Tuesday, 2026-08-04

gouthamrtc-members: a gentle reminder that our weekly IRC meeting will be hosted here in ~53 minutes. Agenda is here: https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Agenda16:07
bauzasgouthamr: I'm fried with meetings back-to-back so permission to skip16:24
gouthamrack bauzas, ty for letting me know16:27
dansmithI'm here, minute late checking in16:59
gouthamr#startmeeting tc17:01
opendevmeetMeeting started Tue Aug  4 17:01:18 2026 UTC and is due to finish in 60 minutes.  The chair is gouthamr. Information about MeetBot at http://wiki.debian.org/MeetBot.17:01
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.17:01
opendevmeetThe meeting name has been set to 'tc'17:01
gouthamrWelcome to the weekly meeting of the OpenStack Technical Committee. A reminder17:01
gouthamrthat this meeting is held under the OpenInfra Code of Conduct available at17:01
gouthamrhttps://openinfra.dev/legal/code-of-conduct.17:01
gouthamrToday's meeting agenda can be found at17:01
gouthamrhttps://wiki.openstack.org/wiki/Meetings/TechnicalCommittee17:01
gouthamr#topic Roll Call17:01
noonedeadpunkIn the airport right now, so can t participate, sorry17:01
gouthamrnoonedeadpunk: ack17:01
spotz[m]o/17:01
frickler\o17:01
mharley[m]\o/17:02
gouthamrnoted absence: noonedeadpunk ba u zas 17:02
gouthamrcourtesy-ping: cardoe, mnasiadka17:02
dansmitho/17:02
cardoeo/17:02
mnasiadkao/17:02
fungiahoy!17:02
gouthamrgreat, good crowd.. let's get started17:03
gouthamr#topic Last Week's Action Items17:03
gouthamrbarbican-ui's revival seems to have stalled? or at least i'm not sure of the current state17:04
gouthamr#link https://review.opendev.org/q/project:openstack/barbican-ui+status:open 17:04
gouthamrwe'll probably need to revisit that.. my fear is that lack of clear direction is off-putting to new/existing contributors 17:05
gouthamrwill look for updates outside of this meeting and reprise this action item next week17:05
gouthamrfrickler proposed new changes to the proposal to extend AC status automatically to +/-2 +W reviewers17:06
gouthamr#link https://review.opendev.org/c/openstack/governance/+/995535 (Reviewers as Active Contributors)17:06
gouthamrthis needs roll call votes please, if you haven't already looked.. 17:07
gouthamrit hits the 7-day mark tomorrow, and i'd like to merge it promptly because it's been widely discussed 17:08
fungiespecially since this is veering close to making electorate decisions right before nominations start which rely on the definition in order to qualify candidates17:08
gouthamr+117:09
fungi(nominations begin tomorrow)17:09
gouthamryes.. we have ~three weeks before the email deadline, and we formalize the electorate17:09
fungiodds are any candidates will not need to rely on the new rule in order to be verified, but it's not guaranteed17:10
gouthamryeah, super rare to see this.. but it has happened, that led to us working on this 17:11
fungi(for ptl nominees i mean, tc nominees aren't constrained by being contributors)17:11
gouthamrbut the TC electorate is.. 17:11
gouthamr"The electorate is requested to confirm their email address in gerrit, review.opendev.org > Settings > Contact Information > Preferred Email, prior to Aug 19, 2026 00:00 UTC so that the emailed ballots are mailed to the correct email address."17:11
gouthamr^ i'd take this as the deadline17:11
fungiright, thankfully that doesn't come into play for a few weeks as you noted17:11
gouthamrthe deadlines are on: https://governance.openstack.org/election/ and on the Indri release calendar for reference17:12
gouthamrspeaking of, the release team has published the calendar for 2027.1 "Indri":17:12
gouthamr#link https://releases.openstack.org/indri/schedule.html (2027.1 Indri Release Schedule)17:12
fungibut just to be absolutely clear, the tc is going to be making a change to the rules for ptl candidacy during the nomination period17:13
mharley[m]And which change would be, fungi ?  Sorry if it's an obvious question.17:14
gouthamryes, through our weekly emails, I've been communicating that this is what we want for weeks now though.. somewhat using the same verbiage :) 17:14
fungimharley[m]: that being an active core reviewer qualifies you to be a ptl candidate17:14
mharley[m]Gotcha.17:14
mharley[m]I remember that discussion.17:14
fungi(even if you didn't have any of your own changes merge)17:15
mharley[m]This is a big turning point. :-)17:15
gouthamri'll ask project teams to start adding their own internal deadlines to the release calendar.. although some of them might wait until the PTG to set those17:16
gouthamrconsider the runtime update in your reviews, because we'll be quickly planning capacity/CI pipeline changes etc to accommodate it:17:17
gouthamr#link https://review.opendev.org/c/openstack/governance/+/996925 (Define testing runtime for 2027.1 release)17:17
gouthamrnext action item, security liaisons17:17
gouthamrwe merged the change that added security liaisons to most teams: 17:18
