without checking, Gates’ wealth is probably tied up in a lot of MS stock, and he could probably walk into the office and ask the intern to get him a coffee. but yeah i think mostly retired.
Linus is still active is maintaining the Linux kernel.
and yes, this is fluff, not some kind of summit
ah that makes sense
eh Bill doesn’t really pull any levers at MS these days, and Linus already got the moral victory of the Linux kernel running on Windows and basically powering their entire cloud business.
fuckin weird that an extension would inject invalid JSON into an API payload. if you’re gonna make a shady plugin at least test it lol
anyway, if that’s truly the issue i’d be worried about what my extensions were doing, personally.
i would start by seeing what the actually API response is. i haven’t used OpenWebUI, but to me this looks like some kind of error response from the server. you could use an API tester like Bruno. also check your Ollama logs to see if it’s getting the request and any other output there.
if you really want to stick it to Google you have to go for Firefox or something derived from it. Chromium gives Google a ton of leverage to push features to all of their downstreams. not sure what engine these are using, but i also prefer to use Firefox because it’s open source. if these were open source you could easily see which engine they’re using.
wow i had no idea development had stalled by this much. we are just now evaluating allocators for our Rust cloud services, and there are a lot of tribal debates between jemalloc and mimalloc with little to biased evaluation. i’m curious what a sufficient analysis even looks like
pretty scant on details. what is this doing for me that Podman or Containerd aren’t? “oPtIMizeD fOR aPPlE SiLICon” is fluff
it’s so much worse than the normal search. i would search for “dog” or “pasta” or “house” and get a pretty good result, but this conversational shit is just plain worse. and the “conversational” aspect is useless
pretty common misconception about how “AI” works. models aren’t constantly learning. their weights are frozen before deployment. they can infer from context quite a bit, but they won’t meaningfully change without human intervention (for now)
is there such a thing as “legitimate criticism” against an entire race of people? this writer is bonkers, and you can tell from the intro. seems like the actual content of the post was buried beneath the first paragraph where a rare few would find it. maybe it was wrong or illegal to fire this guy for being a racist asshole (being a state funded org or something?), but couching it in this narrative of “cancel culture” and “a violation of the first amendment” has fashy vibes to me. institutions should be allowed to control the narrative set by their employees. i understand that as part of my company my words reflect on them, and it’s up to their discretion whether they want to continue to associate with me based on the things i say. you have every right to say racist shit on your favorite fascist-owned platform, but everyone else has the right to tell you to fuck off.
also big downvote for posting this in a technology community
god help the poor soul tasked with maintaining a JDK 8 system in AD 2034