Tuesday, 2025-05-13

opendevreviewGoutham Pacha Ravi proposed openstack/election master: Create candidates/2026.1 placeholder directories  https://review.opendev.org/c/openstack/election/+/94955403:05
*** tosky_ is now known as tosky12:27
opendevreviewIvan Anfimov proposed openstack/openstack-manuals master: Documentation main page - add link to Glossary  https://review.opendev.org/c/openstack/openstack-manuals/+/94705815:15
*** ralonsoh is now known as ralonsoh_out16:03
gouthamrtc-members: a gentle reminder that we'll have our weekly IRC meeting here in ~52 minutes16:08
gouthamr#startmeeting tc17:00
opendevmeetMeeting started Tue May 13 17:00:04 2025 UTC and is due to finish in 60 minutes.  The chair is gouthamr. Information about MeetBot at http://wiki.debian.org/MeetBot.17:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.17:00
opendevmeetThe meeting name has been set to 'tc'17:00
gouthamrWelcome to the weekly meeting of the OpenStack Technical Committee. A reminder that this meeting is held under the OpenInfra Code of Conduct available at https://openinfra.dev/legal/code-of-conduct.17:00
gouthamrToday's meeting agenda can be found at https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee17:00
gouthamr#topic Roll Call17:00
frickler\o17:01
gtemao/17:02
fricklerhmm, pretty empty around here17:03
gouthamrcourtesy-ping: noonedeadpunk, gmaan, spotz, cardoe, mnasiadka, bauzas17:03
gmaano/17:03
gmaanboard meeting still going on but should be closed soon17:03
mnasiadkao/17:03
gouthamrgmaan: ack.. 17:04
bauzaso/ but distant 17:04
gouthamrspotz[m] should replace spotz on the ping list, don't know if she has notifications set for just "spotz"17:04
noonedeadpunko/17:04
gouthamralright, most of us are here, lets begin.. 17:05
gouthamr#topic Last Week's AIs17:05
spotz[m]Board meeting is still going but here17:05
spotz[m]Ok we're done there17:06
gouthamri have a few on me that i am going to punt to this week too, lest someone can lend me an extra pair of hands/time :D 17:07
gouthamr- mailing list discussion about Skyline SBOM17:07
gouthamr- adding VMT process links to the "New Projects Requirements" doc17:07
gouthamr- reach out to all project teams to refresh security liaisons and coresec groups17:07
gouthamrin similar vein i think fungi was taking a look at adding escalation documentation for unresponsive liaisons and PTLs within the VMT docs17:07
fungiyes17:07
gouthamri think you had a busy week, so we can work on these when/if you have time this week.. i'll keep tracking this17:08
gouthamrany update on "ansible-collections-openstack"? 17:09
gouthamrdid anything change wrt the review situation there? 17:09
gtemawhy should it change just magically?17:10
cardoeo/17:10
gouthamrnot magically, just checking if anyone here worked on this further17:12
noonedeadpunkI was not :(17:12
gouthamrack, ty.. 17:14
gouthamrwe took an AI to update our OpenInfra profiles with current affiliations.. this helps a couple of things we're trying to do/formalize before the next elections:17:14
gouthamr#link https://review.opendev.org/c/openstack/governance/+/949432 ( Require declaration of affiliation from TC Candidates)17:15
gouthamr#link https://review.opendev.org/c/openstack/election/+/876738 (Add affiliation lookup functionality)17:15
gouthamrty for the reviews on the proposal so far gmaan and JayF.. i'll post an update soon17:16
fungiah, yeah i was just about to mention that change17:16
gmaanso main goal is here to make affiliation visible to community when voting right? and off course to have it up to dated in openinfra profile.17:16
fungiif anyone wants to take it over and flesh it out, i'm happy to un-wip that17:16
fungiit's sort of bare-bones at the moment17:16
gouthamrgmaan: yes17:17
gmaanI am thinking if somehw we can show it (from openinfra foundation) to TC member table also?17:17
gmaanI know we discussed it in past but hardly recall why we did not do17:17
gouthamroh, i didn't know of that discussion17:17
gmaan#link https://governance.openstack.org/tc/#current-members17:18
gmaanthis one ^^17:18
