Closed gouttegd closed 3 years ago
Think 3 years is a good amount of time :)
Shall we automatically close existing issues that have not been updated in the past 3 years, as we have recently done for Uberon?
If so:
* [ ] Shall an editor do a pass on those issues before we trigger the closing script, to protect some worthy issues from auto-closing (again similar to what we have done for Uberon)? * [ ] Should we use the same age threshold as for Uberon (3 years)?
For information, as of today there are 182 issues that have not seen any update in the last 3 years.
I had a quick glance at some of those 182 tickets. At least some of them look worthy of consideration before auto-closing en masse. E.g., several tickets look neuron-related, so may be in scope for specific funded projects. Other may be in scope for the Human Cell Atlas. Others still may have been addressed, or out of date, or ok to close anyway. If we're willing/able to put in a bit of our time, we could split say the 100 oldest tickets among as many of us as possible and take a quick look? Even 5-10 minutes per ticket would make a difference, even if we can't fully address all requests. Then apply Damien's workflow on what's left. And repeat once per quarter?
Happy to do so :) I've seen some of the neuro ones and think they can probably be closed (or maybe some discussion continuing that imo is secondary) - havent done so just cause if we are going to implement the above, it might be better use the automated thing to ping the people assigned on them to see if they have anything else to add/change. Happy to split tickets and go through instead :)
(...) it might be better use the automated thing to ping the people assigned on them to see if they have anything else to add/change.
Unfortunately, most old tickets are unassigned, and the "creator" is mostly GoogleCodeExporter (or Chris when he migrated issues from SourceForge). So the automated message will go, broadly, to anyone monitoring the CL tracker. But that's probably ok - we're cleaning up the tracker a little anyway. :-)
We are now at 117 issues that would be closed with a threshold set at 3 years.
I propose to give us one more month to check some more of those issues (and either fix them directly or at least tag them or comment on them — anything to mark them as "active"), and then by the end of October I unleash the auto-closing script.
Would everyone be happy with that?
For reference, here are the issues that would be closed as of today:
Is adding a new label enough to spare from closing? There are multiple kidney related items that should be addressed as part of the KPMP and HuBMAP project work. I am adding the 'KPMP' label to these.
Yes, adding a label counts as an "activity" from the point of view of the auto-closing script, so it's enough to prevent auto-closing.
We are down to 53 old issues inactive for the past three years (well done !)
Unless anybody objects I will forcefully close them on Monday, as agreed above.
44 old issues have now been closed. They are all tagged with the autoclosed-unfixed
label.
Thanks to those who contributed to cut down the number of inactive issues by 75% (down from 182 issues when this ticket was opened) in the last two months!
Now that we have an automated procedure to regularly flag and close stale issues, there will be no more need for mass closures like this one.
From the CL meeting on August 18th, 2021:
Shall we automatically close existing issues that have not been updated in the past 3 years, as we have recently done for Uberon?
If so:
For information, as of today there are 182 issues that have not seen any update in the last 3 years.
CL editors who were not present in the August 18th meeting are invited to weigh in.