Tuesday, 2018-08-28

*** rosmaita has quit IRC01:30
*** annabelleB has joined #openstack-tc02:23
*** annabelleB has quit IRC02:26
*** ricolin has joined #openstack-tc03:08
*** Bhujay has joined #openstack-tc04:18
*** Bhujay has quit IRC04:19
*** tdasilva has quit IRC04:27
*** pabelanger has quit IRC04:55
*** annabelleB has joined #openstack-tc05:01
*** annabelleB has quit IRC05:01
*** dims has quit IRC07:08
*** dims has joined #openstack-tc07:10
*** cdent has joined #openstack-tc07:39
*** tosky has joined #openstack-tc07:58
*** jpich has joined #openstack-tc07:59
*** dangtrinhnt_x has joined #openstack-tc08:06
*** dangtrinhnt_x has quit IRC08:10
*** e0ne has joined #openstack-tc08:43
ttxtc-members: office hour starting09:00
ttxmugsie: we have freezer and karbor no our plate for health updates, we should probably split the effort09:02
* cdent waves09:02
ttxstarted a bit on Freezer, so if you can take karbor that would be great09:02
mugsiettx: sure, will do that today09:27
ttxmugsie: thx09:29
ttxdid we arrive at some decision regarding campaigning at the PTG ? I had a hard time telling from the logs09:30
mugsieI don't think there was much of a decision, other than no offical debate / meet the candidates session. someone suggested a meet the TC session where candidates who are in Denver could mingle with interested people09:40
ttxhmm ok09:42
ttxI'll wait a bit more before confirming/abandoning that Wed 5pm slot09:42
*** dtantsur|afk is now known as dtantsur09:44
cdentyeah, what I was able to gather was that an informal "governance gathering" on wednesday was an option. but if we made it official campaigning that would present problems.09:44
* cdent locates sustenance09:45
*** cdent has quit IRC09:45
*** cdent has joined #openstack-tc10:20
eumel8ttx: Are yml files requirements?10:53
eumel8 https://github.com/ttx/openstack-map/blob/master/openstack_components.yaml10:53
openstackgerritAndreas Jaeger proposed openstack/governance master: Add -E to sphinx-build invocation  https://review.openstack.org/59704711:14
*** ricolin has quit IRC11:24
*** rosmaita has joined #openstack-tc11:59
ttxeumel8: not sure i understand your question... Do you mean, was it necessary to use YAML files ? No, it just meeds to use structure machine-readable data files, and we usually like YAML... Now the website code reads those YAML files so I'd rather not change them12:08
eumel8ttx: Questions from my side are always I18n related ;) Would be interesting to have a multi-language site on the stage. That's why I looked into your proposal to set up a new repo for www.o.o stuff12:13
*** mriedem has joined #openstack-tc12:35
ttxeumel8: it's for www.o.o which is not translated right now, but I suspect that could be a future evolution.12:38
*** cdent has quit IRC12:46
eumel8ttx: Zanata doesn't support yml. it should be xml, txt or po files12:47
ttxeumel8: noted. we'll transition to another format if we end up translating the site.12:49
*** annabelleB has joined #openstack-tc13:01
fungithough by then, zanata might also grow yaml support13:04
*** dangtrinhnt_x has joined #openstack-tc13:11
*** tdasilva has joined #openstack-tc13:13
ttxfungi: We have Barbican on our plate for health updates. Was that something you planned to cover soon ?13:14
fungittx: oh, yes i can get that one added. meant to earlier--thanks!13:15
ttxfungi: cool, thanks13:15
*** cdent has joined #openstack-tc13:29
TheJuliare: ironic health update, would it be help full if I put my assessment as the ptl in?13:33
*** annabelleB has quit IRC13:34
*** purplerbot has joined #openstack-tc13:34
smcginnisBut "telephone" is such a fun game. :)13:41
dhellmannwell, the idea is that we have someone "impartial" looking at things, so we do want the PTL input of course as long as we also have the objective review13:45
*** dangtrinhnt_x has quit IRC13:45
*** dangtrinhnt_x has joined #openstack-tc13:47
*** dangtrinhnt_x has quit IRC13:52
mnaseri guess we'll have to update the list14:03
mnaseras i'm a ptl of a project i was watching14:03
*** amotoki has quit IRC14:03
smcginnisIf the point of this exercise was to make sure the TC was aware of things going on with the projects, having the TC/PTL seems good to me. Are we really concerned about having an objective review of each others input?14:04
evrardjparen't there two person on each project? Shouldn't that be enough?14:06
dhellmannyeah, I figured for this first round the 2nd person would be good enough14:06
evrardjpand +1 on smcginnis comment14:06
dhellmannI'm not concerned that someone would intentionally misrepresent a project's status, just that they may not see a problem because they're too close to it14:06
evrardjpI'd presume good will :)14:06
evrardjpI see14:06
*** ricolin has joined #openstack-tc14:17
openstackgerritAndreas Jaeger proposed openstack/governance master: Add -W to sphinx-build invocation  https://review.openstack.org/59704714:24
mnasermriedem: from what I see, your email is usually "mriedem.os@gmail.com" but under the upgrade checker goal, i see it listed as "mriedemos@gmail.com" .. shall i fix that up for you?14:33
*** Bhujay has joined #openstack-tc14:33
mnaseror was that planned to avoid getting any emails about it :P14:33
*** Bhujay has quit IRC14:34
*** Bhujay has joined #openstack-tc14:35
cdentmnaser: they are the same14:36
cdentgmail doesn't care about the .14:36
mnaserthat explains a lot of emails i get to mohammed.naser@gmail.com ..14:37
*** pabelanger has joined #openstack-tc14:38
cdentmnaser: yeah, it's really annoying if you have an apparently common name like chrisdent and people don't know their email address when signing up for things14:43