gmaanat least it will help if any affiliation change in between and it will be easy to know by everyone17:18
gouthamrthat table is statically compiled.. 17:19
fungipublishing it on the governance web page was considered iffy because it makes it look like tc member seats are assigned to specific companies17:19
fungior that tc members are expected to act as representatives of their employers17:19
JayFI wonder if you could combat that with only showing the aggregate data.17:20
fungiwe tried to avoid giving that impression, since there definitely are other open source projects which operate that way17:20
gouthamr#link https://github.com/kubernetes/steering 17:20
gouthamr^ as an example17:20
spotz[m]And how you word it could make folks think we can't police the diversity rule ourselves17:21
JayFe.g. "TC membership is in compliance with corporate diversity requirements, with X members from Y company, A from Z,..." 17:21
JayFinstead of trying to map people:companies, just express that we have diverse corporate perspectives; which is the point?17:21
gmaanyeah, showing affiliation is not very new things and we know we should be community first not just as TC but  a community member17:21
spotz[m]But the problem is if folks affiliations aren't up to date then it's harder to verify17:22
fungiif you show the totals for different organizations, still would want to include a disclaimer that these are not x number of seats assigned to employer 1, y number of seats assigned to employer 2, et cetera17:22
gmaanwe are asking everyone to add affiliation in TC candidacy and profile. Showing it in governance page is more of making data easily visible to everyone. 17:23
gmaanthere is nothing we need to hide or make less visible about it17:23
gouthamrmaybe we just propose a change and debate on it? i do think adding disclaimers will be useful, and we should trust our community not to misconstrue this17:24
fungibut publishing it without providing clear context can be more confusing than not putting it front and center on the main page of the technical governance site too17:25
JayFgouthamr: I am not concerned about *our community* misconstruing this, I think the concern is how it appears to external folks17:25
gmaanyeah, if we are doing for election candidacy then we should do it on TC member table also.17:25
gouthamrack, i think the foundation could be concerned about external perceptions.. and i'm sympathetic to their view given all the campaigning they do to grow the community17:26
gmaanwe can highlight the main motive of org diversity, column name can be "Organization diversity"  or something17:26
JayFYeah I think it's all implementation details. Please add me as a reviewer on the proposed change and I'll happily help.17:26
fungii already hear somewhat regularly from people in other communities that "openstack is not openly governed" because they saw that companies get to purchase seats on the foundatition board, not realizing that the board doesn't decide the technical direction of the project17:26
gmaanwe do have affiliation shown for board members also in main page so it is not just for TC17:27
gmaanJayF: ++17:27
gouthamrtrue, fungi17:27
gouthamri actually want to fight this misconception17:27
gouthamron a tangent, people assume everythng the board adopts is automatically applicable to OpenStack17:27
gmaanI think if we see that then it will be valid when we put it in TC election candidacy 17:28
gouthamrit isn't, the TC still governs the project independent of the board's decisions, and we need to accept/adopt their recommendations/stance17:28
spotz[m]Historically it was, now it's not:)17:28
gmaansomeone can say having affiliation in candidacy also influence the votes?17:28
gouthamrit should :D 17:29
gmaanwell, we should not think only min requirement of org diversity  .17:29
gouthamri mean, we are hoping if company X has 6 candidates for an election, the electorate votes keeping in mind that only 4 of them will ever be on the TC17:29
JayFIt's completely reasonable to allow folks to vote against candidates based on their employers actions in/attitudes towards the OSS community. 17:30
