Given the small number of developers and packagers currently working on Open Mandriva it would be a useful thing to be able to reduce the amount of effort required of the Dev group to resolve a bug. How can this be achieved?
- One Idea might be that the QA team tries to investigate the issue so that they can suggests easy resolutions like 'upgrade, patch with (elsewhere) existing patch, remove from packages, ...'. So the devs will have less work figuring out..--Arisel (talk) 22:25, 7 May 2013 (UTC)
- We could enlist the help of competent users to help with this work; see below--Colin Close (talk) 23:30, 7 May 2013 (UTC)
- I think we should try to be the first line of defense, clearing away unneeded bugs. But if we cannot, they'll still have to go to the developers... --Rxu (talk) 15:18, 7 May 2013 (UTC)
- A possible idea to reduce the load on the devs may be to encourage users to participate. There must be more people around like Ben and I who have used linux for a while and can at least determine why a program or system is failing we may not know exactly how to fix it but we can nail the source of the trouble. Many bugs are generic and occur on other systems; sometimes a couple of searches on Google can reveal the cause. There is a bug like this in Cooker at the moment in fuse which manifests itself when trying to mount an ntfs filesystem. A bit of Google research showed that the problem is due to the use of the GNU Gold linker which does not search for indirect symbols. Rebuilding fuse and ntfs3g with the bfd linker allows it to work (with the Gold linker).
I envisage something like a list of users ranked by skill we would initially review the bug and pass it to someone with sufficient experience to at least diagnose it if not to post a fix. Users who fix lots of bugs gradually increase in skill and are allocated more difficult (interesting) bugs. --Colin Close (talk) 23:30, 7 May 2013 (UTC)
- Categorisation of bugs. Something I have not see in a Bugzilla issue tracker is the categorisation of bugs by type. i.e. Segfault, missing symbol, failure to start This categorisation if adhered to would possibly may it esy to guides them to the most efficient resolution. --Colin Close (talk) 23:30, 7 May 2013 (UTC)
- So essentially a bugfixing section of QA that we can allow not only ordinary users to join, but to help solve bugs? Also, if they're initially triaged, then we should be able to categorise. --Rxu (talk)
- That's the general idea. Do you think it would fly? As far as I know it hasn't been done before but I'm pretty parochial so I expect it has!!