smcginnisEven less common names - I get a surprising number of those.14:44
*** dangtrinhnt_x has joined #openstack-tc14:48
ttxdhellmann: so what's the final call on that campaigning meeting ? Should I keep the slot on Wednesday evening, or just encourage people to network ? I can do a call-out to TC candidates on the Monday post-lunch presentation if we prefer to keep it low-touch14:53
dhellmannwhat would the wednesday evening thing be? a mixer?14:54
dhellmannI'm torn myself. I understand potential candidates not wanting to participate in something very social, but at the same time it's important for the TC to have some visibility.14:55
dhellmannI don't think making it a campaign event is necessarily good, for all of the privilege reasons others have stated.14:56
*** e0ne has quit IRC14:56
*** Bhujay has quit IRC14:58
*** jroll has quit IRC15:05
*** jroll has joined #openstack-tc15:06
ttxdhellmann: that would be in the lunch room, so more amenable to a panel format than a mixer.15:06
ttxMaybe it's simpler to tell everyone that the TC members and candidates will be around at the official happy hour15:07
mnaserpanel format means (imho) some will avoid it from the pressure (maybe!) and others can miss out on that opportunity if they cant make it to the ptg for $reasons15:07
ttxthat happens at the bar15:07
mnasermaybe tc candidates who want to make themselves available can just send an email if they want to saying they'll be at the bar :p15:08
mnaseror wherever they feel comfortable being15:08
* mnaser just isn't a fan of making it an official thing15:08
mnasermaybe we should find a way to vote on this in a quick and simple way15:08
ttxSo two options: wednesday 5pm in the PTG lunch room, with a stage and a mike, or just crash the official happy hour on Tuesday at 5pm, saying TC people will be around15:08
ttxI'm tempted to just two the latter15:08
* cdent has given ttx his disease15:09
jrollI also like the latter15:10
mnaseri feel like saying the tc will be at $x is okay, i would hope that naturally candidates would want to be present there without us having to say that15:12
dhellmannttx: yeah, let's just encourage tc folks to be around for the mixer15:14
* cdent gets ready for the sock hop15:17
mnaseri liked how it was called in vancouver15:24
* mnaser looks if the schedule is still there15:24
mnaserit was something along the lines of openstack leadership something?15:25
*** david-lyle has quit IRC15:33
*** dklyle has joined #openstack-tc15:34
*** dtantsur is now known as dtantsur|afk15:44
fungihttps://review.openstack.org/596619 reveals that we should likely get https://wiki.openstack.org/wiki/LegalIssuesFAQ#OpenStack_Foundation_Copyright_Headers (and maybe some other things from that article as well) moved into a more formal location15:56
fungii can put it on my to do list15:56
fungiunless someone else wants it15:56
fungii can start it with an ml thread15:57
*** e0ne has joined #openstack-tc16:06
*** Bhujay has joined #openstack-tc16:17
*** Bhujay has quit IRC16:22
*** dangtrinhnt_x has quit IRC16:23
*** Bhujay has joined #openstack-tc16:30
dhellmannfungi : ugh, yeah. where do you think we should put that? project team guide? governance?16:33
fungimight go alongside our licensing document16:33
fungii haven't thought that far yet. still working up the e-mail to the list16:33
*** jpich has quit IRC16:39
*** Bhujay has quit IRC16:43
*** Bhujay has joined #openstack-tc16:44
fungiposted http://lists.openstack.org/pipermail/openstack-dev/2018-August/133929.html as well as a heads-up to legal-discuss at http://lists.openstack.org/pipermail/legal-discuss/2018-August/000507.html17:04
notmyname...and replied :-)17:11
smcginnisShould we propose a revert for that openstack-helm patch?17:14
fungii'd prefer for the author to propose that revert17:14
smcginnisProbably best17:14
funginotmyname: that's a great point, and one for which i don't believe we've ever gotten any clear legal advice17:15
fungii'm not actually sure whether assigning copyright to a non-legal entity (neither a natural person nor a legal company/organization) is appropriate, though i've certainly seen to done all over the place17:17
notmynameI'm ready to tell any lawyer flat-out that we have never and will never start reviewing patches based on copyright headers. so tell me the way to do it, and I'll propose a global copyright header update today17:17
fungier, seen IT done17:17
notmynameyou get fun questions like "dose *any* change add a copyright claim to the code? does removing all lines edited by a particular individual remove their copyright claim to the file? do any rights change as employers change? what about when acquisitions happen?17:18
notmynameI'm sure there are good answers here. but I don't care and don't want to be bothered with them. I just want the legal equivalent of "run `git blame` to figure out what legal departments you need to go talk to if you have an issue"17:19
evrardjpnotmyname: so I heard that when acquisitions happen, recopyrighting has to happen, but I am not a lawyer17:19
evrardjp"don't want to be bothered with them" -> who wants that except a lawyer? :p17:21
notmynameI am 100% sure that with no advice or with advice that equates to "list every holder individually", we will continue to be completely wrong with whatever is at the top of every file in whatever repo exists17:21
smcginnisAt one point I was also given the mandate for an internal project that any time a file is touched the date at the top needed to be updated.17:21
evrardjpsmcginnis: I heard that one too.17:21
