Hi
As of 2022 July, Chromium has 4531 OWNERS files containing 6850 names. These include inactive owners, which are one of the sources of slow code review latency. One year ago, we cleaned up inactive owners and removed ~500 inactive owners. I propose running the clean-up process again to keep the OWNERS files updated.
Specifically, a person is identified as an "inactive" owner iff:
The person didn't commit or review any CLs in the directory they own while there were 100+ CLs that touched the directory in the past 6 months (as of July 6, 2022).
Last year, I gave the inactive owners an option to flip the decision manually to stay as an owner, but for this cycle, I'm planning to remove the inactive owners unconditionally. The rationale is 1) if the person made no contribution on a very active directory for 6 months, it will be reasonable to say that the person is inactive, and 2) if there is any special reason for it and the person needs to stay as an owner, the person can show evidence that they are meeting the owners expectations and be readded through the standard OWNERS nomination process.
Specifically, people listed in this spreadsheet are identified as inactive owners and will be removed.
I understand this is a tricky proposal. Having your name on OWNERS is an award for your previous amazing contributions, and I understand your feeling about your name being removed. However, I think it's important to keep the OWNERS files updated so that Chromium developers can find active owners and improve the code review latency.
If you have any questions / concerns, please let me know. Thanks!
----
You received this message because you are subscribed to the Google Groups "blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABg10jyArLjDp0ixPu%2BCSZ9NVrn0M1GwNFiJqiPGRE1f0mrbfQ%40mail.gmail.com.
I'm worried that this process excludes/penalizes folks who may be OOO for extended leave (incl long stretches of parental leave, bereavement) and have that in their Gerrit status. This should not be a source of review latency, if it is Gerrit should better surface that they are OOO.Are any of the inactive owners, who did opt out last time, a source of review latency? I.e. are reviews assigned to them but they don't review them within some SLO window? Otherwise I strongly suggest we let folks decline the OWNERS removal (at other OWNERS' discretion who should probably review removal CLs).
----
--
Chromium Developers mailing list: chromi...@chromium.org
View archives, change email options, or unsubscribe:
http://groups.google.com/a/chromium.org/group/chromium-dev
---
You received this message because you are subscribed to the Google Groups "Chromium-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chromium-dev...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CAHtyhaTNC4tgQbqbUq%2BQdFfcORr3aFobjgbeE%2BTaVf7eDgU2Bg%40mail.gmail.com.
--
Chromium Developers mailing list: chromi...@chromium.org
View archives, change email options, or unsubscribe:
http://groups.google.com/a/chromium.org/group/chromium-dev
---
You received this message because you are subscribed to the Google Groups "Chromium-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chromium-dev...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CAGFX3sFB9G8R2MyHT6rjVtEFRAKMeyCTH6Yu0DYqUOfLPCxCBw%40mail.gmail.com.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/0a2a01e2-652b-4e31-895c-f020e7b46358n%40chromium.org.
This list includes per-file owners, did the script look for 100 CLs in those files named by the rule when deciding to remove the person?
I'm worried that this process excludes/penalizes folks who may be OOO for extended leave (incl long stretches of parental leave, bereavement) and have that in their Gerrit status. This should not be a source of review latency, if it is Gerrit should better surface that they are OOO.
Are any of the inactive owners, who did opt out last time, a source of review latency? I.e. are reviews assigned to them but they don't review them within some SLO window? Otherwise I strongly suggest we let folks decline the OWNERS removal (at other OWNERS' discretion who should probably review removal CLs).
Maybe it would make more sense to identify OWNERS who are not active globally in chrome/, instead of owners not active in a particular directory? How common are OWNERS active in Chrome, but high latency only for specific directories?
I'm asking as someone who was recently inundated by auto-generated removal CLs, the majority of which did not make sense (admittedly, I believe it wasn't based on activity). The tool even seemed to want to remove all owners from some directories.
The data in the table seems off, what is considered a "review": is that a "Code Review +1" or is that any review comment?
