BL4CKP1XX13

The people behind Asahi are so painfully underappreciated, not only is their work some of the hardest across the entire Linux space, but what they’re fighting for is basically the last remaining escape out of Apple’s walled garden.

Yes, unfortunately they are, and I do think efficiency is going to be a significant research front for open-weight models. The nature of this topic is highly speculative as our compute capabilities have only recently reached what is required to consider running generative AI models, what we have today are very crude first implementations of what I personally believe will become an everyday tool for developers, and more.

And we have seen this, there are models now capable of running on an individual’s hardware (and not particularly expensive hardware either) that can outperform what ChatGPT initially launched with.

Unfortunately not yet, no, true FOSS models are likely many years away, but I would argue that that follows typical FOSS lifecycles. Emerging technology is typically outperformed by proprietary endeavors, which creates an audience, and then that audience undertakes a FOSS implementation that initially underperforms, then just about competes, then eventually overtakes (i.e, GNU/Linux).

If the model is adequately FOSS, i.e. open weights, and can run on a single consumer GPU (or NPU), and the “author” (quotations because I’m personally undecided if one can claim code generated by an aforementioned model is theirs) understands it, then I really don’t see what the problem is…

Personally I have nothing against SLMs/LLMs as a technology, to me my grievances against ChatGPT or Claude are mostly about their environmental impacts and selling us back our own art, also keeping knowledge behind a for-profit black-box - if those aren’t appropriate for a specific model, then I say using that model is fair, and good for productivity.

I am not unfamiliar with Windows in any sense thank you. Windows is objectively bad, there is no good reason that one must maintain their own custom image to install an operating system to a usable capacity because the vendor image is so disgustingly bloated that getting any real-world use is a chore.

And it’s not even recent Windows enshittification that is the problem, it is the foundational architecture of the NT system. With a Linux operating system, you install, and you’re done, on Windows, you install, then you need to go out and do 15+ updates that for some reason aren’t done during install, then you need to go out and get all your drivers for your hardware. It’s poorly designed, slow, clunky, and, especially considering that Windows doesn’t have a package manager and a curated software repository (Winget + MS Store does NOT count), unbelievably insecure.

My organization has pretty much decided that Windows is on the way out, it’s just going to be a very long time for us to transfer to Linux-based operating systems. Even after getting Windows to a just-about acceptable standard, it’s still far heavier than any Linux-based OS would be, which for our thin clients is genuinely proving a problem as many of our staff have high memory usage workflows and are frequently seeing programs crash due to memory limitations.

Unless you are severely, clinically deluded, there is no way you can genuinely state that what you hear in Linux communities is “batshit”; everyone knows Windows is dying, even Microsoft…

Bootloader? I hardly know her!

Running Windows programs in WINE bo’les innit.

I work in IT and half the time spent deploying a new device is getting around Windows’ bullshit.

Fine, you know what, corporations gonna corporate, yes you need our companion app so we can datafarm your phone and continue to extract wealth from you after purchase.

Okay, but I ask WHY does this not have offline local storage to keep operating when offline?

The Spiral? Is that a Magnus Archives reference?

doesn’t even matter he’ll sleep through the whole thing

Damn right it’s a national security concern, let the US keep it’s spyware to itself.

Paint.

Orange to be precise.

When I was a kid, a friend of mine hosted a birthday party on a paintball range. We played for I don’t know how long, large group of us, was really fun.

One thing the instructors told us, keep your mouth shut.

You see, we wore face masks, however for us to breathe of course there’s a grill where you mouth goes.

Needless to say, I got shot right on that mouth grill, and naturally due to the shock of it I opened my mouth…

The paintball burst on the grill, and I had the whole volume of paint launched deep into my throat.

It was so… ungodly… salty. For the rest of the day all I could taste was salt (the blow occured around early afternoon). And that’s coming from someone who also on another unrelated occassion ended up with a deep mouthful of sea water.

Trust me my code aint worthy of being “art”.

If yours is, I totally get where you’re coming from, AI-generated code is pretty bad right now, albeit it does often work.

The way I often do things is have an agent create a rough first implementation according to my own architecture, so I have done all that high-level thinking that it currently struggles with; then I have a dedicated improvement agent come and clear that up substantially, and then I review whatever is left.

Grow up.

It does exist, but its small and separate from the “conglomonet”, if you will.