fungismcginnis: heh, i've actually implemented a git commit hook to check that in the past17:22
fungiwhich basically refuses to commit if you're changing a file with a copyright date earlier than the current year17:22
smcginnisAnd there is at least one storage company that for a time was making each person, regardless of change, to get legal approval and add their own personal copyright line appending to the list of people that have in any way modified the file.17:22
mnasermy view is17:22
fungitbh, i consider that absurd17:22
mnaserthe simple rule is17:22
notmynamesmcginnis: sounds like it's ripe for svn-style keyword expansion :-)17:22
mnaserdon't upload anything with OpenStack Foundation, that's all17:22
smcginnisfungi: Agreed!17:22
cdentaw. I miss svn keyword expansion17:23
smcginnisnotmyname: But then the lawyers can't review the final text first. ;)17:23
fungiif rcs and cvs didn't have it, i don't need it ;)17:23
mnaserIMHO the email that fungi sent should be considered as one coming from the foundation saying "dont do that"17:23
fungiexcept for the disclaimer i put at the top of that message saying i'm not legally representing the osf?17:24
mnaseri see that the same way as if some group of users changed all copyright stuff to $random_company17:24
*** Bhujay has quit IRC17:24
mnaserfungi: well, i'd think that maybe we need someone officially from the foundation saying "don't do this, thanks!"17:24
notmynamemnaser: sure. fine. I'll still not care if someone submits a new patch with a new file and copy/pastes an openstack foundation copyright from an existing file17:24
fungibut yeah, so far the only straightforward advice we've gotten is that no new content is to be assigned copyright for "openstack foundation" unless it's actually written by someone at the openstack foundation17:24
smcginnisOh, can we actually change them all to the literal string $random_company? :}17:25
* mnaser starts filing for $random_company, LLC17:25
mnaseroh the databases we would break..17:25
smcginnisI actually kind of like the idea of a company called Random Company.17:25
mnasersounds like something musky would come up with17:26
funginotmyname: to your point, if someone asserts a copyright in a file update, you don't have an easy way of knowing whether it's correct anyway so i concur with not reviewing it. my main concern is with projects doing mass updates thinking they need to assign copyright to the osf without actually asking anyone17:27
mnaseryup i think if someone innocently copy pastes stuff, no one really minds that much17:27
mnaserbut the action takes there was to globally moves copyright over to the OSF which is a bit more "serious"17:27
evrardjpsmcginnis: you'll get random employees too17:27
mnaserif i understand fungi correctly17:27
fungiwhen it's a misconception at a leadership level within a project team, that's what i want to address17:27
evrardjpfungi: it's definitely something that was misunderstood17:27
mnaserfungi: i'm wondering how you managed to catch it in the first place17:28
fungimnaser: clarkb was apparently flipping through irc channels and happened to see the gerritbot message about the change merging. i take no credit for spotting it17:29
evrardjpI am glad that we have fungi, if he could catch all the potential big fails, I'd like to wire him before gating :p17:29
notmynamefungi: as someone "at a leadership level within a project team", I can easily point out many places in swift's repo where copyright headers are wrong. and I always feel I have a misunderstanding of the right way to do it. hence my refusal to review it (or really pay attention to it) and my email asking for a simple "go look in `git blame` for it"17:30
fungievrardjp: i'm afraid my meat-based processor couldn't handle the throughput17:30
funginotmyname: there's also the tension with downstream consumers (particularly redistributors/distros) who want to know who's making copyright claims on the source code. in actuality it's a futile endeavor, but try telling lawyers that (i mean, i have tried telling lawyers that, but they seem to have some sort of mental block regarding such matters)17:32
fungicopyright law, as i understand it (at least as berne convention, buenos aires convention and similar signatories implement) is that all creative works are subject to copyright of some creator(s) whether or not they're enumerated. the best we can do as a project is to remind contributors to get advice from their employers (in cases where they're contributing on behalf of an organization) as to what17:37
fungicopyright they should claim when making copyrightable changes/additions17:37
fungias reviewers we can't do much more than assume they've done it, and that whatever they claim is correct (excepting cases like the current example where the rationale _was_ explained and was clearly counter to advice we've already gotten for the organization which the authors were attempting to say claimed ownership of those rights)17:39
notmynamefungi: that is all true and my undestanding too. but these conversation always tend towards those sort of comments (ie technical reasons for the Way Things Are), but they are unhelpful at actually making the code we maintain more legally correct17:39
fungisure. as is often said, "the law" is not scientific, and cannot be approached with the same analytical skills as one would apply to technical endeavors such as software development17:40
fungiattempting to review copyright assignment claims is not going to work out particularly well17:40
