| *** benj_80 is now known as benj_8 | 10:49 | |
| opendevreview | Monty Taylor proposed openstack/project-config master: Add cargo-siblings repo https://review.opendev.org/c/openstack/project-config/+/1011914 | 17:43 |
|---|---|---|
| opendevreview | Merged openstack/project-config master: Add cargo-siblings repo https://review.opendev.org/c/openstack/project-config/+/1011914 | 18:56 |
| JayF | Trying to gather information for the "make github stop saying openstack doesn't have a security policy" task. I was wondering if we had any community-commons tools for file syncing; e.g. if we did end up with a SECURITY.rst file in a shared repo we wanted to copy through to all repos, is there any automagic to do this? | 21:07 |
| clarkb | the closest thing that comes to mind is the old requirements syncing which was for ongoing syncing | 21:11 |
| clarkb | you can probably ifnd the historical remnants of that in the requirements repo history? I don't think we sync those anymore though | 21:11 |
| JayF | ack; I don't really think I'd *want* that unless it already existed | 21:11 |
| JayF | like "add a file to this list" == yes please | 21:11 |
| clarkb | if it is a one time thing it might be easiest to use the gerrit api to create changes with file contents | 21:12 |
| JayF | otherwise I hope there are other options :) | 21:12 |
| clarkb | then you don't have to clone anything | 21:12 |
| JayF | that is likely the route I'd go | 21:12 |
| JayF | or something shaped like that | 21:12 |
| JayF | but right now I'm just gathering info | 21:12 |
| JayF | (and filing a "support SECURITY.rst files, dammit" issue with github) | 21:12 |
| clarkb | https://review.opendev.org/Documentation/rest-api-changes.html#create-change may be helpful | 21:13 |
| clarkb | looks like you may need a followup api call to actually add the content? | 21:14 |
| JayF | I'll cross that bridge when I get there; I'm just putting together the menu of options to help decide :) | 21:14 |
| clarkb | https://review.opendev.org/Documentation/rest-api-changes.html#apply-patch for that I think | 21:15 |
| clarkb | the other option is git clone and make a change and push it using a more human like workflow | 21:16 |
| clarkb | This approach may be easier if you're editing existing content as it will provide all the context to work against. But for a new file I think the rest api may be easy | 21:16 |
| mikal | clarkb: these three patch series might be of interest to you -- https://review.opendev.org/c/openstack/devstack/+/1011794; https://review.opendev.org/c/openstack/oslo.privsep/+/1011795; https://review.opendev.org/c/openstack/neutron/+/1011797. | 21:45 |
| JayF | mikal: I swear I didn't open all those just to put a nit on the commit message /o\ it looks good though | 21:51 |
| JayF | oh, I should hit the whole chain | 21:51 |
| mikal | JayF: if your nit is about Co-Authored-By, I'd be happy to remove it once you provide a link to a clear written policy statement on the matter and explain how that policy has been communicated. That is -- it seems like a decision has been made but I don't feel like its rationale or enforcement has really been explained to me. | 21:54 |
| JayF | I don't know if there is one, and I didn't -1 because those aren't repos I am core on. If it was Ironic, the policy would be "as a core I want only humans to be seen as responsible for commits to this repo". I'm not the only one who thinks it's a bad thing though: https://meetings.opendev.org/irclogs/%23openstack-tc/%23openstack-tc.2026-10-22.log.html#openstack-tc.2026-10-22.log.html#t2026-10-22T16:44:02 | 21:56 |
| JayF | oh, that whole chat was started by one of these changes | 21:56 |
| JayF | I didn't realize that, it's been a WHIRLWIND two weeks | 21:56 |
| mikal | Oh, you're all entitled to your opinions. I just think its super weird that people would NACK patches for a thing which doesn't actually appear to exist as a written down big boy policy. | 21:57 |
| JayF | The policies are written by the board, which is not exactly technical governance on the project. Like I said, my POV has always been from that of a core reviewer setting a bar. | 21:58 |
| mikal | Surely the TC debating what the project wants to do, voting on that, and then disseminating that decision should come before enforcement? | 21:58 |
| mikal | You can set whatever bar you want, but I guess my counter-stance is that as a donor of my time, I want a clear set of consistent expectations because I don't want to have to jump through arbitrary and varying hoops when all I am trying to do is address a concern clarkb raised. The other obvious option is to simply ignore that concern and go back to | 22:00 |
| mikal | watching Battlestar Galactica re-runs. | 22:00 |
| JayF | I enforce lots of things as a core reviewer that I can't draw back directly to a policy. In my mind, keeping LLMs out of authorship is supportive of our Open Community which indicates that building community between developers, users, and ecosystems have value. | 22:00 |
| mikal | That is, I would use the adoption of the DCO as a counter example. It was debated, ageed, communicated, and now everyone knows what the expectation is. | 22:00 |
| JayF | This is a conversation to -tc, I made a review comment without a -1 attached. It's advisory. I'm not core on those repos. | 22:00 |
| JayF | #openstack-tc -- not here, I mean | 22:01 |
| JayF | I'm also not on the TC either. I'm just a man making his way through the world lol | 22:01 |
| mikal | That's fair. And I am just a hobbyist contributor who has been rebasing the same patches in an ever growing set for a year while the only engagement I get from reviewers is to nit pick the formatting of commit messages. So hopefully you can see that sometimes that's just a little bit frustrating. | 22:02 |
| mikal | Honestly I think one of OpenStack's core problems is that its processes were built for a community and rate of change at least ten times bigger than now, and that eventually someone needs to accept that they need to scale back to meet the resources actually available. | 22:03 |
| JayF | If you ever have that experience on a repo I am core on, please reach out directly long before it gets to a year :) | 22:03 |
| *** jonher_ is now known as jonher | 22:09 | |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!