Raphael

@SheDrivesMobility@norden.social Das war das, was die Bedenkenträger damals im Big Data Hype mit " wir wissen nicht, wer in Zukunft was mit den Daten machen wollen wird" meinten. Seufz. Wenn wir wenigstens draus lernen würden ...

@nicolas17@social.treehouse.systems Not what I said, and a false dichotomy.

@mavu@mastodon.social @wtay@fosstodon.org @TheEntity@social.treehouse.systems @lu_leipzig@troet.cafe @downey@floss.social And how would we reliably recognize such contributions? I'd rather they leave the attribution in instead of lie about it (if only by omission). 🤔

@marcuwekling@cultur.social Aber ohne Brunnen!!

@nikclayton@mastodon.social @bms48@mastodon.social @jonny@neuromatch.social @f4grx@chaos.social @srtcd424@mas.to @be@floss.social I was (maybe falsely) conflating most of that under 'discussion'. Basically, "text about that change". I think all of that can and should live on the instance hosting the repo. Why should I post a comment on my instance for a PR on somebody else's instance for a repo on a third? GitHub is not Twitter. Every repo has a clear owner account/org; I need to go to their instance to access and engage. Cross-repo links are ... just links. Backlinks are nice to have, but ultimately a decades old feature of the web with lot less complexity compared to fedi-style federation. Workflows and the questions around who can trigger runs with which secrets on which infra are interesting, and in fact they already are on GitHub. Account/org boundaries remain the same in a distributed setting, so nothing new really. Global search: use a search engine. Global dependency graphs: interesting; luckily, ultimately near-static information. Overlay tools exist for GitHub, could work similarly with a distributed alternative. But I'll grant that this feature requires more than "everything lives at the owner instance". I realize I make some simplifying assumptions, but they cover 99% of what I've seen. What am I missing, beyond the Linux kernel and another few exotics you'll never convince to change their ways, anyway? In summary: I understand that federation is rough, but I don't think we need full fedi-style federation for a suitable (not perfect!) GitHub replacement. The domain lends itself more to "partitioning" distribution than a "social network".

@nikclayton@mastodon.social @bms48@mastodon.social @jonny@neuromatch.social @f4grx@chaos.social @srtcd424@mas.to @be@floss.social What _is_ a PR, though? Basically, literally, a remote Git ref plus a discussion. Git covers the former, no federation needed. The rest would be solved by a shared auth provider; let PR comments live on the instance that hosts the target repo. Of course, shared auth is nothing people want of a distributed system, right? 🤔 Would we want the instances to accept each others' tokens? Which makes me think of how many dev-work-related systems I've logged into with GitHub as auth provider (seemed reasonable at the time), and I don't know. "Social" auth was always a bad idea ... 😪

@jonny@neuromatch.social @f4grx@chaos.social @be@floss.social I've never been hired after someone looked and my GitHub account, I don't think. If they wanted to that, they could check any other platform just as easily. That said, you don't have to delete anything on GitHub. Just start new projects elsewhere. Move active development off, push-mirror to GitHub if you want. Keep the stars while building a new place. Will people follow? Probably not, not at first. For me, it's always a noticeable gap when something it _not_ on GitHub. But we'll get used to it.

@jonny@neuromatch.social @f4grx@chaos.social @be@floss.social My perspective is that we don't really need interop that isn't covered by their APIs or open standards, anyway (Git, OCI, ...). Smooth-ish migration is enough. Unless you want to federate issues and discussions with GitHub. I don't see any reason to do that, though. What am I missing, what sort of interop are you looking for?

@jonny@neuromatch.social @f4grx@chaos.social @be@floss.social I think just a significant fraction of customers moving will be threat enough. And, by my guess, it'll be much easier to convince execs to move back to on-prem or EU-SaaS dev infra than it was to get them "out". For better or worse. If Github went back to being the OSS place it'd be fine; only of course it can't.

@f4grx@chaos.social @srtcd424@mas.to @be@floss.social @jonny@neuromatch.social Yeah. That the internet is and should be free (and open 🙊) is quite the entrenched assumption, isn't it. If we had and were used to paying for things, it'd be a lot different.