fungiso while i appreciate the desire to have some simple rule reviewers can follow for all cases, i don't think it exists. my goal here was not to come up with repeatable, deterministic process. it was to address a legal misunderstanding17:42
notmyname...which is why I want to completely short-circuit that by not having anything other than a generic "those who hold rights are the rightsholders" statement at at the top17:42
persiaI have previously received qualified advice that in some common law jurisdictions, the author is free to assign copyright to any entity they so desire at the time of publication, without necessarily receiving approval from the entity to which copyright is assgined.17:42
persiaI believe that most of the states in the united states of america are considered such jurisdictions, but have not received qualified opinion on a sufficient number of them to make any statements related to whether people can assign copyright to the openstack foundation even if OSF doesn't want it.17:43
fungipersia: that may be, indeed. the claimed rightsholder doesn't necessarily need to exercise those rights17:43
persiafungi: Indeed, they do not :)17:43
persiaAnd, at least in the EU, I have received qualified opinion that they cannot (as such assignment is forbidden).17:44
fungii think the point, though, is the osf doesn't want rights assigned to it erroneously, and the assumption of some seemed to be that it does17:44
persiaI think the point is that the OSF doesn't want rights assigned to it, except where it itself makes such claims.  The cause of my comment is the use of the word "erroneously" (and that any time "copyright" appears in my buffers, I take a closer look).17:46
notmyname"I wrote this while above international waters between the US and EU on a plane registered in Turkey outside of business hours and on my personal laptop but related to my company's work. it completely removes lines contributed by another contributor"17:46
persianotmyname: Doesn't matter.  If a copyright claimant wishes to pursue action against a publisher, the copyright claimant typically needs to show why they have standing, and then how the publisher violated license.  For people that don't care about being claimants, anything can be written.17:50
*** ricolin has quit IRC17:50
notmynamepersia: so let's just remove all copyright headers from all files! (*please*)17:50
persiaGenerally publishers (e.g. OSF and many vendors) prefer contributors to either not make claim (so no action), and to make firm positive statement that they are not subject to another claimant (to provide indemnity).17:51
persianotmyname: Works for me.  Compliant with my read of treaties.  On the other hand, lack of announcement of counterparty is often interpreted as lack of standing to enforce the license.  As a result, publishers may not consider the license actionable as a contract, etc.17:52
persia(this both means that the patent protections may not be enforced and that publishers may seek alternate code they feel more comfortable publishing).17:53
notmynamecommit message dco and remove copyright headers. sounds simple enough that *surely* someone would have proposed that years ago in openstack ;-)17:53
persiaGenerally that is considered insufficient protection for publishers.  If someone wants to publish the code (e.g. provide it to users), they may not feel they have the right to do so.  Go ask someone who gives code to users if they feel safe that way.17:54
notmynamedevs/reviewers are never going to be able to police copyright statements in code. please give us guidance that makes lawyers happy and matches that reality.17:56
persiaNobody can.17:57
persiaI'm not qualified to have an opinion, which lets me state lots of things which may or may not be true.17:57
notmynamewell, maybe "makes lawyers happy" is too much to ask. how about "that doesn't block lawyers from doing their work"17:58
persiaQualified people are either a) restricted to developing opinions that favor specific entites that fund them or b) state actors.17:58
persiaGenerally developers don't trust folk in category (a) (for good reasons: they must act in narrow interest), and are suspicious of folk in category (b).17:58
persiaFor "doesn't block lawyers from doing their work", one solution I've seen used in the past is for a sponsoring fiscal entity (e.g. OSF) to arrange to receive legal opinion regarding a template for the community.17:59
persiaIt can't be enforced, but it might be sufficient.  Whether it is actually in the interest of the OSF to publish such a recommendation is more complicated, because there are many jurisdictions, and some guidance may contradict other guidance.  This would also be a *very* expensive undertaking, considering the number of jurisdictions for which research is required.  Is that the best use of OSF's limited funds?18:01
persiaI suspect the question of funding such an exhaustive recommendation was raised as part of drafting https://wiki.openstack.org/wiki/LegalIssuesFAQ#OpenStack_Foundation_Copyright_Headers18:02
notmynameI'm trying to say that I do not care and practically it doesn't matter. either we keep it as-is and know it's wrong (but not how it's wrong), we remove all copyright references, or we are given a generic statement like I said above18:05
notmyname(I mean, I care that someone is thinking about it in the same way I care at some level that my food supply doesn't poison me. I don't care about either in my day-to-day things I can control)18:05
persiaIf you don't get paid for this ask a lawyer, and if you do get paid, ask your funding source who to ask, but it may be considered safe to leave copyright headers alone when patching code.  New files are more complicated.18:06
