| pabelanger | and the preliminary results are in: http://lists.openstack.org/pipermail/openstack-dev/2018-March/128626.html | 00:33 |
|---|---|---|
| pabelanger | ttx: fungi: how does one now notify foundation for trade mark check for the naming? | 00:34 |
| dmsimard | aw yeah, solar | 00:36 |
| * dmsimard cheers | 00:36 | |
| *** diablo_rojo has quit IRC | 00:43 | |
| persia | The results of trademark searches are informative: while I didn't check the areas of interest, it takes a while to get far enough down on the list to get a term that isn't registered somewhere. Now to wait for qualified counsel. | 00:44 |
| *** diablo_rojo has joined #openstack-tc | 00:45 | |
| *** annabelleB has quit IRC | 00:48 | |
| *** annabelleB has joined #openstack-tc | 00:54 | |
| *** annabelleB has quit IRC | 01:02 | |
| *** harlowja has quit IRC | 01:31 | |
| *** liujiong has joined #openstack-tc | 01:46 | |
| *** liujiong has quit IRC | 01:50 | |
| *** mriedem has quit IRC | 02:07 | |
| fungi | pabelanger: i've brought your poll results to their attention, thanks! | 02:20 |
| fungi | stein was a close second | 02:24 |
| * fungi is not so secretly hoping stein still becomes the top _viable_ pick | 02:24 | |
| fungi | speaking of people who have the freedom to collaborate at odd hours, it's time i started this last kernel rebuild and let it run whilst i get some shuteye | 02:30 |
| *** harlowja has joined #openstack-tc | 03:32 | |
| *** harlowja has quit IRC | 03:54 | |
| *** persia has quit IRC | 04:13 | |
| *** persia has joined #openstack-tc | 04:14 | |
| *** slashnick has joined #openstack-tc | 05:07 | |
| *** slashnick has left #openstack-tc | 05:10 | |
| *** rosmaita has quit IRC | 05:18 | |
| *** diablo_rojo has quit IRC | 07:25 | |
| *** jungleboyj has quit IRC | 07:25 | |
| *** zaneb has quit IRC | 07:25 | |
| *** jungleboyj has joined #openstack-tc | 07:27 | |
| *** zaneb has joined #openstack-tc | 07:28 | |
| *** diablo_rojo has joined #openstack-tc | 07:28 | |
| *** jpich has joined #openstack-tc | 08:54 | |
| *** dtantsur|afk is now known as dtantsur | 10:36 | |
| *** dklyle has quit IRC | 11:49 | |
| *** rosmaita has joined #openstack-tc | 12:08 | |
| *** cdent has joined #openstack-tc | 12:34 | |
| *** mriedem has joined #openstack-tc | 12:50 | |
| *** dmsimard has quit IRC | 13:21 | |
| *** dtantsur is now known as dtantsur|brb | 13:21 | |
| smcginnis | In my amatuer verification, it looks like "solar" should be OK - http://tiny.cc/jgr0ry | 13:31 |
| smcginnis | Well that didn't work well. | 13:32 |
| *** dmsimard has joined #openstack-tc | 13:34 | |
| persia | smcginnis: Depends on the field-of-interest considered by the vetting council. There is coverage for scientific data analytic systems, electronic account services for finance, large computer networks for finance, application service provider featuring computer software for providing access to via a computer network, etc. | 13:40 |
| persia | some of that may overlap with some of openstack. I am not qualified to have an opinion. | 13:41 |
| smcginnis | Good point. | 13:41 |
| smcginnis | I suppose that could be tricky with things like the "Solar Software Summit". | 13:41 |
| fungi | also, in this case the vetting council is a vetting counsel | 13:51 |
| fungi | at least initially, and then i suppose a council when marketing weighs in | 13:52 |
| smcginnis | They can leave the results on the console. | 14:04 |
| * smcginnis goes away before he's hit | 14:04 | |
| persia | fungi: Singular? Interesting. I would have expected a council of councillors qualified to provide counsel | 14:08 |
| smcginnis | It would be a consolation to fungi if stein is picked. | 14:09 |
| fungi | persia: i was referring to the foundation's "legal counsel" (which may consist of multiple counselors) | 14:10 |
| fungi | or rather, may be provided by multiple counselors | 14:11 |
| persia | Ah, yes, counselor, rather than councillor. | 14:11 |
| persia | I was mostly having fun with words, although it is my experience that often folk like to have their opinions checked for this sort of thing. | 14:12 |
