| |||||||||||||
Reasoning behind the "preferred form" language in the GPLReasoning behind the "preferred form" language in the GPLPosted Mar 7, 2011 5:50 UTC (Mon) by foom (subscriber, #14868)In reply to: Reasoning behind the "preferred form" language in the GPL by dlang Parent article: Commitment to Open (Red Hat News)
> I also agree that just releasing one modified tree instead of a series of patches is not an issue
Well..I think really that *is* the main issue. That's what started this whole conversation after all.
It may not be illegal or "against the spirit of the GPL" to just release the modified source rather than a git repo or set of patches, but it's not very nice. The legality or spirit-of-gplness of releasing those broken out patches to customers only is actually kind of a distraction from that main issue...
(Log in to post comments)
Reasoning behind the "preferred form" language in the GPL Posted Mar 7, 2011 6:03 UTC (Mon) by dlang (✭ supporter ✭, #313) [Link]
I disagree. i don't think there is any problem with having the SRPMs not take the time to apply patches when used.
I also think that going down the road of requiring access to version control repositories that are used internally is a very bad idea.
think about the case where the source is distributed along with the binaries instead of being available for download separately. is the supplier requried to include a complete version control repository? what if that repository is using proprietary software? you could get into the situation where the source is being provided in the manner that the devlopers prefer to work with it, but the resulting files are completely useless to anyone who isn't willing to spend the money (and agree to the terms of service) of some proprietary software package. and to make things even more interesting, what if that VCS system only runs on windows?
you can't have it both ways. either the GPL does not require access to the VCS (in which case releasing one big patch, or a modified tree is not a problem), or it does (in which case the problems above can apply)
Reasoning behind the "preferred form" language in the GPL Posted Mar 7, 2011 12:54 UTC (Mon) by foom (subscriber, #14868) [Link]
Well, I don't really believe the GPL requires it, but I still think it's quite poor for RH to do this. Just because something isn't litigable doesn't mean you have to like it.
I think it's a problem because other kernel developers are complaining about it making their life more difficult, and the reason for the change was indeed to make other people's life more difficult.
|
|||||||||||||