@f4grx@chaos.social @srtcd424@mas.to @be@floss.social @jonny@neuromatch.social That's a core difference, isn't it. The lock-in effect is a lot weaker. As an individual, you _can_ just leave, it's just a little work. You have local copies of your code, by design. Community ... there isn't one on any of _my_ repos, that's for sure. And I don't need to host my stuff there in order to engage with any other project. Basically, each repo is much more independent compared to, say, an FB account.

@f4grx@chaos.social @be@floss.social @jonny@neuromatch.social TBH, I don't see that happening. So far, there's no tangible benefit to leaving GitHub. It's a bet. Plus, non-paying devs/org leaving won't hurt them, not in the short term. High-paying companies will have an even harder time leaving. What a shame, too, after "we" struggled for years to make deciders embrace SaaS in general and GitHub in particular ... 😪

@UKFilmNerd@mastodon.me.uk Apparently the first human to ever turn 100, if you can believe the internet. 🤦

@jukkan@mstdn.social _Is_ there even headless tooling by MS to work with Office docs? I wouldn't know, but I wouldn't be surprised at all if there wasn't. Certainly no OSS, given through how many hoops I've seen colleagues jump to implement the inevitable "make it export Excel" requirements.

@srtcd424@mas.to @be@floss.social @jonny@neuromatch.social Aye. It's my understanding that Gitea/Codeberg have a mostly compatible CI/CD system. GitLab would be harder. Integrations with APIs would be rougher. But yeah, rebuilding all the stuff coupled to GitHub is a lot of work. Still dwarved by "let's give users N months to migrate now that we have it working", by my guess, though. 😉

@srtcd424@mas.to @be@floss.social @jonny@neuromatch.social I don't have any reason to think they couldn't migrate off of GitHub. First, mirror elsewhere. Then, move (reading) clients to the mirror. Then, move (writing) operations & communication over to the new system; GitHub becomes a mirror. Then, deprecate GitHub; make old clients emit a warning if you can (probably not). Finally, turn off GitHub. Possible, but this takes a _lot_ of time if you want it to be graceful, never breaking clients. In fact, each step will take months: planning, execution, communication, bugfixes & refinements. The whole thing will take years. Absolute mayhem if you're forced to move or, even worse, restore elsewhere without warning.

@be@floss.social @jonny@neuromatch.social If GitHub collapses while the majority of OSS is still there we are in trouble.

@catrinity Datenpunkt: Konto bei der Postbank, App von Wero selbst. Zum Verbinden der Accounts (nur mit Email, keine Handynummer) mit den Bankkonten brauchte ich keine Banking App (hab keine), vom 2FA der Bank mal absehen (dafür hat man Optionen). In den Einstellungen kann ich einstellen, dass der Nachname ausgesternt wird. Konnte noch nichts damit bezahlen oder empfangen, kann also leider nichts dazu sagen, wie sich die App verhält bzw mein Account in anderen Apps erscheint.

@cryptkiddie @dickenhobelix Der frühe Vogel fängt sich eine

@dickenhobelix Zu viele Follower oder was?!

@danluu Apparently the slopmachines can't be trusted to credit themselves reliably? Go figure. 🤭

@tagesschau@ard.social Wie wärs mit PV auf der selben Fläche und stattdessen Ladestrom subventionieren?

@raven667 @alexanderkjall Sounds like reasonable criticism to me. But then, an extended group of volunteers can only be expected to do so much. If we want SLAs, we have to pay people.

@alexanderkjall And there I sat, thinking it was just me being too dumb to figure out whether I had a patched kernel without running their bespoke, obfuscated script.

@nerdherz Wir hatten ihn nicht verdient.

@yosh Nothing is safe from the slopmachines 😞

@MugsysRapSheet @ironicbadger I have the opposite experience. There's a learning curve, but I can usually find commands to debug and fix problems on my Linux boxes. For Windows, the internet is full of variants "try reinstalling". And yes, I consider the CLI a boon in that regard. Even if I brick GUI login for some reason, I can proceed to fix. Also, there will never be useful GUIs for rare problems. Plus, to be fair, niche OSS GUIs are usually not good, with notable exceptions.

@GossiTheDog They explicitly say so in their blog post, so ... ?

@joelvanderwerf @unaegeli @signalapp Yes, it is. And it still trains the wrong response, I think. Users _will_ repeat the thing that made the dialog box go away last time. Even if that last time was a different app.