| fungi | but you're right in that it's not just limited to legal counsel, but also passes through marketing and public relations vetting as well so we avoid names which may be legally unencumbered but reflect poorly on the project for other reasons | 14:13 |
| fungi | (c.f., the "lemming" debate) | 14:13 |
| *** hongbin has joined #openstack-tc | 14:13 | |
| * fungi still wanted lemming to win, has poor taste in names | 14:13 | |
| *** dtantsur|brb is now known as dtantsur | 14:15 | |
| *** dklyle has joined #openstack-tc | 14:15 | |
| persia | I don't recall precisely, but when the results were posted, I was surprised how far I had to go down the list before finding an awkward reference. I do hope the folk actually vetting are less wary of potential complications. | 14:18 |
| persia | Err, before not finding an awkward reference, rather | 14:18 |
| fungi | they've generally been fairly forgiving in the past | 14:18 |
| fungi | i worry about spandau mainly for the wwii era use of spandau prison (which in english searches at least is the first hit for "spandau") as a forerunner of the nazi concentration camps | 14:20 |
| hongbin | hi, are tc going to move the links as proposed by this patch https://review.openstack.org/#/c/555179/ ? | 14:22 |
| fungi | so might carry negative connotation for reasons similar to why we crossed off meiji | 14:22 |
| persia | "meiji" makes me think of chocolate, but I do recognise I may have been exposed to different media than some folk. | 14:24 |
| persia | Plus Mitaka is really cool, because 三鷹 contains a character officially banned by the education ministry. | 14:25 |
| fungi | nice! | 14:25 |
| persia | (it means "Hawk") | 14:25 |
| fungi | is the alternative to just spell it out in kana? | 14:26 |
| persia | But it is considered too complicated to teach in school. | 14:26 |
| persia | Yes, the official government position is that it should be written in kana. There is an exception for proper names, which is how the district preserves the character (and so it was on swag) | 14:26 |
| fungi | hongbin: i have no idea. someone has been going around proposing lots of tiny changes to canonicalize urls which are currently referring to redirects. i tend to ignore those changes but others may end up reviewing/approving them at some point | 14:27 |
| hongbin | fungi: ack, thanks | 14:28 |
| fungi | it seems to me more like the sort of change someone would make when generally updating that file rather than being important enough to change on its own, but i recognize that my opinion is not necessarily shared by everyone | 14:29 |
| cdent | tc-members and the rest of the world, it's office hours time | 15:00 |
| EmilienM | o/ | 15:00 |
| fungi | you just beat me to the enter key | 15:00 |
| cmurphy | hiya | 15:00 |
| TheJulia | o/ | 15:00 |
| smcginnis | Howdy. | 15:00 |
| ttx | o/ | 15:02 |
| dhellmann | o/ | 15:02 |
| ttx | We have a bunch of changes that will be finally approved tomorrow, so now is the last chance to object to those | 15:02 |
| ttx | That includes the EM proposal, the Interop test location proposal | 15:03 |
| ttx | The PowerStackers rename | 15:03 |
| smcginnis | Ajutant seems like it is going to need a lot more discussion. | 15:04 |
| ttx | Yes... triggers a bit of "product fit" discussions we ahve shied away from historically | 15:04 |
| ttx | but which I think we should have in that case | 15:04 |
| ttx | Like is OpenStack better with it in or out | 15:04 |
| ttx | more or less confusing, more or less interoperable... | 15:05 |
| EmilienM | smcginnis: I was about to propose we discuss it today :) | 15:05 |
| smcginnis | Agree. My biggest concern is the overlap with mistral. It sounds like it can do a lot of great things, but I need a better picture of what it can do that mistral can't. | 15:05 |
| dhellmann | I'm still trying to understand exactly what Adjutant *is* | 15:05 |
| smcginnis | And why it would be better to have a new project for those differences vs just working within mistral. | 15:05 |
| fungi | i like that it brings a missing feature for public cloud providers: specifically, self-service account creation | 15:05 |
| ttx | I know what it was, but it seems to have grown a more fuzzy scope | 15:05 |
| fungi | but i agree it seems to suffer from scope creep | 15:06 |
