Yeah, that did nothing to sell me on the concept, but if there's one great thing about open source it is that there's no single right way of doing things. If this is what you want for your system(s), you get to do you.
And those of us who run on cobbled together systems composed of other people's e-waste get to not waste massive amounts of system resources we don't have on abstractions ten layers deep. I can live with the potential for occasional transient system instability if it means I can run my base desktop system without involvement of flatpak, snap or - Jesus Christ - docker.
KDE Linux is an "immutable base" operating system that does not include a traditional package manager. Apps can be installed from Flatpak, Snap, or AppImages.
Oh, wow. Okay. I can see why some people might want something like that, but I'm empathically not one of them.
"Look, we'd like for you to come in for a private meeting alone in a room with a bunch of old white men in power. There's no way you'd be averse to something like that, right?"
It sure would be terrible if this atavistic knuckle dragging neanderthal took another swim in raw sewerage and accepted its loving murky embrace forever.
Just terrible. I mean, the waterways are polluted enough as it is.
No, that's a good point. We all bloody well know there isn't a single provider of LLM's that aren't sucking the entire Internet dry while gleefully ignoring robots.txt and expecting everybody else to pay the bill on their behalf, but the AI providers are getting really good at using other people IPs both to mask their identity and to evade blacklists, which is yet another abusive behavior.
But that's beside your point. So forget the class-action lawsuit in favor of the relevant Ombudsman.
Either way, this cannot go on. Donation-driven open source projects are being driven into the ground by exploding bandwidth and hosting costs, people are being forced to deploy tools like Anubis that eats additional resources - including the resources of every legitimate user. The cumulative damage this is doing is no joke.
Nobody said they replicated by authoring the replica from scratch, which seems to be what you're assuming. A generative AI is ultimately a lump of code and a statistical model. Surely you're not saying that it cannot copy files given file system access.
Because copying some files and starting new processes is all it really has to do to 'self replicate'.
Yeah, that did nothing to sell me on the concept, but if there's one great thing about open source it is that there's no single right way of doing things. If this is what you want for your system(s), you get to do you.
And those of us who run on cobbled together systems composed of other people's e-waste get to not waste massive amounts of system resources we don't have on abstractions ten layers deep. I can live with the potential for occasional transient system instability if it means I can run my base desktop system without involvement of flatpak, snap or - Jesus Christ - docker.