@signalapp@mastodon.world While I'm generally on your side here, that argument is bollocks. "We told you four years ago in that message you left on your old phone!" That just doesn't work. You have users losing their chat history because they ignore instructions. That's _users_ for you, but that's also human beings who have no concept of how crypto systems work (and who can blame them?). "But WhatsApp doesn't need me to do that?!" Sigh. FWIW, I don't think there's a technical solution to the education problem. We, as a society, need to learn the basics of IT and how to remain somewhat safe in an IT system, the same way we have an idea of how a stove works and how we've learned to turn it off again.

@Diogenes Hab womöglich bissl viel gelacht. 😁

@kplx Autsch.

@xahteiwi Wenn solche Worte fallen, vermehrt Resignation. Wer nicht hören will ...

@xahteiwi@mastodon.social Noch paar Minuten, dann ruft jemand "emacs!" 🤭

@sindarina Convenience to the detriment of others and the planet is pretty much the theme of the last 150y or so, as is convenient ignorance of that fact.

@b_cavello @zkat Difference is how easy it is to say "no". GenAI is, as of yet, not structurally baked into our society. We can just choose not to do that, and bear the FOMO.

@mattgemmell@mastodon.scot How to implement that without horrible side effects, though?

@ScientistRebellion_NL@social.rebellion.global Humanity, always trying to use linear regression on non-linear processes.

@danirabbit@mastodon.online More like asbetos, what with all the long festering and then tumors.

@tante They're already halfway done in some places. Some folks can't read, write, or do basic arithmetics without tool support anymore. Let alone summarize, argue, or conceptualize.

@alex@mytoot.de @fribbledom@mastodon.social As if work was the only source of stress.

@eribosot@mastodon.social @blackoverflow@norden.social Buying _less_ is not an option, hm?

@eribosot@mastodon.social @blackoverflow@norden.social Closer! On average, we don't vote for those politicians. On average, we buy cheap, not sustainable. That's on us.

@blackoverflow@norden.social @eribosot@mastodon.social Ressource-constrained people have little choice about many things, true. However, when given a choice, how do they act? Badly, most of us, most of the time. Latest experiment at scale: GenAI. Nobody had to start using ChatGPT. _Nobody._ And yet here we are. The "global north" societies are so complicit in global-scale extraction and exploitation it's hard to comprehend. You yelling at billionairs? Imagine what a random person from the "global south" has to say to you or me. An extremely humbling experience that would be, by me estimate.

@rebane2001 What kind of low-quality hosting is this? I see no blanket, nor complimentary water.

@petko I'll take purchase, thank you. Not that you can actually purchase and own much of anything anymore.

@InternetEh GenAI, that is. Some AI variants do have error bars.

@iris_meredith I would also add that while definitely many loud people jump on hype trains in tech -- maybe because of some internal malleability as you propose, maybe out of cynical financial tactics -- it's not been my impression that _most_ people follow along, not really. Most people seem to just keep doing their jobs; keeping your head down and letting the hype pass is in many ways safer (and easier) compared to voicing loud disagreement with the Cool Kids. Come to think of it, _that's_ a lesson nerds learn in school, isn't it? 😕 Go along, shut up, or be isolated.

@iris_meredith Can confirm considerable cognitive dissonance between what I think creates value in the world and what "tech" is asked/made to build most of the time. Can also confirm strong conflicts of this line of work with needs of body and soul. Some of that applies to any desk job, I think. On top of that, there's definitely many people who invest most of their free time to "stay on top" of developments (which of course normalizes that behaviour and creates expectations of techies to always know their way around the new stuff). And _of course_ we make up narratives in an effort to make all of that make sense. How could we go on otherwise? Are these narratives more malleable in tech than elsewhere? Interesting thought. 🤔 "Strong opinions loosely held" _is_ a meme, and adjusting opinions in the face of evidence _is_ a strength (imho). Problem is: There's so little science behind software "engineering" it's almost embarrassing. I can see how in a field of low-evidence best practice, any new idea can shift opinions with little evidence in turn. Not sure how that's necessarily related to the cognitive dissonance / dysphoria line of thought, though. 🤔

@azonenberg @InsertUser @osm_tech Given who controls the app stores, courts may be more willing _and_ faster.