persiaSadly, because of how law works, more precise statements aren't actually meaningful (and anyone qualified to make them probably isn't allowed to tell you except in strict confidence and awareness of your particular interests).18:07
fungimy take, for what it matters, is that we generally accept whatever contributors put in as copyright claims under the assumption that they've received legal guidance which has told them what to claim (if anything). not claiming copyright over some change/addition is a perfectly valid choice, but also some employers may have very specific requirements about having their employees claim some specific18:07
fungicopyright. we can't really know18:07
persiafungi: In that sentence, who do you mean by "we"?18:08
persiaRather, in "we generally accept"?18:08
fungisorry, "we" as reviewers of those changes, in the general sense18:08
persiaI suspect most reviewers don't even make that assumption, and, as reviewers are typically not publishers, simply don't care at all, but I like the thrust of your take :)18:09
fungithe main reason i see it as important to consistently remind people not to try to assign copyright of their work to the osf is that we want to actively avoid the perception that openstack projects demand copyright assignment from contributors18:09
persiaI agree that this is an important message.  Sadly, it is a negative message.  Worse, I cannot immediately think of a positive declarative statement that provides the right tone.18:10
fungithe more new files we add which are copyright "openstack foundation" the more likely it is for the uninformed public to make such assumption18:10
notmynameso the practical effect is that when contributors make small changes to copyright statements, we don't care and just add/update the boilerplate, despite not being able to test/check it and it having no impact on the ability of the code to run. or someone makes a big change that causes someone to see it, and we have our bi-annual copyright conversation18:10
fungipretty much18:11
persiaMaybe something like "The OpenStack community encourages all contributors to claim copyright as appropriate, maintaining their own standing in assisting with the enforcement of licenses for our shared code."?18:11
persianotmyname: Precisely.18:11
fungibut at least we can surface the guidance and reasoning we have received, as vague as it may be, somewhere more prominent than the bottom of a drawer in a locked filing cabinet in a disused lavatory with a sign on the door saying "beware of the leopard"18:12
fungier, i mean, the wiki18:12
notmynamepersia: read as "reviewers, please continue to ignore the cruft building up at the top of your code files"18:12
fungiyes, my call was directed not at change reviewers but at change authors18:13
fungiauthors: please don't assign copyright to the osf unless you know you're actually supposed to. reviewers: keep on keepin' on18:14
persianotmyname: Roughly.  Also "publishers, be aware that there's a long list of people who can sue if you violate the license", and "contributing orgs, feel free to pursue your complex partial-public-source revenue models, including being a copyright troll, if you prefer", and "contributors: know that no entity is going to take your code and do something special with it not described in the license".18:14
notmynamebut that resolves the conversation in the most unsatisfying way: keep the status quo, despite knowing that it leads to wrong results. guess we'll schedule this conversation again in 2020 ;-)18:15
notmynameah.. guess it will free up the time for the annual "what is openstack?" conversation ;-)18:15
persiafungi: Yep.  Well targeted.  Although in 2020, maybe try to phrase it without the use of either "not" or "n't", for reasons having to do with human cognitive processing :)18:15
cdentwe could ignore the lawyers, remove all the headers, and when they say they need to be there, point them at git and gerrit18:17
notmynamecdent: I like it!18:17
persianotmyname: I think that part of the issue is that nobody wants to know if it lead to any wrong results.  OSF might not want to hold much copyright (for the messaging reasons mentioned above), but I can personally say that I'm delighted not to have any idea who actually holds rights over which lines of code: that knowledge might make me less comfortable using OpenStack (because of weird differences in law related to ignorance).18:18
persiacdent: That has been suggested in other communities.  Generally this causes downstream publishers to stop publication.  Before Berne, that was a more viable model.  Now, default is that one can't publish unless one has a contract with a claimant to permit one to publish.18:19
notmynamepersia: so we know the current state is inaccurate but we'd prefer not to resolve it because resolving it would make us uncomfortable with who actually holds rights. therefore we continue to ignore it18:19
notmynameisn't there a pair of glasses in Hitchiker's Guide that is expressly for this purpose?18:20
cdentpersia: My comment was meant to indicate that current reality is not what I aspire to.18:20
persianotmyname: Precisely.  In the event that anything was actually wrong, OSF might have to stop publication.  That would probably frustrate many of us: one of the first impacts is that there would be no CI, and no code review.18:20
cdentOr if you must: I was kidding.18:20
persiacdent: Heh :)18:20