gmaanI feel adding it in election is more externally concern then having elected TC member data showing17:31
gmaanJayF: against is fine, but who stop them to be in favor when they see the affiliation as one of the data for voting17:31
JayFgmaan: Ah, I hadn't thought about it from that perspective; but I suspect that if that sorta stuff is happening it's going to happen internally if not externally. Bad actors will find a way to act bad.17:32
JayFgmaan: and I say this as a person who had their vote specifically solicited by a former employer while I worked there via internal mailing lists during early, early OpenStack days :)17:33
gmaanI mean if "openstack is not openly governed" is perception then adding affiliation in election does not solve it instead it bump it more17:33
fungiyeah, we had at least one case (long ago now) where a candidate was censured because their employer was contacting customers suggesting voting for that employee17:34
spotz[m]I personally just remind folks to vote because we have a bad turnout issue. I don't care who they vote for:)17:34
gouthamr^ deeper problem :D 17:34
gouthamri mean = our voter turnout17:35
noonedeadpunkindeed...17:35
spotz[m]Yeah I know what you meant:)17:35
gmaanI was ok initially but seeing the points here especially fungi mentioned make me change my vote not to make affiliation more visible especially during election. 17:35
gmaanlet's keep it same as it is now, do diversity checks and advertisement on ML, IRC etc17:35
gouthamrokay, lets work out a proposal and discuss this some more17:36
gouthamrwe have other topics to get to today17:36
gouthamrty for engaging here, but please hold your pitchforks for the gerrit changes :D 17:36
gouthamrnext AI: we had a couple of "AI" meetings in the past week!17:36
gouthamrat this point, who's to say we're not sending ai agents to all these meetings.. but i did see spotz[m] and bauzas at these meetings, and i don't know if there's anything to share here beyond the summary i posted with the TC weekly email17:37
bauzascool indeed 17:38
gouthamr#link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/BX7XVZ4I6AAKJETU6Y2TQBNFXO7BUKZ2/ 17:38
gouthamrtl;dr: first paragraph17:38
gouthamron the last AI that i was tracking: "update all project contributor guides to list how to reach core teams" - i sought some action on this through that same email17:39
gouthamri was curious, do we do PTL-only emails like the release team does?17:39
gouthamri am inclined to stick to the list and prefix something with [all] and [ptl]17:40
fungican't hurt17:40
gmaan++, we are not asking any confidentiality to PTL, they can always share it with team or community. at least it will help to have better response/communication. 17:41
mnasiadka+117:41
gmaanI am ok if we send to PTL personal email also17:41
gouthamri did that before the PTG, and i don't know if it was useful or annoying :D no one complained, but people did come to our community leadership forum17:42
gouthamrthat's all the AIs i was tracking17:43
gouthamrwas anyone else working on anything else to note here?17:43
gmaanwe can send ML and the ping PTL who did not respond/take action. like election official did last time. I find that very useful 17:43
gouthamrtrue, good idea17:44
gouthamr#topic 2026.1 Elections kick-off17:44
gouthamryou may have noticed gerrit updates, ty for reviews thus far on election setup that ianychoi and slaweq have been posting17:44
gouthamr#link #link https://governance.openstack.org/election/17:44
gouthamr#undo17:45
opendevmeetRemoving item from minutes: #link https://governance.openstack.org/election/17:45
gouthamr#link https://governance.openstack.org/election/17:45
gouthamr^ the dates for the next election cycle are published17:45
gouthamr#link https://review.opendev.org/c/openstack/election/+/94955417:45
gouthamrwe are creating directories for folks to submit their nominations at any time here17:45
gouthamrsince i'm not contesting these elections in any capacity, i don't mind the collab with ianychoi and slaweq and helping with the TC stuff.. but, if you'd like to do it with me, please feel free and join the #openstack-election channel17:46
