Wannabe Github competitors have two major problems: 1) it's free and it's hard to make a profit competing with free. Never mind that it isn't really free. It's

Wannabe Github competitors have two major problems: 1) it's free and it's hard to make a profit competing with free. Never mind that it isn't really free. It's free enough that making a profit doing the same thing just is not a great plan. 2) it quite literally has close to 100% of all oss projects in existence on their platform. There are no alternative platforms that have anywhere near close to the same network of developers, companies, etc. Being part of that network provides discoverability, easy access to other developers, etc. People care about Github stars. Nobody cares about whatever the Gitlab equivalent is to that. Feature wise, Github is a commodity. Hosting projects doesn't cost a whole lot if you think about it. Most git projects aren't very big. They don't get a lot of traffic. Etc. So the real cost of hosting a project is pretty low to begin with. There are of course some bigger projects that would generate a bit of traffic. But on average, it's not much. Gitlab and others have proved long ago that it's not that hard to replicate what Github did. And that in turn prompted Github to offer it for free because they recognized it was a race to the bottom. Making private projects free entrenched them as the #1 company in this space. And of course they are not for free if you are running big enterprise organizations on them. Sponsoring a lot of tiny OSS projects, libraries, struggling startups, etc. with that revenue isn't that expensive. And it makes sales to companies really easy. Because they and anyone they hire are likely already on Github anyway. It's of course not that hard to roll your own clone that does more or less the same. Lots of companies have actually done that and some of them have a quite nice business offering that as a SAAS service to companies not interested in using Github for various reasons. So, the added value of doing that some more only diminishes its value. To the point that it's a fairly pointless thing to do. Even if you do it well, it's basically just yet another Github clone without a network. People that argue the social network doesn't matter are missing the point: it's the thing that matters most. Software development is a social process. And a social network without any members is just not that interesting. And not getting that you are developing and competing with a social network is not a great premise for coming up with something better. Hypothetically, if somebody were interested in such a thing, imitating what Mastodon is doing to Twitter wouldn't be a bad idea. Git is decentralized by design. But Github isn't, which is in a way the biggest problem people seem to have with it. So fix that problem. The fix is not replacing it with another centralized social network. That would be moving the problem, not solving it. Of course the trick is finding a business model to fund that.