Currently, the Lemmy Project only uses Github for its repositories related to Lemmy’s development (e.g. Lemmy, Lemmy-UI). GitHub is a proprietary service, and it is owned by Microsoft. These facts open the door for a myriad of potential issues across the ecosystem, and community. I would like to clarify, though, that I don’t think that it would be a wise decision, currently, to remove Github as the primary location for development, but I would think that it would be a good move to mirror Lemmy’s repositories to a FOSS service (e.g. Codeberg). I personally would advocate for the use of Codeberg, as it is entirely open source, and non-profit, and they are currently working on implementing federation (through ActivityPub) – all these things, I think, align well with Lemmy’s role in the wider community, and its more general philosophy. In the future, I would ideally hope for a permanent move to such a service, but, in the meantime, I think it would, at the very least, be a wise, if not only benevolent, move.
I decided to post this here, as I felt that it didn’t seem appropriate to post it as an issue in any of the Lemmy repos.
Yea i always thought why every one was using microsofts github when it goes against their ethics and philosophy .
Codeberg and other alternatives are used by 2 people, if not more. If a repo is hosted on such unpopular service, potential contributors must register a new account. This is very frustrating if you want to report just one issue or make one pull request. Self-hosted repos are even worse.
This problem can be solved by implementing federation. GitLab, Gitea and Forgejo already working on it, but really slow.
Codeberg and other alternatives are used by 2 people, if not more
It last reported it has about 400 members (people who pay money) , liberapay shows about 190 supports (and the number is slowly but consistently growing for years).
This is very frustrating if you want to report just one issue or make one pull request. Self-hosted repos are even worse.
It takes about a minute to make an account and store it in a password manager, it might be better because a higher threshold for contributing might mean a higher average quality of contributions.
Is there a problem with GitLab if used on gitlab.com? Would it be worse than GitHub?
Still subject to a company’s will in the cloud. For something like this (not doing anything in a legal gray area) it’s probably fine, though. For now, places like GitLab know they’d dissolve their trust with the world in an instant if they fucked around with a legal projects code (and github knows this too).
I think the point is less worry about corpos and more about “let’s support FOSS since an actual alternative exists”.
deleted by creator
how so?
Most potential contributors are on GitHub.
It’s yet another instance of the network effect.
Honestly I didn’t even knew that codeberg existed. A Open Source gitub would be pretty pog ngl
Okay, I‘ve read this now a few times. What does pog mean?
It means “good”, “agreeable” or “desirable”. It’s modern slang so I don’t blame you.
Just for completeness, “ngl” stands for “not gonna lie” and means something akin to “in my opinion” or “to be honest”.
Short for “PogChamp”, which is a Twitch chat emote featuring an expression of surprise like “😯”. Is used to be a picture of streamer GooTecks until I think he made some comments supporting Jan 6th and twitch changed it (in the worst way possible, but that’s another story)
So saying something is “pog” means you are excited/hyped for it. Other uses include “Poggers” or “Poggies”. Over time it’s worked its way into the larger Internet slang lexicon as Twitch has grown.
Play of the game.
I’m all for open source services, but realistically, what potential issues are there with using GitHub?
Every contributor has a copy of the Git repo, so isn’t the worst case basically losing access to issues and similar data? And even that is very unlikely.I can’t give you an exhaustive list, but I’m, for example, not a fan of:
- Needing an account at Microsoft for reporting issues, contributing code or searching code in repositories. Microsoft operates in the ad/tracking business and is known for violating laws.
- Microsoft will potentially link this account data with LinkedIn or other Microsoft services in the future.
- Microsoft violates the licenses of code hosted on GitHub to train their AIs.
- Microsoft is known for creating lock-in effects and EEE, generally putting humanity worse off for their own profit. I’d rather not contribute to that, neither with code nor socially.
And just in general, Microsoft will enshitify GitHub one day. Its inevitable for every free service run by a public for-profit corporation. You can count on this as much as you can count on climate change.
So why wait until it starts happening? Get started with the move now.
They make a lot off of paid repositories and enterprise contracts, id be shocked if they had to enshittify it
You can self host gitea and begin mirroring the Lemmy repos yourself right now.
Indeed! But, I think that it would be preferable for it to be handled officially.
It’s actually already mirrored to Codeberg, though it seems the devs haven’t pushed the changes for a couple of months.
Note that Codeberg discourages read-only mirrors though.
It’s actually already mirrored to Codeberg
My mistake! I swear that I did check before I posted this! I must’ve improperly searched for it, or somehting, because I really don’t remember seeing it when I looked.
it seems the devs haven’t pushed the changes for a couple of months.
Indeed, that seems to be the case. I wonder why?
EDIT (2024-01-19T00:45):
I think that I found what I did “wrong”, initially: I searched for “Lemmy” under “Repositories” on Codeberg (which didn’t display the Lemmy repos), where I should’ve searched under “Organizations”. Personally, I feel that this is a bit of a UX issue on Codebergs end; the main search should be a fuzzy search with options to fliter afterwards.
Yeah devs should definitely take on extra work that serves no purpose whatsoever.
extra work that serves no purpose whatsoever.
No purpose? I outlined the main rationale in my post, and this comment followed up with more detail.
