Being brought up from the discussion surrounding the RadeonHD driver being vandalized
, which wound up just being a prank by two X.Org developers to torment one of the former RadeonHD developers, was a discussion why more companies don't contribute back to X.Org. Do companies think the X.Org code is too hard? That it's not worth the time? Is it all politics?
Here's the beginning of this new thread
. Reasons expressed by other developers range from Microsoft F.U.D. to vendors just wishing X.Org would go away, provide the vendors with a competitive advantage by not pushing their patches upstream, to their code just being in a poor and unreliable state. There's also the matter that with the modularized X.Org state, it's easier to keep and maintain an out-of-tree DDX driver than it is maintaining an out-of-tree kernel driver on Linux.
In an attempt to increase the X.Org developer base, there are efforts underway to increase the X.Org documentation pool so that the barrier to entry is lower, but still there is a whole lot of areas of the X stack that are currently left undocumented. There's also reported to be a few days prior to the 2011 X Developers' meeting where it will focus upon improving the documentation at large.
Besides the lack of documentation and higher barrier to entry in general than other areas of the Linux stack, what's stopping you or your company from contributing back to X.Org in one form or another? It will be interesting to see if the situation with Wayland
turns out any better since its code-base is significantly smaller and cleaner (though it's now LGPLv2 licensed
) than the aging X11 Server.