| smcginnis | I guess that's the problem with a general purpose and flexible framework. It can be interpreted in many different ways depending on your desired use case. | 15:06 |
| fungi | that age old developer trap of "ooh, i bet i can generalize this api so it can be used for anything" | 15:06 |
| smcginnis | A whole herd of elephants. | 15:06 |
| dhellmann | fungi : i would *love* the project if it was just about account management features | 15:06 |
| smcginnis | dhellmann: ++ | 15:07 |
| ttx | Also before we dive into that rabbit herring -- wanted to plug https://etherpad.openstack.org/p/YVR-forum-TC-sessions again | 15:07 |
| smcginnis | I do think that is a commonly needed and missing component for us. | 15:07 |
| EmilienM | https://github.com/openstack/adjutant/blob/master/adjutant/actions/v1/resources.py is what worried me the most | 15:07 |
| EmilienM | it's doing things that you can already do with a Mistral workflow (and like d0ugal said, we already do it well in TripleO) | 15:07 |
| fungi | the adjutant debate is reminding me of concerns we raised with glare (and its desire to be a general purpose system for "storing stuff") | 15:07 |
| pabelanger | o/ | 15:08 |
| johnthetubaguy | EmilienM: isn't that also done by the "give me a network feature", well sort of, but I get your point | 15:09 |
| smcginnis | fungi: Same. | 15:09 |
| fungi | ttx: fail. you can't lead with the rathole topic and then expect to steer us to paperwork ;) | 15:09 |
| EmilienM | johnthetubaguy: yeah exactly | 15:09 |
| pabelanger | dhellmann: fungi: +1 to just account management features | 15:09 |
| johnthetubaguy | been working with some folks around federation recently, so the account management is a hot topic for me, I should look more closely | 15:10 |
| jroll | so the quota change request got me interested in adjutant, especially given it has an admin approval workflow - can mistral do things like that? also I could be wrong, but I seem to remember adjutant could take actions or sync data between regions or openstack installations, can mistral do those things? | 15:12 |
| EmilienM | i'm calling mistral folks | 15:13 |
| *** thrash has joined #openstack-tc | 15:13 | |
| EmilienM | thrash: hey | 15:13 |
| thrash | EmilienM: hi | 15:13 |
| johnthetubaguy | jroll: isn't that kingbird or something | 15:14 |
| jroll | we have lots of downstream code to handle syncing accounts and such between our openstack clusters, I'd love to have that upstream (and the pluggability is helpful for some of the craziness we do) | 15:14 |
| EmilienM | thrash: we're discussing about https://review.openstack.org/#/c/553643/ and how it overlaps with what provides Mistral already | 15:14 |
| jroll | johnthetubaguy: oh my, I hadn't seen this :) | 15:14 |
| *** d0ugal has joined #openstack-tc | 15:14 | |
| d0ugal | EmilienM: Hey | 15:14 |
| ttx | the rabbit hole now can fit the elephant in the room | 15:14 |
| EmilienM | d0ugal: hey | 15:14 |
| johnthetubaguy | jroll: I have a feeling we need to get lots of interested people in the same room here, somehow! | 15:14 |
| TheJulia | ttx: Just an elephant? | 15:15 |
| EmilienM | d0ugal: we're discussing about https://review.openstack.org/#/c/553643/ today | 15:15 |
| jroll | johnthetubaguy: story of our lives :) | 15:15 |
| * johnthetubaguy nods | 15:15 | |
| * dhellmann wonders about ttx's animal menagerie theme for today | 15:15 | |
| EmilienM | d0ugal, thrash: one of the question dropped here was if mistral was able to manage admin resources/workflows, also if mistral was able to take action between regions or openstack setups (which seems to be doable in adjutant) | 15:16 |
| d0ugal | EmilienM: I think the answer to both is yes | 15:16 |
| jroll | d0ugal: admin approval workflows, to be specific, e.g. user requests a quota change and an admin has to approve | 15:16 |
| fungi | let's not let m o r d r e d know about the animal menagerie | 15:16 |
| jroll | heh, wet cats are so fun though :P | 15:17 |
| * fungi has had enough wet cats for one week ;) | 15:17 | |
| d0ugal | jroll: I'm not sure what specific requirements that would have? | 15:17 |
| TheJulia | heh | 15:17 |
| cmurphy | jroll: why do you sync accounts rather than using a shared external identity provider? | 15:17 |