gouthamr#link https://review.opendev.org/c/openstack/governance/+/996563 (Seed security liaison data from VMT wiki)17:18
gouthamrthe commit message is no longer accurate, because PTLs/maintainers overwhelmingly participated and updated liaisons we took from the wiki17:19
gouthamrwe have a few that didn't, and we'll default to using the PTL as the security contact there.. 17:19
gouthamra related concern that was raised in this meeting was to clean up the telemetry team's core security contacts on launchpad17:20
gouthamrthat was done as well, after the meeting17:20
gouthamr#link : https://launchpad.net/~ceilometer-coresec/+members#active (Telemetry Coresec)17:20
gouthamrname still says "ceilometer-coresec"; but the VMT is aware what projects fall under the telemetry umbrella and who to look to for help with triage/handling of vulnerabilities17:21
gouthamri have a couple of action items regarding the TC vision, and the CEWG.. I don't have interesting updates to share.. these got the slow-burner treatment, and i'll catch up this week17:23
gouthamrthat's all the action items i see, was anyone working on anything else to note here/today?17:24
gouthamralright, let's move to our next topic17:26
gouthamr#topic Bridging the gap effort updates (fungi/ildikov)17:26
ildikovo/17:26
* gouthamr waves at ildikov fungi 17:26
gouthamrwelcome back, at this point, you've done a round of updates with project teams, and it's a nice juncture to fill us in on insights, and any action items the TC can take :) 17:27
ildikovhappy to be here in good company :)17:27
fungimaybe a good intro to this is that my main priority has really been on the long-term sustainability of openstack17:27
ildikov@fungi will give a brief update on the team outreach we've been doing and then I'll touch on some plans we're having17:28
ildikov+117:28
fungiconcerns of the usual maintainer burn-out we've always faced, while maintaining a pipeline of new maintainers as well17:28
fungiwith that in mind, i've reached out to the project teams with the most survey responses (whether maintainer, contributor, or both)17:29
fungidropping into a couple of their regular irc meetings to present team-specific analysis of the survey results and also review system metrics17:29
fungiwith a goal of finding ways to make the experience more positive for maintainers (so they're more likely to stick around) and also contributors (so they might want to become maintainers)17:30
fungipreviously we'd identified communicaion gaps around review and development prioritization, and added some questions to surveys in an attempt to isolate the confusion and find ways to better coordinate17:31
fungithat's seemed to help in at least some cases where teams have focused on improving how they set their priorities and make them discoverable17:32
fungiin this last round (for the gazpacho retrospective) we added new questions about ai/llm use, because there were concerns emerging about how this is altering the way our community interacts and how we collaborate with one another17:33
fungias suspected, we did see quite a bit of llm reliance already both for people developing new changes and also to assist with reviewing changes17:33
fungiit seemed like drafting test cases and analyzing test failures were an especially popular use17:34
fungifeedback from the teams in these outreach sessions has been positive on the whole17:35
fungiand some of the changes teams have been implementing seem to have shown improvements from one cycle to the next (though it's still hard to know with the small sample size and relatively short duration we've been doing these so far)17:36
fungiparticularly things like priority dashboards, scheduled check-ins among maintainers and activities like "reviewathons"17:37
fungianyway, i want to give ildikov an opportunity to talk about future plans, but we can hopefully take a few minutes to answer questions after that, if there are any17:38
gouthamr(the goals seem great)17:38
ildikov+1 thanks fungi, this was an awesome summary!17:39
ildikovas we’re getting to the end of the team outreach period after this last round, we’re putting together another set of improvement ideas and will share them with the community as soon as we can17:39
ildikovwe'll likely highlight these changes that fungi mentioned here as well to make sure good habits have a chance to spread17:40
gouthamr> feedback from the teams in these outreach sessions has been positive on the whole17:40
gouthamr(good to know) 17:40
gouthamr> and some of the changes teams have been implementing seem to have shown improvements from one cycle to the next (though it's still hard to know with the small sample size and relatively short duration we've been doing these so far)17:40
gouthamr(i'm itching to know the specifics here) 17:40
ildikovjust a reminder if folks missed it, we also share aggregated data twice a year17:40