gouthamr^ all the planning/updates are posted here, along with specific gerrit changes.. 17:47
gouthamrany questions/concerns/feedback to share wrt elections? 17:47
gouthamralright, unfortunately, we're pressed for time again, so i'd like to table the topic follow up on the contributor experience to next week.. any objections?17:50
spotz[m]no but I won't be here:)17:50
gouthamra reminder to please push your colleagues to take the contributor and maintainer surveys.. 17:50
gouthamrack spotz[m] 17:51
gouthamr#topic A check on gate health17:51
gouthamr^ any gate health concerns to report? 17:51
gmaannothing much this week. py3.9 drop fixes are settle down now17:51
fricklerwell rpittau made a revert for the py39 drop from reqs and is adding extra work in to keep things updated17:52
fungia setuptools update broke pbr in some situations, work is in progress to address that17:52
gouthamr#link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/DOJZL5WZJMN4NEAD5GLIQZ455D2NUREY/ (wsgi/deployment changes)17:53
fricklerexcept that pbr CI is badly broken17:53
gmaanmost of project py3.9 are dropped or adjusted (c9s ad rocky joibs), not sure what was blocker to re-add py3.9 constraitns17:53
gouthamr#link https://review.opendev.org/c/openstack/requirements/+/948876 17:53
fricklerseem ironic has some more twisted deps17:54
gouthamr"For example, without RHEL 10/CS 10 support in Bifrost, we're risking having no upgrade path for the current RHEL 9/CS 9 users."17:54
gmaannot sure if we will be ready with cs10 by oct when 3.9 is EOL but somewhere we need to hard stop to support it17:55
fricklerthere also was some CI failures due to an osc-lib updates, sample fix is https://review.opendev.org/c/openstack/python-octaviaclient/+/94898117:55
gmaancs9 can be run on 3.11, that is how yatin moved devstack jobs to17:55
gouthamrright now, with the revert, any changes to the requirements repo not compatible with py3.9 will not be allowed, correct?  17:56
fricklerwell you can make version specific changes, like for >=py31017:56
fricklerthere's a lot of those in there already anyway17:57
frickler(for upper-constraints.txt)17:57
gmaanI am not sure requirement test the every bit of 3.9 but a basic checks yes. Many project dropped py3.9 testing like functional tests17:58
gmaanand unit tests from generic python testing template are also gone since start of cycle17:59
fricklerthat check only verifies that all of u-c is co-installable for every python version17:59
gmaanyeah, it is hard to say we support/test python 3.9 it is just we do not break if any testing exist18:00
gouthamr~time check~18:00
gouthamrthanks for highlighting this issue18:00
gouthamranything else to note in the minutes today?18:00
fungi#link https://summit2025.openinfra.org/cfp/ paris-saclay summit cfp closes in exactly one month!18:01
gouthamr++18:01
mnasiadkaWell, we have py3.9 back in reqs - but PTI for 2025.2 still does not include py3.9, right?18:01
gouthamryes18:02
fungialso based on discussion at today's foundation board meeting, sounds very likely we'll be able to drop the icla and ccla in favor of dco (signed-off-by) once the foundation joins the lf18:02
gmaannot just PTI, we removed the testing also18:02
gmaanwe are heading towards situation of py3.618:02
gouthamrfungi: ack, ty because i missed the meeting and will need to rely on minutes that won't be posted until they are approved at the next meeting :D 18:03
fungithere will be an announcement to the foundation ml about it, and i'll bring it up on openstack-discuss once that's posted18:03
gouthamr++18:03
gmaanI think jbryce  will send in on openstack-dicuss also but if ont then yes it will be on foundation ML at least18:03
gouthamrgmaan: i suspect this is temporary, but i share your suspicion that we may not be able to resolve CS10 issues by Oct18:03
gmaangouthamr: mainly to run CS10 test at upstream 18:04
gouthamryes..18:04
gouthamrlets wrap this meeting up, thank you all for joining, and for the spirited discussion.. we have lots to follow up on18:05
