On Mo, Nov 14, 2011 at 13:46:21 (CET), Daniel Holbach wrote: > Hey, > > Am 14.11.2011 12:28, schrieb Reinhard Tartler: >> On Mo, Nov 14, 2011 at 12:05:31 (CET), Daniel Holbach wrote: >>> While it would be nice for debexpo to have a similar feature, it seems >>> more worrying to me that we still advertise a process that is broken and >>> we set wrong expectations. >> >> What exactly are the 'wrong expectations' that are advertised to >> potential contributors? > > Currently the Ubuntu Wiki explains that in order to get a new package > into Ubuntu (and if you don't wish to get it through the Debian NEW > process), you have to upload it to REVU, etc. > > This creates the impression that somebody will review the package on > REVU. This is unfortunately not the case generally. My experience is > that you a) have to know an Ubuntu developer to get help or b) the > software you package needs to be of high interest to a lot of people. > > A lot of us put effort into reviewing packages on REVU in the past, but > it is happening less and less. I got the impression that a lot of Ubuntu > developers feel that pushing packages directly into Debian is more helpful.
I agree with this part. In order to solve this problem, we need more people that actually review (more) packages. > >>> Personally I'd rather like to see REVU closed, the documentation changed >>> and for packaging review (or general code review) any VCS be used, where >>> you can very easily track changes in packaging, without incrementing >>> packaging version numbers. >> >> Why do you think that they will be met (better?) with "any VCS"? > > Sorry, I wasn't saying that REVU should be replaced with "any VCS", only > that there is other ways to serve the need of "differing packaging > without new version number". > > > REVU has served us well in the past and I'm thankful we had so many > people working on its code and maintenance and others reviewing packages > on there. > > That said, it's quite harsh on new contributors who expect to get a > package into Ubuntu and they never get a reply. I feel we'd better get > involved in the Debian review process of NEW packages and solve the > problem together. > > (I hope that was a bit clearer now.) I think it is. Still, I disagree with the idea of shutting down a service because of lack of manpower. I imagine that there still are reviewers that appreciate the features of REVU for reviewing new packages. Shutting down REVU will most likely not result in such reviewers reviewing more packages. Quite the contrary, I expect this to reduce the available reviewing ressources, making the problem you've identified above worse. I'd rather propose to update the documentation wrt. to what users can expect. For instance, let's suggest that contributors without upload priviledges should solicit for reviewers in advance on mailing lists or IRC, and ask for the reviewer's preferred review method before uploading to a bzr branch on launchpad or to REVU. Cheers, Reinhard -- Gruesse/greetings, Reinhard Tartler, KeyID 945348A4 -- ubuntu-devel mailing list ubuntu-devel@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
