Archives
Contributors
contributors@odoo-community.org
Re: Reciprocity in PR opening vs reviews; banning contributors
Re: Reciprocity in PR opening vs reviews; banning contributors
Re: Reciprocity in PR opening vs reviews; banning contributors
Hello!
I like the idea of using statistics to show the path we expect contributors to follow. However, I don’t think we should ban or limit users who don’t follow that path.
In my experience, newcomers rarely meet these expectations at the beginning. It often takes around a year before people start collaborating the way we would like (doing reviews, participating in discussions, etc.). Some people get there faster, others slower, but we shouldn’t limit them because of that.
Obviously, we can enforce rules like a 2 reviews : 1 PR ratio for PSCs or maintainers (and IMO, we should), but not for everyone.
Collaboration data is very interesting, but relying only on data can be counterproductive and, in some cases, unfair. Reviews and contributions are hard to evaluate properly. Should we ban someone who does 3 high-quality reviews (with thoughtful comments and valuable points) and 10 fix PRs (including migrations), while promoting someone who does 10 reviews with no comments and 1 PR that creates a lot of extra work for others due to their way of working?
This kind of data can be useful when looking at large numbers and trends, but with smaller samples, it’s easy to draw the wrong conclusions. IMO, using it can be a double-edged sword.
So if we decide to use this data, I think it should be for promotion and visibility, not for banning contributors.
My 2 cents.
El mar, 27 ene 2026 a las 9:12, Tom Blauwendraat (<notifications@odoo-community.org>) escribió:
On 1/27/26 8:57 AM, Jairo Llopis wrote: > IMHO statistics should be used as a prize, not as a weapon. Agreed! And it still could achieve the same: if a PR or contributor is valued very highly, his/her PR's float to the top of the sorting/filtering of reviewers anyway. And also the message to the contributors is clear: *this* is how you can get a higher ranking. So it still might achieve the same. Holger, what do you make of this inverted carrot approach rather than a stick?_______________________________________________
Mailing-List: https://odoo-community.org/groups/contributors-15
Post to: mailto:contributors@odoo-community.org
Unsubscribe: https://odoo-community.org/groups?unsubscribe
Enric Tobella Alomar
CEO & Founder
by Enric Tobella Alomar - 09:35 - 27 Jan 2026
Reference
-
Reciprocity in PR opening vs reviews; banning contributors
Hi all, on the OCA days, we discussed that the current situation with having way too much input for way too few reviewers is untenable. This has not improved since, quite the opposite. It's really hard to find the gems in the noise. Back then, I called for better automation for this, so here my proposal: Have a github action that counts lines of PRs somebody opened in a repo, vs the lines of PRs the person reviewed in that repo. Everyone must review at least twice as much as they submit. If after asking for more reviews, no reviews come, close the user's PRs automatically (in the repo, not all OCA) after some time. Also add a manual mechanism for banning users who try to cheat with bullshit reviews or otherwise undesirable behavior. PRs by banned users are closed automatically. I implemented both in https://github.com/hbrunn/social/blob/18.0/.github/workflows/reciprocity.yml resp https://github.com/hbrunn/social/blob/18.0/.github/workflows/ban.yml You can test this by creating PRs against my fork after cloning my version of the 18.0 branch. Banning works by adding a handle to a file .banned.txt in the repo's root. Before proposing this to oca-addons-repo-template, I'd like to hear some input from you. Best regards, Holger -- Your partner for the hard Odoo problems https://hunki-enterprises.com
by Holger Brunn - 08:56 - 23 Jan 2026