| persia | cmurphy: Not everyone supports shared external providers? | 15:18 |
| fungi | but i suppose the question is why someone would choose not to support shared external providers | 15:18 |
| jroll | cmurphy: I don't honestly know all the details well enough to say, we do have an external provider but there's also data we need to sync I guess | 15:18 |
| * jroll is still new here | 15:19 | |
| dhellmann | after my initial confusion, I'm less concerned with the overlap with mistral's features than I am with the sense that adjutant is being set up as a "put all of your non-interoperable APIs here" service | 15:19 |
| fungi | that way lies madness | 15:19 |
| TheJulia | cmurphy: it could just be operational or security requirements, like environments could sync/live and have data and some could be abandoned with only older accounts from before the sync was turned off | 15:19 |
| ttx | dhellmann: I'm a bit concerned by both | 15:19 |
| thrash | dhellmann: I would agree with that... In my very small about of reading... | 15:20 |
| dhellmann | the feature that distinguishes mistral is that cloud consumers can upload workflows | 15:20 |
| ttx | Also lack of clear focus is not the best way to attract new contributors | 15:20 |
| dhellmann | so that's workflow as a service, vs. deployer workflows | 15:20 |
| thrash | *amount | 15:20 |
| cdent | dhellmann: yeah, I think that's main distinction | 15:20 |
| cmurphy | TheJulia: i'm trying to understand the specific case and how keystone isn't able to fulfill it with what it provides | 15:21 |
| persia | dhellmann: That is probably a distinction that is meaningful to many audiences. Is it sufficient to avoid feature overlap? | 15:21 |
| TheJulia | cmurphy: makes sense | 15:21 |
| dhellmann | persia : it avoids feature overlap, but opens up the issues of interoperability and lack of focus | 15:21 |
| dhellmann | is adjutant *a* service, or a framework for building lots of different services? | 15:21 |
| persia | cmurphy: The use case I heard most from Public Cloud is self-service signup, which they reported was hard in keystone. That might be fixed by now. | 15:21 |
| dhellmann | and if it's lots of services, how does that fit into the interoperability picture? | 15:22 |
| fungi | what was the network-oriented api project which distinguished itself by supporting extensible api definitions? now i'm forgetting the name | 15:22 |
| ttx | dhellmann: one question I haven't dived into yet is whether Adjutant functions could have been implemented in Mistral, and if yes, why they have been done separately | 15:22 |
| d0ugal | dhellmann: I got the feeling it wasn't a service, not like the others | 15:22 |
| dhellmann | fungi : I think I know the one you mean but can't remember the name either. I think that one was extensible as a development feature, rather than a deployment feature. | 15:22 |
| cmurphy | persia: we don't have self-service signup solved, but i was more curious about the syncing | 15:22 |
| dhellmann | at least that was the messaging I got from one of their more recent presentations | 15:22 |
| dhellmann | if it's the same service | 15:22 |
| johnthetubaguy | persia: cmurphy: FWIW, been looking at how EGI uses co-manage to delegate who belongs in your project, in the external IdP, but not sure that is the case here | 15:23 |
| dhellmann | d0ugal : interesting. I got the opposite impression. | 15:23 |
| ttx | If it's a way to easily add one-off APIs I feel like it will hurt interoperability | 15:24 |
| d0ugal | dhellmann: well, I mean, it is a service - but for operators only I guess. That is probably why it feels different | 15:24 |
| thrash | I think the part that strikes me the most is the use of the exact same terms as Mistral... Workflow, Tasks, Actions. | 15:24 |
| dhellmann | d0ugal : except that users trigger the APIs, right? | 15:25 |
| dhellmann | "I need my password reset" was one of the examples | 15:25 |
| d0ugal | right | 15:25 |
| d0ugal | I think I'm finding it hard to explain what I mean, I tried in my comment on the review. | 15:25 |
| dhellmann | now if there was a canonical password reset API with deployer-specific implementation, that would be OK | 15:25 |
| thrash | dhellmann: that is definitely solvable with Mistral. | 15:25 |
