The backlog is our usual problem; the number keeps increasing but it is true that we don't have a better idea of how well/bad we are doing.
A metric sounds like a good idea and putting in the report every quarter will help us have a permanent trace. Percentage of open pull requests after three months is a good one. Another could be the number commits from non-committers per quarter [1] which would give something like the following: Q1 2017:44 Q2 2017:43 Q3 2017:27 Q4 2017:36 Q1 2018:39 Q2 2018:49 Q3 2018:46 Q4 2018:45 Q1 2019:43 Q2 2019:53 Q3 2019:77 Q4 2019:87 Q1 2020:36 Q2 2020:52 Q3 2020:37 Q4 2020:1 [1] https://gist.github.com/zabetak/413cf1538ca9b026622e62596e094f6e On Tue, Oct 6, 2020 at 8:37 PM Julian Hyde <jh...@apache.org> wrote: > The report looks good. > > The backlog of pull requests continues to be a concern. I think we > should track a metric so we know how we are doing, and strive to > improve it. How about "percentage of pull requests that are open after > three months"? > > Julian > > On Tue, Oct 6, 2020 at 7:37 AM Stamatis Zampetakis <zabe...@gmail.com> > wrote: > > > > Thanks for the feedback guys! > > > > @Vladimir: Indeed the sentence does not make sense :) > > It is a residual from another paragraph that I already removed. > > Thanks for catching that. > > > > On Tue, Oct 6, 2020 at 10:54 AM Vladimir Sitnikov < > > sitnikov.vladi...@gmail.com> wrote: > > > > > Stamatis> A small decrease > > > Stamatis> in closed issues can be attributed to those opened after > applying > > > static > > > Stamatis> code analysis frameworks > > > > > > Frankly speaking, it sounds puzzling. > > > It is not clear how "opening new issues" might result in "decrease in > > > closed issues". > > > > > > Vladimir > > > >
