Would you consider using a Github app such as Stale?

The project has quite a large number of issues, many of which may be low priority or require more information and will go stale. Other large projects use apps to automate closing inactive issues to help with this.

Would this be something worth considering or maybe it has already been discussed and ruled out at this point.

[original thread by Gareth Whittaker]

No, we discussed and tried it and it makes more harm by closing relevant issues and force us to review such issues when we have no time.

[Gareth Whittaker]

Yes, I thought that might be the case. Thanks.

We’ve talked about it in the past https://github.com/eclipse-theia/theia/issues/3563, and use it in theia-apps but as Anton mentioned it might cause more harm and overhead in the main repository.

[Gareth Whittaker]

There are a little over 300 issues that haven’t been updated for a year or more. I guess it all depends on the preferred approach, as it could be determined that all of these should be closed even if they are relevant because their lack of activity suggests they aren’t high enough priority to address.

[Gareth Whittaker]

[Gareth Whittaker]

I was interested to see how the issue count fluctuated over time, but Insights only allowed me to go back one month. Looking at this date range I see closed vs new issues have actually been pretty even.

I try when I can to revisit old issues and see if they can be successfully closed:

  • if they are reflecting problems that no longer exist

  • if they are issues that have been addressed but not closed

A lot of the issues are still enhancements and items we want to be tracking. Just because they are old does not mean that we do not want them resolved at some point, and we fear that if we close them, we will somehow forget about them and never work on them.

One can sort issues by oldest and review some of them each day. But usually it is hard to miss important issues, such issues get commented and upvoted.