I generally agree with your stance about the situation in eastern Europe, but there's no particular reason to connect that to this incident. There's no public evidence that this was state-sponsored or even connected to Russia at all. They definitely provide sponsorship and cover to a lot of attacks, but still are not the only source of online crime. We have plenty in the US and the UK and everywhere else.

He was not a good person. He was one of the great writers of this age. Two things can both be true.

Sure but do we need idolize a pedophile in the pages of Nature? I think not.

You cannot have peace without justice.

Justice is in the eye of the beholder. There has to come a time of acceptance

This is a terrible article about what sounds like a legitimate problem. Even in the section, "What is Chat Control?", the answer to the question is buried in the middle of the seventh paragraph.

If the writer of this post wants people to oppose it, they really should do a better job of explaining at the very top what "it" is.

I think you're being hyperbolic. It wasn't terrible, but I do agree, I had to dig for "What is Chat Control". It read to me like a panicked person, repeatedly saying, "You've gotta hear this..." over and over, before getting to the point.

The question makes as much sense to me as "why teach literature in the age of typewriters?" Not that the analogies are perfect, but the idea that it's not worth learning something because a related technology has advanced significantly is a non sequitur.

There may be good reasons to learn or not to learn calculus, or literary theory, or anything else, but the existence of some related technology isn't it. I'd go so far as to suggest that perhaps calculus is even more important for some folks to learn in the age of AI (e.g. applications in neural networks), and we don't know who those folks will be in advance.

I got all excited thinking this was related to real-world archaeology and somehow providing a git-like representation of that. (No idea how it would work, which is why I was excited).

Should Archeologist rebase or merge when new evidence is undercover?

These look like fun technology, but I don’t even know what the use cases would be anymore. I don’t need something to control my lights: I have an actual light switch for that. I don’t have endless terabytes of media that I need to serve within my household, so I just don’t know anymore what I would use this stuff for. Twenty or thirty years ago, I loved having a home lab, but these days I’m just not sure what I would do with it.

It's definitely not something for everybody, and I suspect not really a "revolution" outwith the dedicated enthusiasts segment of the population. (-:

If one wasn't likely to have a rack of desktop PCs at home, one probably will be hard pressed to have a reason for a collection of mini PCs.

On the other hand, there are a few more use cases than just over-egging a light switch or being a big file server. One can replaced dedicated hardware such as network gateways/routers, and do self-hosting, for examples.

There are RaspberryPi systems that one can get with extra Ethernet ports that can be set up to do application gatewaying and routing with general-purpose operating systems like FreeBSD, NetBSD, and such. And obviously filtering HTTP/other proxies incorporating spam/advertising/malware blockers are a well-known use case.

There are oddball mini PCs in some parts of the world with loads of serial ports, useful as terminal hosts if one has a lot of systems with no disploy/HIDs. (I saw one mentioned on the FediVerse the other day. It turned out that it was an old Russian point-of-sale system, with 6 serial ports.) More of a use case for someone who already has lots of PCs (with serial ports), of course.

But yes, it's not going to be everyone's cup of tea. Not everyone is Kitboga with a server farm in xyr garage running speech synthesis engines and language models to call scammers. (-:

On the gripping hand, I swapped out someone's under-the-desk tower for a mini PC years ago simply for space reasons.

> what I would do with it

I set up one to run Frigate [1] to detect motion over my several security cameras and send me notification emails with still images and video clips attached. It works well, and I hate the idea of sending my private videos to the cloud for processing with usual security camera setups.

[1] https://github.com/blakeblackshear/frigate

That's not much of a disability-related use case to insult me about.

You can't imagine a disabled person might have difficulties to get up and reach a light switch?

Absolutely, but you suggested that it was because you hate to turn off music and then walk out.

I did no such thing; that was a non-disabled use case just to show you it has use for everyone.

I swapped from Linux to MacOS when the M1s came out, and I love the integration with all the iCloud stuff (particularly Messages). Occasionally I miss being on Linux, as somebody who did so for 20+ years before making the switch. But on Mac, stuff actually does Just Work.

Reading this makes me a little misty-eyed and I miss my solid old Thinkpads from 10-20 years back.

The point is that you call the LLM to generate the code that lets you talk to the API, rather than writing that glue code yourself. Not that you call the LLM to talk to that API every time.

Exactly.

The article talks about that at the end, then says:

> Let models talk to each other directly, making their own case and refining each others’ answers. Exemplified in patterns like Multi-Agent Debate, this is a great solution for really critical individual actions. But XBOW is basically conducting a search, and it doesn’t need a committee to decide for each stone it turns over whether there might not be a better one.

In general, this seems reasonable to me as a good approximation of what works with humans, but with _much_ faster feedback loops in communication.