| d0ugal | thrash: the naming could be very confusing. | 15:25 |
| cdent | dhellmann: the apis are behind web pages was my read | 15:26 |
| dhellmann | but if every deployer writes their own thing and the APIs are all different, and then maybe besides account management there's a "do something random to my network" API then that starts to feel odd | 15:26 |
| cdent | so human interaction, not code interaction | 15:26 |
| johnthetubaguy | cdent: they do ship a CLI though | 15:26 |
| dhellmann | cdent : ok, that's interesting. I assumed there would have to be a UI, but didn't pick up that this service was providing the UI, too | 15:26 |
| ttx | It's definitely one we need to dive into before making a call | 15:27 |
| johnthetubaguy | how were you thinking we "dive into" it? schedule in person discussion? | 15:28 |
| cdent | dhellmann: I think the difficult is around what it can do versus what it does do | 15:28 |
| persia | Are the Adjutant folk expected to be represented at Forum? Maybe that could be added to the TC session list? | 15:28 |
| EmilienM | summit is not that far, we could maybe discuss there? | 15:28 |
| EmilienM | forum* | 15:28 |
| dhellmann | cdent : as I said on the review, all of our services are written in a language we can use to implement anything. I'm more concerned with having a good idea of what the team may end up wanting to make it do | 15:29 |
| pabelanger | yah, with CLI tools, it is natural for me to say, maybe this is something ansible could do. But behind a web page, doesn't really work | 15:29 |
| * cdent nods at dhellmann | 15:29 | |
| pabelanger | unless you look into something like tower to trigger something via API | 15:29 |
| dhellmann | the way it's pitched right now it doesn't feel like any of our other deliverables | 15:29 |
| dhellmann | it's not a library | 15:30 |
| dhellmann | it's not a client | 15:30 |
| dhellmann | though it has one | 15:30 |
| dhellmann | it's not a fixed-API service | 15:30 |
| dhellmann | it's not a "Foo as a service" thing | 15:30 |
| d0ugal | so, what is it? :) | 15:30 |
| dhellmann | right. and maybe we need a new category? but maybe not. | 15:31 |
| johnthetubaguy | isn't it a set of REST APIs with a client and a UI? | 15:31 |
| cdent | I'm reminded of the discussion here: https://github.com/cncf/toc/issues/85 | 15:31 |
| thrash | feels more like a bolt-on than a core component | 15:31 |
| cdent | maybe we should just wait until there is demand, rather than trying to bring it about | 15:31 |
| ttx | If it's an API factory, it's more like Gluon, and back then we rejected it because it was more of a differentiation engine | 15:31 |
| dhellmann | johnthetubaguy : is it? all of the docs talk about deployers adding their own rest APIs to it. | 15:32 |
| fungi | dhellmann: aha! now i remember what it is (was?) called: gluon | 15:32 |
| dhellmann | ttx: thank you, yes, gluon | 15:32 |
| dhellmann | :-) | 15:32 |
| fungi | oh, wow ttx brought it up just as i finally found the name via a web search | 15:32 |
| dhellmann | from your fingers, through google, to ttx's brain | 15:33 |
| dhellmann | scary | 15:33 |
| fungi | well, thankfully not google, but yes | 15:33 |
| fungi | (maybe duckduckgo got it from google, i don't really know) | 15:33 |
| smcginnis | So most of the worlds search engines are powered by ttx's brain? | 15:33 |
| johnthetubaguy | dhellmann: true, I am unsure | 15:34 |
| ttx | I prefer DuckDuckGo | 15:34 |
| fungi | smcginnis: i think that must be it | 15:34 |
| ttx | smcginnis: reality is actually a lot more scary | 15:34 |
| * dhellmann wants a "Hey, Thierry" voice assistant | 15:34 | |
| fungi | speaking of which, it does seem that gluon is still under (slow but continuing) development. most recent activity was early january | 15:34 |
| pabelanger | I'd also be interested if there was any negative feedback from using the public polling for 'S' release, over all I think it went well. But some people did complain about 'already voted' due to shared IP addressing | 15:35 |
| ttx | I'm very happy that Gluon exists... I just think it would hurt "OpenStack" if it was made a part of it | 15:35 |
| ttx | by introducing competing APIs and fostering non-interoperability | 15:36 |
