2012/2/13 Martin Dobias <[email protected]>: > On Mon, Feb 13, 2012 at 7:24 AM, Paolo Cavallini <[email protected]> > wrote: >> Il 12/02/2012 23:15, Larry Shaffer ha scritto: >> >>> didn't want to break it trying out new ideas (i.e. start doing forks and >>> merges). It's my opinion that most new dvelopers won't even need it for most >>> small projects. >> >> -1: I think an SCM is a necessity, to speed up collaboration. E.g., in case >> of API breakage, a dev can go round all the plugins and fix them quickly, >> whereas waiting for all authors to understand and apply the fix can be very >> long. > > I share Larry's opinion - for a small one man show project SCM is not > necessary. Of course at some stage it is useful to introduce it. > > Regarding API breakage and fixing plugins by someone else - I do not > support that. Even though plugin code may be hosted at hub.qgis, only > the plugin author is the real owner of the code - it is not in public > domain. Therefore doing some changes without author's permission (and > releasing new version on his/her behalf) does not sound good to me.
Martin, I couldn't agree more with you. We've already seen all the troubles and confusion that were introduced by trying to push plugin authors to have a GIT instance on redmine, I would keep this as a soft suggestion (even softer than it is now), nothing more. If an author doesn't want to maintain a public code repository it's his right to do it, and this is not a good reason for not hosting his plugins on plugins.qgis.org, IMHO we should implement other methods to evaluate the quality of a plugin, a voting system is already in the pipeline and could be implemented during the next code sprint in Lyon. -- Alessandro Pasotti w3: www.itopen.it _______________________________________________ Qgis-developer mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/qgis-developer
