> Again, this is false. It is *mindshare* and *momentum* that keeps everyone > concentrating on the same branch, not technical quibbles like revision > control methodology. Linux is a good example of mindshare and momentum over > distributed development - the distros all stick to similar branches (and not > always Linus' branch; -ac was the relevant one during early 2.4). > > Look at OpenOffice.org. The distros ship ooo.ximian.com's branch, not the > Sun branch. Why? Certainly not revision control centralisation - Sun didn't > manage to bottle that genie! ooo.ximian.com has the momentum, and that's > what keeps it ahead. > > At any time, someone could fork or branch Helix or Evolution, and if done > the right way (and with good rationale), they could capture the momentum and > decimate the old branches, cf. XFree86 vs. Xorg.
If you get the people to join your fork, then you are the project, it doesn't matter who has the repo or what version control system it is under, Jeffs examples are all good, can you provide some of companies going the other way? I know when I contribute code to projects I research their licensing and stuff before hand, I expect most other FOSS contributors do the same and decide based on the project and project history whether they want to work on that project.... and if they are happy for their code to be used by the copyright holder.... granted I've never signed a copyright release but I also understand that all the work I put into the drm and Mesa and Xorg can all end up inside someones commercial product.. but I'm happy that the benefits to the community outweighs the advantages any big evil corp is going to get.. Dave. -- David Airlie, Software Engineer http://www.skynet.ie/~airlied / airlied at skynet.ie pam_smb / Linux DECstation / Linux VAX / ILUG person -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
