On Thu, Feb 15, 2018 at 3:05 PM, Christopher <ctubb...@apache.org> wrote:
> On Thu, Feb 15, 2018 at 2:55 PM Mike Drob <md...@apache.org> wrote: > > > On Thu, Feb 15, 2018 at 12:31 PM, Christopher <ctubb...@apache.org> > wrote: > > > > > While I think Spark's reasons might be interesting, I find it hard to > > > believe that any reason they might have would be so compelling as to > > > prevent us from even performing a trial. I think should still > experiment > > > with GH issues, as a trial, regardless of what additional input we > might > > > get from other-than-hands-on experience during that trial. > > > > > > I can think of several. > > > > Let's not assume that we have ASF policy blessing to do this. Somebody > > should research that. Or tell me that they've already researched it. It's > > likely ok, but I don't want to do dive headfirst into something stupid. > > > > > GH issues are supported by INFRA. > > Cool, that gives me more confidence in our ability to succeed. > > > What do we do if the trial is wildly successful? Is there a migration > plan > > for our currently open issues? We have almost 1000 of them. > > > > > As Keith said in the other thread, we don't need to have all the answers up > front. > You're right, we don't need to have all of the answers up front. This is one that I'd like to have some thought put into though. There's lots of things that are fine to handle as we approach it, but this one seems like it will lead to us having split issue trackers for _years_ down the road. > > > And this part is a little glib, but what would be a sign to the community > > that Apache is not the right fit for Accumulo? There are countless > > successful open source projects using GitHub as the canonical repo > > (complete with CI, issues, wiki, and site). Projects can use Google > groups > > for mailing lists when necessary instead of ASF lists. I see a slow > trickle > > in the Accumulo community that some given self-service option is better > > than the equivalent provided by ASF Infra. Sure, there's lots of > > alternatives out there for much of the tooling, but I like the > convenience > > of having it handled for me. Some people can disagree here, and that's > ok. > > > > > Both issue trackers are supported by ASF INFRA. > > > > Anyway, please please address my first two point, and at least consider > the > > third. Not necessarily you, Christopher; open question to the proponents. > > > > Mike > > > > > > > > > On Thu, Feb 15, 2018 at 1:17 PM Mike Drob <md...@apache.org> wrote: > > > > > > > Before switching to GitHub issues, I would like somebody to do an > > > > investigation into why Apache Spark isn't using them. They are heavy > > > heavy > > > > heavy users of PRs, but don't use issues and I'd like to see if > they've > > > > already done the leg-work on figuring this out. > > > > > > > > Not saying that we need to be like Spark in every way, but it's a > > > community > > > > that may have already gone down this path. > > > > > > > > Mike > > > > > > > > On Thu, Feb 15, 2018 at 12:12 PM, Christopher <ctubb...@apache.org> > > > wrote: > > > > > > > > > +1 for a trial... because we shouldn't be afraid to experiment with > > our > > > > > workflows. I also like working on GH, and want to see if it will > > result > > > > in > > > > > a valuable change or not for Accumulo, and if it is, I'd like to > > switch > > > > to > > > > > it at some point (I like the trial first, because I don't know if > > there > > > > > will be unforeseen problems). > > > > > > > > > > Additionally, I think that: > > > > > GH issue tracker could potentially lower the bar to reporting > > issues > > > > > since many devs use GitHub already. > > > > > It would allow integration/exposure via external initiatives, > like > > > > > Hacktoberfest by using GH labels. > > > > > It could improve readability of commit logs by avoiding lengthy > > links > > > > to > > > > > JIRA issue numbers. > > > > > No need to create an issue for the sole purpose of "checking the > > box" > > > > to > > > > > have an issue to mention in the log (since PRs are issues). > > > > > It would improve linking between PRs and issues. > > > > > Markdown in GH is much more friendly than JIRA's syntax. > > > > > Reduced mailing list spam from redundant emails and integration > > with > > > > > JIRA. > > > > > JIRA is frequently offline for maintenance. > > > > > > > > > > Some potential downsides: > > > > > Lack of multiple milestones for an issue (there are workarounds) > > > > > Lack of batch issue changes > > > > > Change in momentum could be confusing to contributors > > > > > > > > > > I think a trial period could help us understand whether the > potential > > > > > downsides are worse than the benefits, and whether or not we can > > > develop > > > > > workflows that mitigate against those downsides. > > > > > > > > > > On Thu, Feb 15, 2018 at 12:06 PM Mike Walch <mwa...@apache.org> > > wrote: > > > > > > > > > > > I would like to open discussion on moving from Jira to GitHub > > issues. > > > > > > GitHub issues would be enabled for a trial period. After this > trial > > > > > period, > > > > > > the project would either move completely to GitHub issues or keep > > > using > > > > > > Jira. Two issue trackers would not be used after trial period. > > > > > > > > > > > > > > > > > > > > >