ildikovMetrics analysis from the last round: https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/message/CVLQIFEVNNJJ7BJSFZJ4RNZZA7ELQKDS/17:41
ildikovSurvey analysis: https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/message/3POML6PYPDWQNYM5IFB54O4ASXEPK4Z5/17:41
ildikovWe understand that these emails are long and not always easy to digest, so we're looking into visualizing the data to make trends easier to follow, now that we have some.17:41
ildikovWe are also looking into further adjusting the surveys, including:17:42
ildikovAssessing maintainer experience a bit more directly, to understand a bit better how people navigate their current load and responsibilities, and how taxing these might be17:42
ildikovUp until now we've focused mainly on challenges both maintainers and contributors face, which is only one piece of the puzzle17:42
ildikovSome teams have mentioned that they’re planning to be more intentional about extending their maintainer group, so we’re considering to include a question in the contributor survey to try and assess the interest folks might have to grow into this role17:43
ildikovWe are also trying to finalize the adjustments to the survey questions as soon as we can, and try to open them earlier in the release cycle to get better engagement17:43
ildikovthis is most of it in a nutshell17:44
gouthamr++ now's a good time for instance for the next survey, or as early as we can.. we can wait to analyze, but, people are probably going through some of the maintainer/contributor pain now and may be willing to share.. 17:45
ildikovI expect fungi and I will work closely with folks in the new CEWG onwards as well17:45
ildikovgouthamr: +117:45
fungiyes, though i do think that focusing on contributor experience is potentially missing the opportunity to make things better/easier for maintainers (though as noted it can provide a necessary pipeline to adding maintainers, which does ease things for them in the long run)17:46
fungiso from the btg side of things i do still want to make sure we try to better understand what maintainers are going through and help them find ways to cope17:47
gouthamrabsolutely, and i'm glad to have this two pronged approach.. 17:47
fricklerbtg?17:47
fungibecause when maintainers are happy and efficient, contributors will have a more positive experience too17:47
ildikov+1, the maintainer survey will get a little bit longer, however, we're also working on making sure it doesn't get out of control17:48
fungifrickler: we had originally dubbed the effort "bridging the gap between community and contributing organizations"17:48
fricklerah, etoomanyabbrvs17:48
spotz[m]2 pronged is the best path forward17:49
fungi(because representatives of foundation member orgs were providing feedback/examples on business development calls about how much trouble their employees were having getting their changes to merge to projects)17:49
gouthamrhas that part continued? 17:50
gouthamri.e., member orgs reaching out to the board with examples of pain points17:51
fungito some extent yes, we were taking those examples as opportunities to identify where the inevitable communication breakdowns were occurring in each case and find commonalities between the cases17:51
fungiin many of those it was a mix of contributors not knowing how to navigate the community dynamics and maintainers being too busy to really give them enough guidance to overcome that17:52
fungiwhich is why we've been trying to focus on education and documenting things like priorities17:53
gouthamrack, 17:53
spotz[m]I know there's the new WG but a lot of the not knowing what to do could be found and updated in the old OUI repo17:53
gouthamrack spotz[m].. i don't know what content this is, but good call out for me check and ask questions17:54
spotz[m]We used to cover governance, project structure, and the git and gerrit stuff17:55
gouthamri see17:55
ildikovI think it's a topic for the CEWG, but one thing to prepare for in the new AI era is to have content that is somewhat for AI agents to pick up17:56
ildikovas many people tend to ask AI first about what and how to do17:56
gouthamr^ an important topic, and has been somewhat been the focus of the TaCT SIG17:56
fungiprobably with the goal of convincing those agents not to overwhelm the limited maintainer bandwidth in our projects17:56
ildikov+117:57
JayFildikov: I would suggest that asking humans to accomodate AI agents more and more is a source, not a solution, for burnout.17:57
gouthamrtimecheck: we're nearly at the hour17:57
gouthamrmharley[m]: do you mind sharing a blurb of your discussion to the ML perhaps? and we can chat here next week?17:57
gouthamror, if you'd like, this channel loves async conversations as you can tell (like everyone else working on openstack)17:58