| *** annabelleB has joined #openstack-tc | 15:36 | |
| ttx | and my fear (but I might have understood badly, which is why I want to dive into code and run Adjutant a bit) is that Adjutant is in the same category | 15:36 |
| EmilienM | If I reached out Adjutant's devs, would we have space at Forum for human interactions? | 15:37 |
| ttx | EmilienM: I would be definitely interested | 15:37 |
| ttx | +Mistral devs | 15:37 |
| ttx | CatalystIT is NZ-based though, so not sure they will travel to Vancouver | 15:37 |
| EmilienM | d0ugal, thrash : do you know who (if not you?) is going from Mistral folks? | 15:37 |
| d0ugal | EmilienM: hopefully thrash is going | 15:38 |
| thrash | EmilienM: I'm not slated to go just yet. | 15:38 |
| ttx | cdent: the main difference with that cncf toc issue is that they are blessing separate open source projects, while we are arguably discussing addition of components into a product | 15:39 |
| EmilienM | first thing is I can reach them out and see if they are going | 15:39 |
| EmilienM | then I could try to arrange some time to meet and talk, maybe 30min or something | 15:39 |
| ttx | EmilienM: I'm a bit concerned to stall that discussion until May | 15:40 |
| persia | My memory of the start of the gluon project was that the other API didn't meet needs, As the other API matures, the need for gluon reduces. (I haven't been involved with gluon since 2016) | 15:40 |
| EmilienM | ttx: I'm sure we could make progress in the meantime, maybe we could ask them to join a TC office hour, to have interaction on IRC maybe? | 15:41 |
| ttx | EmilienM: I would continue async and book a slot for discussing it at the Forum, just in case | 15:41 |
| ttx | I don't feel like we NEED in person discussion to make progress right now, the discussion is just starting | 15:42 |
| ttx | We just expect to need in-person discussion at one point, so let's provision space for it | 15:42 |
| EmilienM | ttx: how can we book a slot for the Forum? | 15:43 |
| jroll | it seems like we have a couple of solid questions - 1) what does this do that mistral cannot? 2) how do we ensure interoperability (or can we at all)? seems like someone could write these up on the ML to start | 15:43 |
| ttx | EmilienM: glad you ask! | 15:43 |
| fungi | pabelanger: my experience with the public polling via civs model is that it worked well. i'm still mildly concerned that we need to be looking into at least having the ability to run an instance of civs ourselves in case something ever happens to the one at cornell | 15:43 |
| ttx | EmilienM: add it to https://etherpad.openstack.org/p/YVR-forum-TC-sessions | 15:43 |
| jroll | (I say ML because I see no emails ever about adjutant) | 15:43 |
| persia | +1 on self-hosting CIVS. | 15:43 |
| cdent | ttx, yes, I know, it was only a tangential reminder, more as ammo for asking the usual question of "what if we just wait until there's demand?" | 15:44 |
| fungi | persia: the -1 on self-hosting civs is that it currently serves as an (ostensibly) unbiased third party when we're conducting elections | 15:44 |
| fungi | so there's of course a trade-off | 15:44 |
| jroll | cdent: how much demand would we wait for? clearly at least one cloud provider demands this | 15:45 |
| dhellmann | jroll : good point. Adrian has been pretty responsive on the review, but it's not necessarily the best forum for more detailed discussion. | 15:45 |
| ttx | ++ | 15:45 |
| persia | fungi: The ostensibility is the part that makes me not care, but I can see how perceptions may differ | 15:45 |
| pabelanger | fungi: yah, assuming everybody is okay with self hosting, might be a good thing to do | 15:45 |
| cdent | jroll: I don't mean demand for use. I'm suggesting that we should wait until there are plenty of users before something becomes official | 15:45 |
| fungi | persia: though maybe as the ci/cd community absorbs more of the service-running side of the infra team, we can present a little more potential for appearing unbiased about hosted election tooling | 15:46 |
| cdent | And I'm not suggesting that as "the right thing to do" but rather as a strawman | 15:46 |
| cdent | to see how it feels | 15:46 |
| jroll | cdent: I see :) | 15:46 |
