the megacorps stand wayyy too much to lose here and would fight tooth and nail to prevent anything like this.
- The megacorps could just bribe their way out of it like they do every other form of legal scrutiny.
So, this is the place where I’m going to be generally hanging out and trolling around, while my Pixelfed, Mastodon, and Blacksky accounts are going to be primarily art posting accounts, and I also have a Nooki account.
I’m also going to start to be active more on here than on lemmy.org, so I’m making this my primary Lemmy account now.
I’ll link the other socials I’m varying levels of active on below, plus my lemmy.org account which I’m demoting to my secondary account if that instance is going to be more unstable from now on.
the megacorps stand wayyy too much to lose here and would fight tooth and nail to prevent anything like this.


deleted by creator


deleted by creator
I didn’t because I just wasn’t interested at the time, and now with how far universities in my country have fallen in the last 14 years, I feel pretty vindicated in that decision.


Or even Mad Max for that matter, which is the more likely scenario to happen IRL the way things are going currently; Mad Max’s apocalypse is caused by societal collapse due to people fighting over resources, guess what could happen IRL with AI datacenters using up most of the resources right now?


Code generated by a large language model or similar technology, such as GitHub/Microsoft’s Copilot, OpenAI’s ChatGPT, or Facebook/Meta’s Code Llama, is presumed to be tainted code, and must not be committed without prior written approval by core.
While it is not prohibited to use AI as a learning aid or a development tool (i.e. code completions), extension developers should be able to justify and explain the code they submit, within reason.
Submissions with large amounts of unnecessary code, inconsistent code style, imaginary API usage, comments serving as LLM prompts, or other indications of AI-generated output will be rejected.
Although I prefer NetBSD’s stricter ban on AI code over GNOME’s, it’s better than nothing on GNOME’s side.
Also, I hand-drew my pfp unlike one other commenter in here, and the original is pinned on my IRL wall right now, so that should hint at how I personally feel about AI.
Update:
As there seems to have been recent confusion about [AI integration into Firefox], just a quick “official” toot to then pin: we haven’t and won’t support “generative AI” related stuff in LibreWolf. If you see some features like that (like Perplexity search recently, or the link preview feature now) it is solely because it “slipped through”. As soon as we become aware of something like this / it gets reported to us, we will remove/disable it ASAP.
Waterfox also made an official stand against AI a bit after LibreWolf did.


I came over to Fooyin from Foobar2k, so the UI translated right over.


Just like I said in another post related to this, I hope this doesn’t kill LibreWolf, IceCat, and Waterfox.


Fooyin’s great so far.
Honestly, from curiosity and messing around with stuff, playing with Crunchbang on an old Win9x PC. (this was eons ago as Crunchbang wasn’t BunsenLabs yet at the time)
Yes, really, the last time I actively ran Windows for any reasonable length of time was with Win9x, specifically 98se.
I messed around with Win10 LTSB for a bit on a laptop (this was in 2016, so when Win10 was still new and LTSC was still called LTSB), but eventually went back to running Linux, and given Windows’ current trash-fire state, I’m not touching it on my hardware outside of a VM ideally, or a dedicated burner box if a baremetal install is ever needed for anything.


You could still stick an NVMe drive on an older system as a secondary drive, eg. as a /home drive if you’re running Linux on it, by sticking it on a riser card, although you’d still need to boot off a SATA drive, and you’d take up one of your expansion slots doing that.


Phoenix2 APUs like the R3 8300G and R5 8500G are the worst offenders in the ‘cutting PCIe lanes’ department.
The R5 8500G only has 14 lanes, for example. The FX-8350 and 8370 from a decade earlier, would’ve had 32 lanes available on the 990FX chipset, and half that on the 990X and 970 chipsets per contemporary reviews from when those CPUs were new, but they were all PCIe 2 as AM3+ was a PCIe 2 platform.
This is the specific review I’m going off of for this. FX-8350 review
Per that review, 990FX would’ve supported 2 x16 or 4 x8 slots, while 990X would’ve supported 2 x8 slots, and 970 would’ve only supported a single x16 slot, but of course configs varied by the board makers, and there would’ve been nothing stopping someone from making a 990FX board with a single x16 slot, three x4 slots, and two x2 slots, for example, nor a 990X board with a single x16 slot or a 970 board with a single x8 slot and two x4 slots.


both Intel and AMD have historically been pretty bad about being stingy about PCIe lane availability


I’m just speculating on what could happen if this stuff gets worse.


Eventual discontinuation of more PC parts to appease the AI grifters until all that’s left for consumers is mini PCs or ARM black boxes.


The supply shock is going to be as bad as COVID.


Next step, modular desktops as a concept will die, probably.
I hope people like locked-down black boxes they can’t upgrade and can’t run their own OS on in the future, so byebye Linux and BSD in that scenario outside of niche devices.


So even the enthusiasts would rather, or are probably going to be forced to, give up their custom-built PCs, which they can do whatever they want with and run whatever OS and software they want on, for locked-down black boxes they don’t even own anymore, even possibly to the degree where they can’t even install their own OS anymore, so no more Linux or BSD outside of some niche hardware? That sucks.
Wouldn’t that efficiency thing largely depend on what silicon you’re using though? Like, for example, if you have an old AM1 board sitting around, those are 25W APUs and shouldn’t use anywhere near that if used as the base for a router.
Meanwhile if you’re using, say, an AM3+ board and an FX-4300 for your router, the FX-4300 is a 95W part and more likely to cost more to run by contrast to, say, the Athlon 5350’s 25W on AM1.