this doesn’t do that. It just hurts a bunch of people (both the maintainers, and the people using their code) for no benefit whatsoever.

My bet is, it’ll be Saturday that goes, finally achieving a 6-day work week.

To add about the distro framgentation, and particularly:

If I run into a software I need and it specifically indicates it’s for another flavor of Linux than the one I run, how likely is it that I can get it to work on another distro without any real trouble?

You might have. Some software is distributed as a portable binary and can run on any distro. However, many installers are distro-specific (or distro family-specific, since they’re made for a specific package manager). For example, a software packaged for Ubuntu as a .deb file would install fine on Ubuntu or Mint, and probably install fine on Debian, but if you want to install it on Fedora or Arch you’ll have to manually re-package it.

Most distro-specific software usually ships debian or ubuntu package - so you might go with that for that reason. Or Arch/Endeavor: while you’ll rarely see an official Arch package, most often someone will have already re-packaged it and put it on the AUR.

That said, for the major distros, the desktop environment makes much more difference than the distro.

I’m not sure where the Linux kernel part comes from, but if I open the article and search for “linux” or “kernel”, there are no matches…

Technically, “enforced pay it forward” is called credit. Your debt would then be “the amount you still have to pay forward”.

Of course, this defeats both the spirit and the purpose of a pay it forward scheme.

I don’t know - but I’m willing to get those instances where people were saved weren’t calls from anonymous voip numbers.

“Just works” is not a mentality imposed by Microsoft, and has nothing to do with loss of control. It’s simply (a consequence of) the idea that things which can be automated, should be. It is about good defaults, not lack of options.

It’s certainly good, I’m not arguing that. My point is, if the wine team is interested, they can fork the unmaintained project, and work on that. Eventually, people will switch over to the active fork. What Microsoft is doing, is helping the process along, and making it easier. So it’s good, and helpful - but not really a “donation” to winehq.

I guess it’s simply the framing: It was a not very actively maintained open source project. So they’ve decided to turn it over to a new maintainer. Calling that ‘donation’ is a bit pushing it

I’m confused - why is Microsoft trying to - or expected to, by the article authors - patch a vulnerability in GRUB?