mnasiadkaMaybe it's just me, but gmaan's replies are invisible to me via Matrix bridge - I'll need to re-read the meeting again in the logs ;-)18:05
gouthamr#endmeeting18:05
opendevmeetMeeting ended Tue May 13 18:05:16 2025 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)18:05
opendevmeetMinutes:        https://meetings.opendev.org/meetings/tc/2025/tc.2025-05-13-17.00.html18:05
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/tc/2025/tc.2025-05-13-17.00.txt18:05
opendevmeetLog:            https://meetings.opendev.org/meetings/tc/2025/tc.2025-05-13-17.00.log.html18:05
gmaanmnasiadka: oh18:05
gouthamrmnasiadka: oh, yikes.. gtema/spotz[m] - do you see the same issue?18:05
fungipossible something's gone sideways with the bridge, maybe like a partial netsplit18:06
gtemagouthamr - seems to be same on my side18:06
fungiin oftc proper i see his replies fine18:07
fungiso the meeting log should include them at least18:07
mnasiadkait does :)18:07
fungioh, tc-members earlier today i approved a message through moderation to openstack-discuss from someone at another project asking whether we would relenquish control of https://pypi.org/p/quantum to them (for those who weren't around in the long-ago, that was the original name for neutron)19:28
fungihttps://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/2OHLYITMHMMQU5O2XKMCDNZCX4E3UOJD/#2OHLYITMHMMQU5O2XKMCDNZCX4E3UOJD19:28
gouthamri think we can relinquish the name, no one should be using quantum 2012.2 ; and this is akin to https://pypi.org/p/reddwarf20:31
gouthamrhttps://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/message/UIPSKKPCRH6E6ZVTSJW2UQPMNJ7ETELG/20:32
JayFWe can't "let them have it" in general though; the reason we don't use it anymore is that it is not a unique trademark. We can give the user the name in pypi; but it isn't any less ambiguous for them to use it than it was for us :)20:34
gouthamrfor sure20:34
gouthamrbut that's not our issue to impose?20:34
gouthamrwe can tell them there's an issue, as sean mooney and you have done, but its their problem to sort it out :) 20:35
JayFNo, it's not. But it also doesn't mean we should let someone run full speed into a brick wall without us warning them it's there :D 20:35
gouthamryep20:35
* JayF was hoping someone would have something beefier than what we said20:35
spotz[m]Yeah I just knew we renamed due to trademark issue, so while we can give it to him my concern is it's use would come back on us20:36
gouthamrhttps://quantum-journal.org/quantum-is-now-a-registered-trademark/20:37
gouthamrhttps://www.quantum.com/en/terms/20:37
gouthamrhttps://uspto.report/TM/9836235320:37
fungiif memory serves, the foundation's trademark lawyers were concerned we wouldn't be able to register it because there was a global storage hardware/software company of the same name, and so openstack's storage components could be seen as competitors (even though we wanted the name for the network component, that didn't really matter)20:37
gouthamrah https://www.quantumstorage.com/20:38
gouthamr:D20:38
gouthamrokay, computer storage20:38
gouthamrwhich is these guys: https://www.quantum.com/20:38
fungicorrect20:38
fungithe latter20:39
gouthamr> "while we can give it to him my concern is it's use would come back on us"20:39
gouthamrhow?20:39
fungithey made consumer hard drives originally i think, i had a few of theirs back in the '90s20:40
JayFthey made the bigfoots, the thinner but 5.25" style hard drives20:40
spotz[m]Well he found us through pypi:)20:40
JayFif you've never seen one, google it, it's pretty crazy to see 20:41
TheJulia... Yeah20:41
TheJuliaerr20:41
gouthamryes, and the people we transferred https://pypi.org/p/reddwarf to deleted the project?20:41
JayFhttps://en.wikipedia.org/wiki/Quantum_Bigfoot has a photo20:41
TheJuliafungi: yeah, I looked up the number of marks today for quantum and went "wow"20:41
gouthamrso maybe we can delete the contents and metadata and transfer stuff?20:41
gouthamrah, sasquatch of storage20:42
fungistorsquatch20:42
spotz[m]I wonder if the trademark is still an issue? Not that we'd go back to it20:42
