I would be -1 on removing committer status due to inactivity, but whatever we can do to encourage continued contributions is good. Thanks for the tag (and everything else you've been doing)! This is a poor excuse, but I've still made use of a fairly limited portion of Calcite's code base so I find most PRs are in code I'm really not comfortable with reviewing. As I work on the notebooks I posted earlier, I'm hoping to expand my familiarity so I'll be more comfortable with some other pieces.
-- Michael Mior [email protected] Le mar. 25 déc. 2018 à 16:19, Francis Chuang <[email protected]> a écrit : > Julian, thanks for doing the legwork to get those issues tagged. I agree > that we need to keep up with the pull requests in order to encourage > contributors to contribute to the project. > > Stamatis, it's Apache policy that committer status never expires. I > think some projects turn inactive committers into emeritus committers, > but I don't think it's common. See > https://www.apache.org/dev/committers.html#committer-set-term > > On 25/12/2018 2:35 am, Stamatis Zampetakis wrote: > > The main problem seems to be the fact that there are not enough active > > committers. > > > > Between Calcite 1.17.0 and Calcite 1.18.0 the stats regarding the reviews > > and patches are as follows: > > Committer Reviews Insertions(+) Deletions(-) Modifications > > Julian Hyde 52 9039 2304 11343 > > Vladimir Sitnikov 19 865 207 1072 > > Jesus Camacho Rodriguez 7 657 292 949 > > Andrei Sereda 1 472 38 510 > > Michael Mior 3 284 18 302 > > Volodymyr Vysotskyi 1 52 19 71 > > Grand Total 83 11369 2878 14247 > > > > Unfortunately, there is no easy way to fix this. The only measure that > > could engage committers is to adopting a policy for removing inactive > > members (I do not know if there is already one in place). It can be > > something very light, just to ensure a minimum interaction with the > > project. It will not dramatically increase the previous numbers but if > > every committer did at least one review there wouldn't be so many open > pull > > requests and the work load of people in the above list could be > > significantly smaller. > > > > > > > > Στις Κυρ, 23 Δεκ 2018 στις 1:52 π.μ., ο/η Julian Hyde <[email protected]> > > έγραψε: > > > >> I went through JIRA and added the label "pull-request-available" to > >> about 30 PRs that had been marked "fix for 1.18" but we didn't get > >> to[1]. > >> > >> I suggest that we start using the "pull-request-available" label more > >> often. > >> > >> But my bigger point is that we letting our community down when we have > >> such a large backlog of un-merged patches. The committers of this > >> project, collectively, are not staying on top of their work load. > >> > >> Personally, I have been doing as much as I can, including taking on > >> the release manager role when no one else stepped up, and personally > >> committing a very large fraction of the 200+ commits in this release. > >> Maintaining Calcite is NOT part of my responsibilities at my day-job, > >> so I am doing this all in my personal time. It makes me angry that I > >> am shouldering this much of the responsibility. Something needs to > >> change around here. > >> > >> Julian > >> > >> [1] > >> > https://issues.apache.org/jira/issues/?jql=project%20%3D%20CALCITE%20and%20labels%20%3D%20pull-request-available > >> > >