| EmilienM | ttx: done | 15:46 |
| fungi | jroll: the review suggested at least two cloud providers are running it actually | 15:46 |
| clarkb | pabelanger: my take away from it was that only 300 something people voted which implies maybe we don't need to do a massive 20k user poll? | 15:46 |
| cdent | Generally speaking I'd prefer we just accept people who aren't obviously horrible, if they came with resources | 15:46 |
| dhellmann | I think if this existed when we were building the cloud at Dreamhost we would have been very interested in it. | 15:47 |
| persia | cdent: Even if the resources are constrained to their special corner of the table? | 15:47 |
| clarkb | if we want to go back to private we could make the poll opt in via a sign up list maybe? I imagine the list of people interested would be manageable in that case | 15:47 |
| cdent | persia: In this case I was thining rather globally like "ways to compensate for their ci load" | 15:48 |
| persia | clarkb: public poll or massive-private-poll are both easier to administer | 15:48 |
| pabelanger | clarkb: yah, that too. Seemed like a very small number of users actually voted. I wonder if that was also due to not emailing them directly | 15:48 |
| cdent | which I agree is rather vague | 15:48 |
| clarkb | persia: except that civs melts down when we ask them to send 20k emails | 15:48 |
| clarkb | persia: but 400 emails should be doable | 15:48 |
| dhellmann | how many people voted in the last naming poll? | 15:49 |
| dhellmann | if we're getting roughly the same numbers, maybe a public poll is just fine? | 15:49 |
| pabelanger | dhellmann: not sure, looking for that info | 15:49 |
| persia | clarkb: Difference being that folk have to coordinate the electoral roles, provide active support to outliers, etc. In practice, it's usually only one or two for the first few days of a poll, but still. | 15:50 |
| pabelanger | https://civs.cs.cornell.edu/cgi-bin/results.pl?id=E_e53f789ff7acc996 | 15:51 |
| pabelanger | 1794 votes for R | 15:51 |
| dhellmann | hmm | 15:51 |
| fungi | i suppose having a ballot land in your inbox does impact turnout | 15:51 |
| dhellmann | the next question is if we care that the turnout levels changed, given that the same number of people (or more) were *able* to vote | 15:52 |
| pabelanger | 1899 for Q and 2681 for P | 15:52 |
| pabelanger | so, we are down a fair bit | 15:52 |
| dhellmann | can we address that with more publicity? maybe emailing the 20k people the link to the public poll? | 15:54 |
| ttx | dhellmann: another issue with public polling is the restriction to one vote per IP | 15:55 |
| persia | dhellmann: We can certainly do that, but we'd need to make arrangements with the spam folk to make sure they knew we were doing it responsibly. | 15:56 |
| pabelanger | if we email 20k people the public link, why not just do the private? | 15:56 |
| dhellmann | ttx: ah, true | 15:56 |
| pabelanger | I think it was the emailing of 20k people was the issue, not voting of 20k people | 15:56 |
| persia | pabelanger: cornell servers take a long time to send that much mail. | 15:56 |
| dhellmann | pabelanger : I thought civs had trouble with that many ballots | 15:56 |
| ttx | dhellmann: not really | 15:56 |
| pabelanger | dhellmann: i think it just too a while like persia says to email them out | 15:56 |
| pabelanger | took* | 15:56 |
| dhellmann | ah | 15:57 |
| ttx | We'd likely need to patch the mail sending routine to make it more reliable | 15:57 |
| ttx | Can't wait to patch some Perl, will remember the days i was young | 15:57 |
| * smcginnis shudders | 15:57 | |
| dhellmann | well, this is the first time we did it. the turnout was lower, but the results seem similar (in that the top contender doesn't in any way evoke the location) | 15:57 |
| pabelanger | Yah, I had high hopes for spree, but landed #3 | 15:58 |
| smcginnis | pabelanger: Based on past naming, that's still got a pretty good chance. | 15:59 |
| pabelanger | smcginnis: agree, lets see what foundation thinks | 15:59 |
| dhellmann | wasn't the top contender "solar"? | 16:00 |
| * dhellmann doesn't have the link handy | 16:00 | |
| jroll | yeah, solar, stein, spree | 16:00 |
| dhellmann | thanks | 16:04 |