persiaYep.  Same idea.  As long as something is done unintentionally, and nobody makes a tort claim, it's usually considered acceptable.18:21
funginotmyname: i think your real beef is with copyright law anyway (or perhaps intellectual property law in a broader sense)18:21
fungialso, probably, lawyers in general ;)18:21
* persia is willing to argue that there is no such thing as "intellectual property law"18:21
persiaAnd, for that matter, the framework under which lawyers operate that actively prevents sensible answers to questions like "is foo legal?"18:22
* fungi has existential arguments which prove there's no such thing as lawyers ;)18:22
notmynamefungi: lol. I think I'm wearing my own peril-sensitive glasses so that I just continue to ignore it. I *think* I just want to actually acknowledge that instead of prentending that we don't know the current state is wrong18:22
*** tosky has quit IRC18:23
persianotmyname: If you feel the current state is wrong, I encourage you to solve the "ignorance is a defence" problem, rather than trying to solve ignorance about this specific issue.18:23
fungii'm well aware the current state of things is wrong, or more specifically that the desire of copyright lawyers cannot actually be fulfilled in any meaningful and _they_ get to pretend they don't know it's all wrong18:23
notmynamesomething more like what cdent said: "we aren't equiped to maintain the current list of rightsholders, so if you are equipped, then go start with git and gerrit and knock yourself out"18:23
persianotmyname: How does that provide the right to publish?18:24
fungithe information in git is not actually useful for determining who the rightsholders are. only for determining who their proxies might be18:25
persiaActually, it's only useful for determining who the proxies claimed to be, which may or may not be accurate (I can publish a change with any strong in "Author" I wish, as an example).18:25
fungilawyers want to be able to assume that any legitimate claimants of copyright have expressed that desire somewhere. as long as nobody actually wants to challenge them, they're perfectly happy to pretend it's sufficient and go care about other things instead18:26
notmynameI don't know the legal definition of "right to publish". I know that there is a social contract that we work together on a codebase and part of that contract is that we get to use the code18:26
persiaSadly, many of the folk actually working to solve these problems are distracted by the "there is a right answer, and I can derive that mechanically from SCM" problem, so few compliance suites provide a means to store rights claims effectively, and there is little consensus on an appropriate format in which to store such claims.  When more progress is made, it would be appropriate for the openstack community to adopt the standards that might come18:27
persiato exist.18:27
fungiwe do at least have the icla (the dco should similarly be sufficient) providing a legal agreement that the person submitting the work is not doing so under false pretenses18:27
persianotmyname: Generally speaking, copyright is the right to make a copy (publish).  Anyone without copyright generally can't publish anything.  There are limited cases (e.g. code licenses) in which those without copyright can make copies.  Most licenses are a contract between two parties (the copyright holder and the publisher).  With multiple copyright holders, this becomes more complex, and is usually interpreted as a set of independent parallel18:29
persiacontracts for specific sections of the published material.18:29
persiaThe social construct in this situation is embodied in the local copyright law in a relevant jurisdiction (which can be the home of a copyright holder, the home of a publisher, or a location in which publication happens).  Since there is wide variation in law (although the vast majority is compliant with several international treaties), there are a wide variety of answers to whether someone can publish something in some location.18:31
persiaThat our community might operate under a different social contract doesn't waive the liability of any publishers to meet their license obligations to the copyright holders as determined in any jurisdiction that may have standing (in some of which, action may be brought without involvement or agreement of either the copyright holder or the publisher).18:32
persiaHence the need for *something* to provide a counterparty so that the publishers can publish, and the massive value in having *no idea* whether that information is correct to enable lots of publishers to republish without concern.18:33
persia(which, admittedly, is totally broken, but see above).18:33
persia(oh, and in our case, the specific contract in place is Apache 2.0)18:34
TheJuliaso silly question while my brain is on it. Extra ATCs, do we have any documentation around that?18:54
smcginnisTheJulia: We put a little snippet in the release schedule - https://releases.openstack.org/rocky/schedule.html#r-extra-atcs18:57
smcginnisBut other than that, I don't know of anything.18:57
jrollTheJulia: idk about docs, but it's fairly easy, it looks like this: https://github.com/openstack/governance/blob/master/reference/projects.yaml#L67618:59
TheJuliaIt seems totally subjective to me :)18:59
TheJuliajroll: yeah, I have the file right in front of me18:59
jrollah ok18:59
jrollI think it's designed to be subjective19:00
smcginnisYep, it really is a subjective thing.19:02
smcginnis"This person has really helped out and I think they should be recognized for it"19:02