I also have an edge case where I'm mostly interested in several files in a folder where other files are being changed more frequently, should I be optimizing OWNERS to list myself as per-file?
I recently tried a similar automated audit of inactive owners - I looked for anyone who hadn't reviewed or authored a CL in 12 months anywhere, regardless of activity in the directory and found (as list, Google internal only) many accounts that no longer exist (or perhaps never did) in OWNERS. It probably has different false positives than the proposed set above. Maybe the intersection of the two sets would be sensible?
Thanks all for the input!Dana:This list includes per-file owners, did the script look for 100 CLs in those files named by the rule when deciding to remove the person?Thanks for pointing this out! I'll exclude per-file owners from the list for now.Peter:I'm worried that this process excludes/penalizes folks who may be OOO for extended leave (incl long stretches of parental leave, bereavement) and have that in their Gerrit status. This should not be a source of review latency, if it is Gerrit should better surface that they are OOO.
Are any of the inactive owners, who did opt out last time, a source of review latency? I.e. are reviews assigned to them but they don't review them within some SLO window? Otherwise I strongly suggest we let folks decline the OWNERS removal (at other OWNERS' discretion who should probably review removal CLs).I think Glen covered this topic very well. As written in this guideline, owners are expected to be an active contributor to the directory ("Have the bandwidth to contribute to reviews in a timely manner. ... Don't try to discourage people from sending reviews, including writing “slow” or “emeritus” after your name."). If you are on an extended leave and removed by this process, you can explain it and re-add yourself through the owner nomination process. Will it work?
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABg10jyCNN1%3DpfT%3DCWPmc4%2Bi9PmGs-%3DbX9e2mUi2bHthF%2B0w-w%40mail.gmail.com.
On Thu, Jul 28, 2022 at 6:07 AM Kentaro Hara <har...@chromium.org> wrote:Thanks all for the input!Dana:This list includes per-file owners, did the script look for 100 CLs in those files named by the rule when deciding to remove the person?Thanks for pointing this out! I'll exclude per-file owners from the list for now.Peter:I'm worried that this process excludes/penalizes folks who may be OOO for extended leave (incl long stretches of parental leave, bereavement) and have that in their Gerrit status. This should not be a source of review latency, if it is Gerrit should better surface that they are OOO.
Are any of the inactive owners, who did opt out last time, a source of review latency? I.e. are reviews assigned to them but they don't review them within some SLO window? Otherwise I strongly suggest we let folks decline the OWNERS removal (at other OWNERS' discretion who should probably review removal CLs).I think Glen covered this topic very well. As written in this guideline, owners are expected to be an active contributor to the directory ("Have the bandwidth to contribute to reviews in a timely manner. ... Don't try to discourage people from sending reviews, including writing “slow” or “emeritus” after your name."). If you are on an extended leave and removed by this process, you can explain it and re-add yourself through the owner nomination process. Will it work?The next guideline (on removal of owners) explicitly excludes owners who are on leave. I don't think we should be adding additional friction for folks who go on leave; the default assumption should be that when they return, they are just as capable of being a good owner as when they left, without them having to re-nominate themselves.
Matt:Maybe it would make more sense to identify OWNERS who are not active globally in chrome/, instead of owners not active in a particular directory? How common are OWNERS active in Chrome, but high latency only for specific directories?My personal opinion is that owners who made no contributions globally in the past 6 months *or* owners who made no contribution to the directory they own while there were 100+ commits in the past 6 months can be identified as inactive owners.Note that this is not an irreversible process. When you have a reason, you can explain it and re-add yourself through the owner nomination process.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAEK7mvr5vjo0FdbYepkOLBDGJN7KLivmauf4G-HQ1rf67f7pSQ%40mail.gmail.com.