fungiyeah, i can give it the same treatment as reddwarf if tc-members are in agreement20:42
spotz[m]I'm fine with it20:43
gouthamr+1 from me, we've good citizens amongst us, JayF and sean-k-mooney 20:43
TheJuliaGo take a search at: https://tmsearch.uspto.gov/search/search-information20:43
gouthamrthey warned the guy20:43
TheJuliaI've got a half drafted email noting the bit about protecting and defedning marks20:44
fungiin my personal opinion, the fweer names we're squatting on pypi for no good reason, the better20:44
gouthamr++20:44
JayFYeah I am like. +0 on this. I feel like doing what the person asks is going to make their life harder, but also I agree with fungi 20:44
JayFit's their business how hard they want their life to be I guess :D 20:44
fungiwell, 1. they may not want a trademark for the brand, but 2. their field of endeavor might not be seen to compete20:45
TheJuliayeah, all the community can do is try to be a good citizen and point out "maybe not one to use"20:46
gouthamrmaybe trademark lawyers and patent trolls collect fat paychecks for saying : "digital storage ~= package management" 20:46
TheJuliaand then release it20:46
gouthamr++20:46
TheJuliagouthamr: Even OSS projects have trademark lawyers... ;)(20:46
TheJuliaerr ;)20:46
spotz[m]I just don't want ours to have to get involved in this:)20:47
gouthamrtrue, and ours is quite conservative thankfully... that said, do we have any guidance somewhere 20:48
gouthamrregarding naming new projects?20:48
gouthamrhttps://docs.opendev.org/opendev/infra-manual/latest/creators.html#choosing-a-good-name-for-your-project20:48
gouthamrdoesn't call out trademarks, or consulting the foundation lawyers via legal-discuss for instance20:49
gouthamrfungi: we still own https://pypi.org/project/python-reddwarfclient/ 20:49
gouthamra few weeks ago, this came up regarding "aetos" : https://review.opendev.org/c/openstack/governance/+/94674420:51
gouthamrthis = the issue if we need legal to weigh in on new project names20:52
TheJuliaI'll throw in my $0.02: When I was at HP, we actually had to run project names by legal before we even proposed them upstream. That won't be same for everyone, and often, we've seen project have to change their names because you don't know sometimes20:53
fungiyeah, they didn't ask for us to give them that one. we *can* delete unwanted pypi projects if there's interest. or mark retired ones as "archived" now recently20:54
gouthamrTheJulia: yes; if we suggested that folks reach out to legal-discuss in the project creators guide, would the OpenInfra Foundation be billed for these requests? 20:56
fungii can find out, but also the answer to that is likely to be different a month from now20:57
gouthamrah, due to the LF transition?20:57
TheJuliaYeah20:57
gouthamrwe can hold off and ask when that's complete, we don't get new project requests often20:57
fungiyeah, would probably go through central lf legal, i hear they have a trademark lawyer on staff full time20:58
TheJulia*plus* some of us on legal disucss have to respond "I'm not a lawyer, but..."20:58
gouthamrin the past few weeks, we created "grain-ui" and "aetos" .. 20:58
fungigrian20:58
fungi(not grain)20:58
gouthamroh yes that20:58
fungithough i think people's spellcheckers keep auto-correcting it20:58
gouthamram my own spellchecker - that's how i defend my typos :D 20:59
gouthamr"Grian is the pseudonym of a British YouTuber, who is primarily known for creating Minecraft content."21:01
gouthamr#TIL21:01
TheJuliaoh my21:01
TheJuliaits not in the USPTO database!21:02
gouthamrhaha, there's EPO, JPO, CHIPA, IP Australia, UK IPO, WIPO, ... and twenty others - unless we explicitly grin at non US intellectual property police21:04
TheJuliasuper true21:04
gouthamrwould the TC like a vote on the matter? or is lazy consensus preferred21:06
gouthamrtc-members ^21:06
TheJulia14 results on Grian in WIPO's trademark db21:06
gouthamr(most of who may have slept)21:07
gouthamryikes21:07
TheJulia9,870 for quantum21:07
gouthamr*flip table* meme21:07
TheJulialol21:07
TheJuliaTFaaS21:08

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!