fungiyes, in my opinion we should be focusing on giving contributors the guidance they need to do more things for themselves before expecting to have changes reviewed, paying attention to team priorities so they don't push changes that are unlikely to get a response at all, and helping relieve the maintainer burden wherever they can17:58
mharley[m]If I knew one single topic would take almost one hour, I'd put mine before it. :-P 17:58
ildikovJayF: I understand, and I'm sorry it came through that way. I just meant that comment about consideration when updating the now somewhat outdated content. Or reconsider what to update and what to remove in some cases.17:58
mharley[m]So, my intention is to start drawing attention towards the PQC subject, and I wanted to communicate a quick status on it.17:59
mharley[m]But I have an appointment now.17:59
gouthamrsure, come back and do it whenever on this channel.. or if it can wait until next week, good.. i'll put this up before anything else on the agenda17:59
JayFildikov: there's just a lot of talk around what an active contributor is. Beyond anything else: they are humans. Contributor experience working groups should ensure they are focusing on the human over the agent IMO.18:00
gouthamrwe're at the hour.. 18:00
gouthamrfor ildikov/fungi - i can see next steps being the analysis for this current survey, asking new questions and opening up the new survey.. driving improvements for contributors through the ML and the CEWG.. but, i'd like to table more maintainers' pain points discussion for here, perhaps in a week or two.. 18:00
JayFmharley[m]: a post to openstack-discuss, targetted to [security-sig], might be a good venue for that as well18:00
gouthamr^ +118:01
mharley[m]Thanks, JayF .  So, next week, gouthamr ?18:01
gouthamrack mharley[m] 18:01
mharley[m]Do I have to edit again the wiki page to include the topic?18:02
gouthamrand ty fungi/ildikov for your hard work, and the recap here18:02
gouthamrmharley[m]: i'll handle that18:02
gouthamrwe're past the hour.. anything else to add to the minutes today?18:02
mharley[m]Cool, thanks.18:02
ildikovJayF: I completely agree. But how can we focus on the humans without taking into account the tools they rely on?18:03
gouthamralright, thank you all for participating.. i'll wrap it up here, we can keep the conversation going!18:03
gouthamr#endmeeting18:03
opendevmeetMeeting ended Tue Aug  4 18:03:53 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)18:03
opendevmeetMinutes:        https://meetings.opendev.org/meetings/tc/2026/tc.2026-08-04-17.01.html18:03
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/tc/2026/tc.2026-08-04-17.01.txt18:03
opendevmeetLog:            https://meetings.opendev.org/meetings/tc/2026/tc.2026-08-04-17.01.log.html18:03
JayFildikov: I think that's a very fine line, especially since some folks are abusing open source projects by sending the agents off autonomously. As an OSS project, I'd rather focus directly on the human interactions. I would posit that a push to facilitate agents, even under the guise of helping humans with the tools they are using, could be seen by some -- I would certainly see it this way -- as the foundation trying to promote the use of 18:05
JayFLLM toolin.18:05
JayFand while we should facilitate whatever tools folks want to do for their work; I have seen multiple times -- in and out of this community -- more LLM be suggested as the solution to LLM-induced (or at least accellerated) burnout18:06
dansmithyeah I really hate that18:53
dansmiththey seem to think LLMs have generated a problem of quantity and only LLMs could be the solution18:53
dansmithwhen really the LLMs have mostly made one thing cheap and everything else much more expensive18:54
ildikovI understand, and I definitely did not mean to suggest to throw more LLMs at the problem19:17
gouthamri don't see what ildikov said as "promote the use of LLM tooling", but not debating your interpretation as well.. I wanted to make sure her views aren't supposed to be construed as the foundation trying to do something too..19:17
ildikovgouthamr: I appreciate that19:18
dansmithI dunno who is responding to who, I was just reacting to JayF's comments :)20:11
gouthamrack, me too, and likely ildikov was too.. 20:19
* gouthamr cues usual notes about IRC not relaying emotions like natural dialog does 20:19
JayFI mainly just wanted to be sure we are watering our crops with water and not BRAWNDO! :)20:45
JayFseriously, it's important to me that we take care of burned out humans even if the answer is "go slower and use the thinking rocks less" 20:46
JayFand I am ... skeptical that will be a message that will ever be transmitted top-down from the foundation, given the folks who own/run LF20:46
gouthamrhmm, i was in a board discussion recently where we reiterated that community means humans but, saying that aloud here is probably in some part my/our duty being in the TC20:48
cardoeWhy did I decide to dig into roles and permissions mapping?23:16
gouthamr?23:31

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!