Hi Bertho, I've been discussing this on the Discord as well, and I think the core issue isn't really about who the custodian is.
If the code is public, it will be scraped. That's true on GitHub, on Codeberg, on a self-hosted server. Changing the custodian doesn't change the fact that our code is open source and publicly accessible. Any AI tool, not just Copilot, can ingest it. The real tension is between open source and AI, and that transcends any hosting platform. It's a legal and policy problem that needs to be addressed at the tool level and through legislation, not by moving repositories. Best regards, Luca On March 27, 2026 10:27:35 PM GMT+08:00, Bertho Stultiens <[email protected]> wrote: >On 3/27/26 2:58 PM, Luca Toniolo wrote: >>> My irony-meter just overloaded and have to appreciate the boldness of the >>> statement :-) >>> /LinuxCNC/ is just a small volunteer-run party. >> Ha, fair point Bertho, I walked right into that one. > >Happens to the best of us ;-) > > >> But that's actually my argument. We are a small volunteer-run project, which >> is exactly why we shouldn't be taking on additional infrastructure to >> maintain. Every hour spent managing servers and CI runners is an hour not >> spent on the actual codebase. >I don't think that managing all infrastructure ourselves is the best way to >proceed either. However, there are many middle ways to get things done for >least extra time. That said, I do not think we need to do all that much work. > >On a side note. Coherently managing the whole project is not only about a git >server or a CI pipeline. There are currently many disjoint servers used, like >web, forum, wiki and buildbot. And there are plenty of problems with that >because we lack the organisation to bundle resources structurally and assure >continuity on many levels. So I can understand the reluctance to add to that >burden. But we can create an organisation to handle it. Others have done it >before and we can learn from them. I'd be happy with that, even if we need to >drop some bells and whistles to get there. > >But what this discussion started is: Who do we allow to be custodian of the >data we generate in developing the project? That is what the copilot thingy is >about. I for one do not trust github as a good and honest custodian. >Github/Microsoft don't really care about what we actually want, just their own >bottom line, regardless who or what they need to trample on. The copilot >controversy exemplifies that very well. > > >-- >Greetings Bertho > >(disclaimers are disclaimed) > > >_______________________________________________ >Emc-developers mailing list >[email protected] >https://lists.sourceforge.net/lists/listinfo/emc-developers _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
