• 0 Posts
  • 45 Comments
Joined 1 year ago
cake
Cake day: September 25th, 2023

help-circle
  • Google and Microsoft didn’t “kill” the decentralised e-mail of yesteryear. They beat it fair and square

    Sure, they might’ve cornered the market fair and square, but they’re certainly doing anticompetitive things in keeping it cornered.

    Just try setting up a mail server not connected to any of the big corpos (Google, MS, Cloudflare or their clients with more niche marketing) and see who will actually recieve your mails. You most likely won’t land into the Spam folder either.





  • Yes, but framing is important. Saying “Oh look what our perfect corporate buddies over at Taylor let us do even though it’s their call” (a huge lie btw.) vs. saying “We finally got this victory, we can finally do part of what we should’ve never have been unable to do due to corporate greed, thank you Taylor for getting some sense, it seems like your scrooges still have some semblance of a soul left” is a big difference. As always, the truth is somewhere in between these two extremes. However, I’m inclned to lean towards the latter more than the former on the spectrum.







  • If anyone deserves copyright over a photo, it’s the people that had their work photographed without permission. Then, the most deserving of the copyright are the camera and film manufacturers that made photography possible.

    I think this is an angle that isn’t pften taken. The advent of photography was a very similar situation to the current advent of AI.

    However, there are some crucial differences. For example, a photo can realistically be taken for personal use, which is either protected by law, or at least tolerated. AI, on the other hand, doesn’t have this going for it (you wouldn’t really go to the trouble of training an AI model for personal use). Even if the model and everything else is fully transparent and open source, it’s still gobbling up copyrighted data for commercial purposes - the model’s authors or the users’. Luckily, there is no AI fair use carveout (and I hope there won’t ever be one).

    Another thing I’d like to point out: in the vast majority of european legal systems copyright isn’t called “Copyright”, but “Authors’ rights”, i.e. its primary purpose isn’t to restrict copying as much as it’s protect the interests of the author (not publisher/corporation, although this unfortunately got bastardised a while ago).

    I can only hope the EU takes a reasonable approach to AI (that is, ban it from gobbling copyrighted work, require current “tainted” models be purged along with corporations paying reparations to the authors, as well as banning EULA clauses along the lines of “by signing up we get to feed all your information into the AI”).

    By my first comment I was trying to point out the fact that the “time invested” argument isn’t that strong. That doesn’t mean there aren’t better arguments or that I don’t agree with the general idea, just that we need better arguments if we want to win this fight.