There are many reasons, but I think the two most important are:
Linux does not have a single source of truth. Yes, there is the tree of Torvalds, but thats just for the mainline development. Every company developing Linux has their own trees. They may merge commits off the Torvalds tree and vice versa, but most of the time those developments have different goals (ie Ubuntu vs embedded car media system). GitHub is too much tied to one source of truth.
Pull Requests. Githubs way of receiving contributions is via Pull Requests. For a development with lots of different contributors and maintainers, this does not work. A contribution to the graphics system goes through at least to levels of maintainers that provide feedback, before it goes to Linus for final merge. Having all contributions for all parts in one place would be too much.
The point is, a fork in the Github sense is to provide changes to thr master copy. As Linux as multiple masters, having all contributions go zo one master copy does not make sense
There are many reasons, but I think the two most important are:
Thank you so much.
I was able to glance at what such a huge project involves.
That’s nonsense. If anything Github has a bunch of features that make it easier to keep track of forks.
The point is, a fork in the Github sense is to provide changes to thr master copy. As Linux as multiple masters, having all contributions go zo one master copy does not make sense
Also see this reply from Linus Torvalds himself: github.com/torvalds/linux/pull/17#...