How and Why to Ditch GitHub (opens in new tab)

There was very little "Why" in this article. It seems like: 1. Microsoft bad 2. USA bad 3. I had a gut feeling.

I think the Actions support being so lacking is a deal breaker for us. I know we can bring our own. But I really don't want to manage infra here and take on that responsibility.

Using alternatives - and asking for (or helping build!) the features that make GitHub so good - makes switching easier for those who come after you.

Let’s let those alternatives flourish :-)

> Compare this to a Workflow. All that YAML, all that code from someone else, all that additional complexity in spinning up containers or maintaining runners—just to build a static site? Nah, we've been played for absolute fools.

> While there are definitely valid use cases for Actions, I suspect they are fewer and further between than we've led ourselves to believe.

I use Actions to run automated tests and build documentation previews for my open source projects, which I think are very common use cases. I think this could be one of the largest obstacles to a lot of open source projects moving off GitHub, which when ignored like this reduces the credibility of a "How to Ditch GitHub" article. Don't misunderstand - I'm not saying that moving off GitHub can't or shouldn't be done even in these cases, but there's going to be some friction that deters people.

The point about static sites is pretty valid, but I'd point out that the method the article describes is just obfuscated FTP. Unless you really want the domain provided by Codeberg, wouldn't it be better to use a simpler free web host (not a software forge) and just upload your built site onto it?

For a self hosting solution I'd recommend gitea: it's a single go binary that needs almost no installation (only a config file), it's very fast, updates work great (replace previous binary and restart), supports ldap and has all the features of GitHub your love and even various features that free GitHub does provide (ie branch protection for private repos).

> GitHub's use of everyone's code to train its language models

Unless GitHub knowningly trains on code in private repositories (not that I am aware of), it does not matter where you host your code, or whether it's GPL or MIT licensed. Someone is training their models using your code. That's just a matter of life.

Does anyone have any experience using Radicle (radicle.xyz) for a non-toy project? I ran across it in an HN comment the other day and have been playing with it, but am curious to hear from someone who's done more than that.