openstackgerritJulia Kreger proposed openstack/governance master: Add Jay Faulkner as an extra-atc for ironic  https://review.openstack.org/59721219:08
*** annabelleB has joined #openstack-tc19:08
*** e0ne has quit IRC19:08
TheJuliabetter to remember to praise sooner rather than later.19:09
fungihearty +1 on JayF! i miss seeing him at events19:17
*** e0ne has joined #openstack-tc19:23
*** annabelleB has quit IRC19:27
*** annabelleB has joined #openstack-tc20:14
*** rosmaita has quit IRC20:33
dhellmanntc-members: the python-dev community is working on a survey of open source governance structures to help make a decision about how to structure their team with Guido stepping down as BDFL20:40
dhellmannI'm contributing a section based on openstack, and I would appreciate your help in making sure it is accurate and relatively complete (there are a lot of subtle things I think we can leave out)20:41
dhellmannI've started a draft in https://etherpad.openstack.org/p/python-dev-governance-survey20:41
dhellmannand the existing PEP content is in https://www.python.org/dev/peps/pep-8002/ for reference20:41
* cdent looks to see if dhellmann has figured out a name for the local style20:41
dhellmanntrying to explain this in a 1-pager was an interesting experience this afternoon20:42
fungisounds like an awesome exercise20:43
cdentdhellmann: after a quick skim it makes sense and is superficially complete. I'll give it a better read tomorrow. Do you know who else is is planning to provide a section?20:47
*** annabelleB has quit IRC20:47
cdentdhellmann: found a fixed a typo: end up in situations20:48
fungithe focus first on apc instead of atc is an interesting choice, but i see where you're going with that20:52
*** e0ne has quit IRC20:58
fungiwe don't elect exactly half of tc seats each cycle, only _roughly_ half (unless someone volunteers to get split into two equal pieces)20:59
fungii question classifying wechat as a "tool" since as an alternative the comparison is to irc (a protocol/platform)21:00
fungiirc clients are tools, wechat is more of a platform/network/forum/system/something21:01
smcginnisEh, wechat makes sense in this context as a tool for Chinese contributors to communicate.21:02
smcginnisI'm just glad it wasn't slack instead of wechat. :)21:02
smcginnisdhellmann: Looks like all the major points are covered in that.21:03
TheJuliaReading through, line 96-109, it feels like it might be good to highlight that individual teams may vary based upon needs21:03
fungii'm probably just being pedantic, but the assertion there is that chinese contributors have trouble accessing irc so they use a different "tool" (that they have trouble accessing tools rather than systems is a questionable implication)21:04
fungithey have plenty of access to irc clients. their irc clients have trouble accessing irc networks. in contrast, other clients (namely wechat) have no trouble accessing wechat networks21:05
TheJuliaIt might be better to state that we _encourage_ use of irc, but recognize it does not work in every case.21:06
fungimay be better to focus on what we require, which is design and development conversations held in public with a record of what was discussed21:07
fungiirc fulfills this requirement and is used by a majority of the contributor community. contributors in some parts of the world end up needing to use other tools to meet this requirement if they have trouble accessing irc21:07
dhellmanncdent : we had a meeting the other day and came up with a list: https://etherpad.openstack.org/p/python-governance-meeting-projects-to-survey21:07
dhellmannfungi : your point on "platform" is valid; I'll change that21:08
dhellmannfungi : I thought we called them APCs now, based on the charter document21:08
fungidhellmann: apcs and atcs are different but related classifications21:09
fungiapcs are the project-specific active contributors21:09
fungiatcs are the aggregate of all apcs21:09
dhellmannTheJulia : does the addition on lines 87-88 address your comment?21:09
dhellmannfungi : ok. That may be a level of detail that isn't necessary to explain for this audience, but I'll give the phrasing some thought.21:10
fungiso ironic apcs elect the ironic ptl. ironic apcs are atcs, as are nova atcs. atcs as a whole elect tc members21:10
cdentdhellmann: this statement is very true (of node): NodeJS have both spent a lot of time thinking about governance and community-based decision making. Do you know anyone from that community was found?21:10
TheJuliadhellmann: perfect21:10
dhellmannI think I said something like "the APCs of all teams"21:10
fungias a point of clarification though, the foundation bylaws refer to atcs but do not concern themselves with election of ptls. apcs are wholly invented within the technical committee's jurisdiction to decide how ptls will be elected21:11
dhellmanncdent : I don't know. If you know someone to talk to, I would appreciate an introduction. Łukasz or Antoine may have someone in mind.21:11
* cdent is inspecting his mental rolodex21:11
dhellmannfungi : yeah, I wanted to talk about the bylaws as the source of the TC authority, but didn't want to focus on that too much because the PSF is very much *not* involved in python-dev21:12
TheJuliafungi: ++ re trouble accessing irc21:12
dhellmannfungi, TheJulia : how about the phrasing on lines 85-86 now?21:13
dhellmannthanks, this is good feedback21:15
TheJuliadhellmann: lgtm21:15
fungithank _you_ for helping spread an accurate picture of our governance model to adjacent communities!21:15