| *** harlowja has joined #openstack-tc | 16:04 | |
| fungi | solar, 🍺, spree | 16:08 |
| *** harlowja has quit IRC | 16:09 | |
| smcginnis | fungi: You're really wanting to be able to use that a lot, aren't you? :) | 16:11 |
| fungi | yes | 16:12 |
| * fungi already has ideas for an OpenStein logo | 16:12 | |
| jroll | if stein ends up on top, we'll need to get openstack-branded steins | 16:12 |
| zaneb | fungi: if you can see the beer through it, it's not a stein by definition | 16:12 |
| fungi | bingo | 16:12 |
| fungi | zaneb: details ;) | 16:12 |
| * jroll will help fund the kickstarter | 16:12 | |
| zaneb | lol | 16:12 |
| fungi | zaneb: clearly this means we need to propose a new unicode codepoint | 16:13 |
| fungi | "BEER MUG, OPAQUE" | 16:14 |
| *** thrash is now known as thrash|biab | 16:15 | |
| dhellmann | jroll : if there aren't openstack branded steins regardless of the name, I don't even know what we're doing any more | 16:20 |
| dhellmann | fungi : ++ | 16:20 |
| jroll | dhellmann: true that | 16:21 |
| * fungi suggests to swag store maintainers | 16:21 | |
| * persia mumbles about teetotallers in our midst | 16:25 | |
| * smcginnis notes you can put tea in a stein, as wrong as that would feel | 16:26 | |
| * jroll thinks coffee would be appropriate in a stein | 16:26 | |
| * fungi has so done that | 16:27 | |
| fungi | (very stout coffee) | 16:27 |
| smcginnis | (or a good coffee stout) | 16:27 |
| jroll | as opposed to a very coffee stout :P | 16:27 |
| smcginnis | ;) | 16:27 |
| fungi | think of it as a gradient | 16:27 |
| smcginnis | Kind of like a sundial? You can tell the time of day based on the liquid in the stein? | 16:28 |
| jroll | heh | 16:28 |
| *** masayukig has quit IRC | 16:29 | |
| dhellmann | smcginnis : way to work the "solar" theme back into the conversation. Nicely done. | 16:29 |
| smcginnis | dhellmann: I guess I've just been on a spree today. | 16:29 |
| jroll | booooo | 16:30 |
| * dhellmann slow claps | 16:30 | |
| * cdent shuns smcginnis | 16:30 | |
| smcginnis | cdent: Go ahead. Pretend like you didn't laugh. | 16:31 |
| dhellmann | we should schedule a forum session for a dad joke competition | 16:31 |
| * cdent has no sense of humor | 16:31 | |
| * jroll registers the Production Undercloud Network SIG | 16:32 | |
| *** thrash|biab is now known as thrash | 16:49 | |
| *** cdent has quit IRC | 17:18 | |
| *** diablo_rojo has quit IRC | 17:19 | |
| *** dtantsur is now known as dtantsur|afk | 17:23 | |
| *** diablo_rojo has joined #openstack-tc | 17:24 | |
| *** kumarmn has joined #openstack-tc | 17:25 | |
| *** jpich has quit IRC | 17:31 | |
| *** harlowja has joined #openstack-tc | 18:15 | |
| *** harlowja_ has joined #openstack-tc | 18:17 | |
| *** harlowja has quit IRC | 18:19 | |
| *** kumarmn has quit IRC | 18:20 | |
| *** diablo_rojo has quit IRC | 18:38 | |
| *** diablo_rojo has joined #openstack-tc | 18:40 | |
| *** cdent has joined #openstack-tc | 18:53 | |
| dmsimard | fungi used an emoji ? /me takes screenshot | 19:35 |
| fungi | dmsimard: i don't know what an emoji is (something like an emoticon?) but i did paste an extended unicode codepoint | 19:36 |
| fungi | utf-8 encoded | 19:37 |
| fungi | because the associated glyph for it is evocative of one of our preferred release name candidates | 19:37 |
| dmsimard | fungi: there's even an open source emoji font: https://en.wikipedia.org/wiki/Emoji#General | 19:37 |
| fungi | i've clearly fallen into a millennial trap | 19:41 |
| clarkb | fungi: the problem with that character is source code pro (my current termianl font) doesn't have a glyph for it | 19:41 |
| fungi | sounds like an obvious deficiency in your font choice | 19:42 |
| fungi | (to be fair, the commodore64 font i use in my terminal lacks a glyph for that codepoint as well) | 19:42 |
| *** kumarmn has joined #openstack-tc | 20:29 | |
| *** kumarmn has quit IRC | 20:37 | |
| *** cdent has quit IRC | 20:45 | |
| *** kumarmn has joined #openstack-tc | 20:58 | |
| *** kumarmn has quit IRC | 21:03 | |
| *** hongbin has quit IRC | 23:01 | |
| *** annabelleB has quit IRC | 23:09 | |
| *** annabelleB has joined #openstack-tc | 23:16 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!