I agree with the policy to not remove owners because they are on leave. Owners plus OOO status should inform owner selection and we still need that because we shouldn't remove owners while they are out for a week but they sure shouldn't be used during that week.If OOO is an efficient signal this should be sufficient without force removing folks on leave. If OOO isn't a sufficient signal and introduces review latency we should look at why that is instead and fix our tooling.If I may spitball another reason for latency we should maybe also surface how long folks review queues are and perhaps use that to help guide owners selection.Also because I'm already up on my soapbox, if you're fast and not overloaded and want to help with the review load, please put something like "fast" or "more reviews plz" in your Gerrit status to encourage more reviews your way (me inspired by ellyjones@ here). Encourage others to do the same and reward this as a community contribution.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABg10jwB528AFfwArSHLKoHWeWt-JOvCS0dnFJNaDycKSc5GDw%40mail.gmail.com.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABg10jwB528AFfwArSHLKoHWeWt-JOvCS0dnFJNaDycKSc5GDw%40mail.gmail.com.
--
--
Chromium Developers mailing list: chromi...@chromium.org
View archives, change email options, or unsubscribe:
http://groups.google.com/a/chromium.org/group/chromium-dev
---
You received this message because you are subscribed to the Google Groups "Chromium-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chromium-dev...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CAHtyhaTNC4tgQbqbUq%2BQdFfcORr3aFobjgbeE%2BTaVf7eDgU2Bg%40mail.gmail.com.
Just for completeness, there are some OWNERS files that require a nomination process. At least third_party/blink/renderer/core/OWNERS requires that.
Maybe it would make more sense to identify OWNERS who are not active globally in chrome/, instead of owners not active in a particular directory? How common are OWNERS active in Chrome, but high latency only for specific directories?
I don't think that 85/6850 OWNERS entries are a large enough problem to bypass local owner approval for this (i.e. land it unconditionally). If some of them are owners in very active folders and have had a lot of reviews assigned to them, sure. This is about 1% of ownership entries if I'm understanding this correctly. How badly are they contributing to review latency in order to override "and they are not on a leave during that time" as a policy? I like this policy, and it's supposed to still be the policy.
I looked into the first 10 entries manually and there are multiple cases where I don't think it is correct to remove them.There are people who are on long term leave and are indicating this in their status message.I would be quite annoyed if I had to figure out all the files I was owner for and restore this when I come back from a leave.Could you check for each person you want to remove if the status message has an indication that the OWNER is only temporarily inactive?There are also otherwise active owners who have a comment that limits their ownership to a subset of files.I wonder why it would be useful to remove people who are otherwise active just because they haven't reviewed files in a certain directory?There are also per-file owners but I guess the list just isn't updated yet?As an alternative to implementing the suggestions above:How about sending CLs with autosubmit enabled for each removal to all owners of the corresponding file and see if they agree with the removal or not?
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CABg10jzMeS604sff1NNK4BK9E3A8A7rn3wxPiqB0G95%3D3xupgA%40mail.gmail.com.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CABg10jzMeS604sff1NNK4BK9E3A8A7rn3wxPiqB0G95%3D3xupgA%40mail.gmail.com.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CABg10jzMeS604sff1NNK4BK9E3A8A7rn3wxPiqB0G95%3D3xupgA%40mail.gmail.com.
I looked into the first 10 entries manually and there are multiple cases where I don't think it is correct to remove them.There are people who are on long term leave and are indicating this in their status message.I would be quite annoyed if I had to figure out all the files I was owner for and restore this when I come back from a leave.Could you check for each person you want to remove if the status message has an indication that the OWNER is only temporarily inactive?There are also otherwise active owners who have a comment that limits their ownership to a subset of files.I wonder why it would be useful to remove people who are otherwise active just because they haven't reviewed files in a certain directory?There are also per-file owners but I guess the list just isn't updated yet?As an alternative to implementing the suggestions above:How about sending CLs with autosubmit enabled for each removal to all owners of the corresponding file and see if they agree with the removal or not?
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CAKAfQi2EPsAZ0FNGE%3DOTqd1ubradPHO%2Bi8q8S9hbrJzSZaD4MA%40mail.gmail.com.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CA%2B4SKEKJsKET9S7LgyvhEoCBV85ynahAy6WeuB5pM7DEG_nB5w%40mail.gmail.com.