persiaOn the definition of "APC" in that document.  Maybe "having a significant documented contribution to a project team within the last six months." instead of "having a change merged within the last 6 months in a repository managed by a project team." to account for extra-apcs?21:22
fungiwell, it's not the last six months either, it's the last *two* development cycles (which gets roughly interpreted by election officials to ~1 elapsed year in most cases)21:23
persiaGood point (although I believe I flubbed that for the recent PTL election)21:25
fungiwell, we got it corrected. no harm, no foul ;)21:25
persiaBut that it is wrong again here makes me wonder if we, as a community, have fully internalised the two-cycle-lagging nature of these things.  Makes me vaguely curious how many folk benefit from that (in that they were inactive for the past 6 months, but previously active and still engaged).21:28
fungiit's a good point, but (at least for the tc electorate) would require a foundation bylaws change to alter in any significant way21:29
fungii think we could choose to define the ptl electorates on a single-cycle qualifying window, but that would then make them more inconsistent with the technical committee electorate21:30
fungieven major alterations of ptl electorate definitions would only require supermajority of the tc to adjust the charter21:31
TheJuliaOut of curiosity, since I'm trying to work up a schedule for ironic's sessions. Do we have an explicit time set for lunch?21:32
TheJuliaeverything I've seen just says "lunch"21:33
fungitime is an illusion; lunchtime doubly so21:33
* TheJulia grumbles21:33
fungiyeah, so far the schedule is vague on lunch start/end times21:34
fungii'll ask around21:34
dtroyerThe Adams is strong in fungi today…21:34
fungioh no, not again21:35
dtroyerdhellmann: that's a great summary, and I happen to know some folk who will also benefit from understanding that better21:35
dhellmanndtroyer : nice, I'm glad it will be reusable21:36
dhellmannpersia , fungi : I copied the definition of APC incorrectly21:36
* fungi has done six exciting things today, but instead of milliways will be rounding it out with a pan-galactic gargleblaster21:36
dhellmannI've fixed that now21:37
* persia can never get the lemon to wrap around the gold brick properly21:37
persiadhellmann: Still says "having a change merged".  Is that what we want?21:38
dhellmannthat's what that says21:38
fungiTheJulia: i'm told lunch is officially 12:30-1:30mdt (they're working on getting the detailed schedule live)21:40
persiaHrm.  I wonder how we justify extra-ATCs then.  I like having them, and I've enjoyed the couple times I've been one, but ...21:40
* fungi combs the bylaws again21:41
persiahttps://governance.openstack.org/tc/reference/charter.html#voters-for-tc-seats-atc has the ATC exception.  Maybe that language should also be in the APC section?21:41
fungii'm more concerned with confirming that the bylaws do actually allow for non-code contributors voting for tc seats, now that you bring it up21:42
persiaOr maybe ATCs aren't supposed to be able to vote for PTL (to disincentivise PTLs from stuffing the rolls with lots of appointed APCs)?21:42
persiaFor TC, there is a clear process, requiring TC approval (likely covered by having the extra-atcs defined in the governance repo)21:43
persiaFor APC, things are murkier.21:43
fungiahh, yeah, appendix 4 section 3 subsection b paragraph ii covers it21:43
fungi"An Individual Member who has made only other technical contributions to the OpenStack Core Project (such as bug triagers and technical documentation writers) can apply to the chair of the Technical Committee to become an ATC. The final approval of such application shall be approved by a vote of the Technical Committee. The term shall be for three hundred and sixty five days after the date of approval21:43
fungiof the application by the Technical Committee."21:43
persiaParagraph i there confuses me a bit though.  It uses both "two prior release cycles" and "three hundred and sixty five days".21:45
persiaThese are commonly not the same, given that releases tend to be scheduled for a specific day of the week, so even if it is the same week that a release happens, that's only 364 days.21:46
fungiit does. we've treated them as roughly the same21:46
persiaUntil someone complains, I suppose :)21:46
fungiif an election official wanted to come up with a defensibly more correct interpretation of that wording, i would readily +121:47
* persia puts that somewhere in the list of things that would be good to do21:47
fungilike, two previous cycles of 365 days whichever is longer?21:48
persiaOr maybe just drop the second sentence of paragraph (i).21:51
fungiwell, altering the _wording_ of that sentence will require a vote of the foundation members, so sayeth foundation legal counsel when we wanted to fix the glaring typo at the beginning of that paragraph21:53
fungimaybe we weigh in with that when they propose the draft update21:53
fungialtering the _interpretation_ as long as it's still a defensible interpretation is something we can do more easily21:54
persiaI think I like the interpretation of "roughly the same", so long as it continues to get no complaints.22:02
dhellmanntc-members: I expect to propose that draft tomorrow morning my time (~15 hours). I will post the link to the PR in case any of you want to follow along22:14
cdentdhellmann: I'll give it another read through in my morning (~9 hours) after I've coffeed up22:16
cdentuntil then, good night22:16
*** cdent has quit IRC22:16
dhellmannthanks, and good night22:17
fungidhellmann: i would love to follow along, thanks!22:45

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!