Skip Navigation

Posts
15
Comments
450
Joined
3 yr. ago

  • The people that use generative Al for art have no interest in being an artist; they simply want product to consume and forget about when the next piece of product goes by their eyes. The people that use generative Al to make music have no interest in being a musician; they simply want a machine to make them something to listen to until they get bored and want the machine to make some other disposable slop for them to pass the time with.

    Good sentiment, but my critique on this message is that the people who produce this stuff don't have really have any interest in producing what they do for its own sake. They only have interest in producing content to crowd out the people who actually care, and to produce a worse version of whatever it is in a much faster time than it would for someone with actual talent to do so. And the reason they're producing anything is for profit. Gunk up the search results with no-effort crap to get ad revenue. It is no different than "SEO."

    Example: if you go onto YouTube right now and try to find any modern 30-60m long video that's like "chill beats" or "1994 cyberpunk wave" or whatever other bullshit they pump out (once you start finding it you'll find no shortage of it), you'll notice that all of those uploaders only began as of about a year ago at most and produce a lot of videos (which youtube will happily prioritize to serve you) of identical sounding "music." The people producing this don't care about anything except making money. They're happy to take stolen or plagiarized work that originated with humans, throw it into the AI slot machine, and produce something which somehow is no longer considered stolen or plagiarized. And the really egregious ones will link you to their Patreons.

    The story is the same with art, music, books, code, and anything else that actually requires creativity, intuition, and understanding.

  • The key point about codeberg as I understand it is it’s meant for foss projects. It’s not really much more complex than that. Want to host non-free software, or want to use it for your company’s private code repository? They don’t want that on their servers, so either find an alternative or self-host forgejo, which is the same code (derived from gitea) that powers codeberg itself.

  • Fact Check: Verified true!

  • Can’t read my emails on it though.

    Great feature!!

  • The statement did not use the term “occupation,” rather using the term “takeover,” due to legal ramifications for the civilian population in the enclave, Ynet reported.

    when the fuck did legality ever matter to israel?

  • I also try, but it is invoked on my behalf. For instance, at work if I make a pull request now multiple AI bots are summoned to give an analysis of my code changes. It's extremely verbose and annoying, and I think basically nobody reads it because all it does is just spam the comments section with way too much text.

    I vehemently hate OpenAI, ChatGPT, et al. At least it's funny when it summarizes my changes as significant improvements that improve code maintainability. I guess getting glazed by the bot in a way my manager can see is helpful to my career? Though honestly he probably also doesn't read that shit. So glad all this energy is wasted for nothing.

  • “The enigma of success” whoever came up with that line has truly mastered the art of auto-fellatio.

  • That’s slick in how straightforward it is. I like the offline element you get from printing it, too.

  • For both our sakes, I hope we can find something that works for us. I don’t need to be on my productivity grind 24/7. I don’t desire that at all. But I really don’t like the feeling of completely misspent time. I want the balance of doing what I want to do AND totally relaxing (physically+mentally) when I feel it’s time to relax

  • Less than 10% of plastic is recycled.

    Less than 10% is recycled and nobody wants to recycle it anyway, because it is difficult to recycle and the resulting recycled plastic is worse.

  • RAM

    Jump
  • I've never seen a distro take more than 2gb RAM ootb (ubuntu gnome and kde are probably the "heavy" contenders), excluding precached files. In either case, Windows or Linux, you lose big time the moment you launch a web browser.

  • I love the thought of productivity being measured as more and more LoC accepted month over month. This month it was 250k, but maybe next month it will be 350! Soon their OpenAI API front end will have more LoC than the Linux Kernel!!

  • Why does this story magically no longer become interesting because of a group that helps defected NKs?

    There is nothing magic about it. The organization that's cited isn't the problem. The problem is the BBC cites that org as proof that this person's claims are true. But neither that org nor the BBC have said, "we have corroborated Jin-su's story." On the contrary, the BBC just admits they didn't or couldn't corroborate the story themselves. So in my mind I may as well have read this article on any rando's blog post, or in the NYT in 2001 under a Judith Miller byline. It lacks credence.

    I wouldn't have had anything to say if BBC said that they reviewed some documents that showed Jin-su's claim. Maybe a few of the "hundreds" of fake IDs that he used, for example. But instead they just read another testimony from PSCORE. Was that other testimony verified? They don't bother explaining. So they just use an unverified testimony from PSCORE and pass that off to make the reader believe that that's good enough in place of actually verifying Jin-su's testimony!

  • Would you expect a news outlet to be able to somehow verify the testimony of a prisoner of war before reporting on it?

    "If the circumstance were different would you expect something different?" is what you are asking me. The interviewee isn't a POW, but a defector. And not an escapee, because according to the article he was already sent abroad, so it's not like he fled with merely the clothes on his back and a story to tell. So I would presume he would have a bit more evidence to share with the BBC than just a story, just as many of the people responding to me seem to presume that because it's been reported by the BBC it's prima facie undeniably true.

  • You're tedious and annoying.

  • In cases like these the journalists can and often do say something to the effect of they were able to corroborate the claims. But you're super right about being careful, because they also can mishandle the data they receive to the point where they dox the anonymous source, too. That's what happened with Reality Winner and The Intercept. They botched it, and she was arrested.

  • I'm not doubting the N Korean scheme to infiltrate IT jobs. There's even that woman who was prosecuted (I think she lived in Arizona?) because she is one person who acted as a facilitator for this scheme. My point is the BBC ran a story with an "anonymous" source then admits in the middle that they couldn't substantiate any of the claims. That's the problem here.

  • You're not addressing the fact that BBC admits they didn't/couldn't substantiate his claims, which apparently is no problem for your own journalistic standards.

  • A TeamViewer shooter coming out to downvote you lol