> Miro Hrončok wrote: > > The devel list is the first place where developers gather feedback. The idea > to > reach a committee (whether FESCo or Mindshare) before reaching to devel is > hence > entirely wrong in my opinion. Committees should rubber stamp community > decisions, not drive them. >
I agree that committees should not drive decisions. I did not mean to undermine the role of the Devel list in discussing technical ideas and getting wider feedback from the development community. I guess there is not an easy equivalent of the Devel list to Fedora's non-engineering community. > As for users communities, yes, that is a good idea. However, so is first > discussing this within the engineering contributors community and only once > there is a consensus, involve advocacy / user communities for more feedback. > Is there a plan or process for collecting advocacy / user community feedback for more feedback? My concern is this step will not be deemed critical or important enough, and it will be glossed over. I'm not opposed to development discussions, but I don't see where user community feedback fits into the existing Change process. Maybe I am missing something obvious. > Hence, I disagree with what you say - discussing this on the devel list first > is > the best thing to do IMHO. This is not how I wanted to represent my position. I'm sorry to have made it appear that way. > Adam Williamson wrote: > > There is no "why" behind this Change "other than eliminating an > admittedly tedious task for the QA team". That's the whole thing. There > aren't any "private RH BZ customer tickets". Why would RH customers > care about whether or not Fedora blocks on physical optical media > booting? The QA team has increasing demands for the ways Fedora is changing, and I also understand how optical drives are not part of those increasing demands. But to me, there is an opportunity to find a middle ground approach that reduces the work for the QA team without causing panic and fear in user communities in Global South communities that optical drives will be unsupported. Also I'm sorry about the "private RH BZ customer tickets" comment. Sometimes I am a little jaded as a community volunteer because there are some decisions made that I cannot always understand or follow. It was distracting for me to mention it here in this discussion and I didn't intend to make a jab. > Adam Williamson wrote: > > If there are all these thousands of people out there who care deeply > about the optical media...wouldn't it be nice if some of them turned up > and helped run the tests once in a while? > It would be nice, but this assumes a lot of the people who are most affected. Contributing to Fedora in free time is a privilege. Some companies strictly forbid employees from contributing to open source despite using it. Using presence of unpaid volunteers as a metric of engagement and interest does not match up to me. That said, I think there is opportunity to improve in attracting contributors. If it is important to continue the optical drive validation work, it would be cool to see Test Day events or a page in the Fedora Docs sites about how to run through the QA tests for optical drive validation. We could rally more people to do these things but we have not made a good-faith attempt yet at reaching those people and inviting them to the table. I'm not sure where to find this test case information myself! > Very very rarely. As Chris Murphy wrote, by pure coincidence one showed > up the day before yesterday (to be clear, it has nothing at all to do > with this Change proposal, the proposers of the Change were not aware > of that bug until well after the Change was submitted). Before that the > last time I can find, by searching my Bugzilla mail box for relevant > words at least, is 2015: > > https://bugzilla.redhat.com/show_bug.cgi?id=1250440 > > the last time we had a *fatal* issue that I can find is 2014: > > https://bugzilla.redhat.com/show_bug.cgi?id=1148087 > > Note, the testing isn't *hard* to do, really, it's just tedious and > time consuming. Not just the act of running the test (though that does > take quite a while, between the burning process and the boot, media > check and install itself), but the fact that it means we need to ensure > we have at least a couple of people who still have access to a DVD > burner and blank media. I'm warming up to an idea that focuses on empowering the community to lead this work in a sustainable way without mandating the QA team to test these every release. If I could suggest first steps to one way to "hand-off" to the community, it might be like this: 1. Publish "Fedora QA test cases" somewhere in the Fedora Docs site for QA: https://docs.fedoraproject.org/en-US/qa-docs/ 2. Write test cases based on the manual work done currently and how to validate success; publish in docs site 3. Write Community Blog post to point out to community that these docs exist If you (or anyone) doesn't like my idea, I encourage you to propose what you think would work. :) I believe there is a winning solution where the QA team is relieved of tedious work among increasing demands, and the distributed Fedora user community does not enter a frenzy when they find out optical drives are "unsupported". Not saying my solution is the winning one but I think this is the desired final-state to work towards. _______________________________________________ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
