| zaneb | zhipeng: as an example, I personally would have merged https://review.openstack.org/565974 because it's a useful, if trivial, change. I probably would never have got around to checking something like that myself because I never look at the README | 00:01 |
|---|---|---|
| scas | the usefulness can be up for debate, but most of the changes i've accepted were generally something additive, albeit trivial | 00:01 |
| zaneb | but then of course there was https://review.openstack.org/576754 and https://review.openstack.org/580435 and apparently it will never end until you merge one or -2 every developer in the whole company :/ | 00:05 |
| zhipeng | So my reply was always, make the trivia ones part of the substantial patch | 00:09 |
| zhipeng | And provide ways for them to contact the team | 00:09 |
| zhipeng | And you know what, they never did | 00:09 |
| zhipeng | Just abandoned and ran away | 00:10 |
| *** jamesmcarthur has joined #openstack-tc | 00:10 | |
| *** jamesmcarthur has quit IRC | 00:15 | |
| *** mriedem_away has quit IRC | 00:18 | |
| *** jamesmcarthur has joined #openstack-tc | 00:31 | |
| *** jamesmcarthur has quit IRC | 00:35 | |
| *** jamesmcarthur has joined #openstack-tc | 00:52 | |
| *** jamesmcarthur has quit IRC | 00:56 | |
| *** Bhujay has joined #openstack-tc | 02:19 | |
| *** Bhujay has quit IRC | 02:19 | |
| *** Bhujay has joined #openstack-tc | 02:20 | |
| *** diablo_rojo has quit IRC | 02:33 | |
| *** Bhujay has quit IRC | 02:59 | |
| *** openstackgerrit has joined #openstack-tc | 05:00 | |
| openstackgerrit | Merged openstack/governance master: Add ansible-role-chrony to TripleO https://review.openstack.org/603516 | 05:00 |
| *** Bhujay has joined #openstack-tc | 05:35 | |
| *** mtreinish has quit IRC | 05:37 | |
| *** mtreinish has joined #openstack-tc | 05:38 | |
| *** mtreinish has quit IRC | 05:56 | |
| *** mtreinish has joined #openstack-tc | 05:59 | |
| *** zaneb has quit IRC | 06:55 | |
| *** diablo_rojo has joined #openstack-tc | 06:55 | |
| *** zaneb has joined #openstack-tc | 06:56 | |
| *** diablo_rojo has quit IRC | 07:27 | |
| *** tosky has joined #openstack-tc | 07:45 | |
| *** jpich has joined #openstack-tc | 07:47 | |
| *** e0ne has joined #openstack-tc | 08:12 | |
| *** dtantsur|afk is now known as dtantsur | 08:36 | |
| *** Bhujay has quit IRC | 09:19 | |
| *** Bhujay has joined #openstack-tc | 09:25 | |
| *** cdent has joined #openstack-tc | 09:44 | |
| *** purplerbot has joined #openstack-tc | 10:08 | |
| *** cdent has quit IRC | 10:08 | |
| *** cdent has joined #openstack-tc | 11:00 | |
| *** tosky__ has joined #openstack-tc | 11:44 | |
| *** tosky has quit IRC | 11:44 | |
| *** tosky has joined #openstack-tc | 11:45 | |
| *** Bhujay has quit IRC | 11:51 | |
| *** Bhujay has joined #openstack-tc | 11:52 | |
| *** Bhujay has quit IRC | 11:53 | |
| *** Bhujay has joined #openstack-tc | 11:53 | |
| *** Bhujay has quit IRC | 11:54 | |
| *** e0ne has quit IRC | 12:09 | |
| *** e0ne has joined #openstack-tc | 12:09 | |
| *** tosky__ has quit IRC | 12:15 | |
| *** mriedem has joined #openstack-tc | 12:22 | |
| *** mriedem has quit IRC | 12:34 | |
| *** mriedem has joined #openstack-tc | 12:49 | |
| *** e0ne has quit IRC | 13:22 | |
| *** e0ne has joined #openstack-tc | 13:26 | |
| dhellmann | not every new contributor is going to be ready to write a substantial patch | 13:32 |
| dhellmann | having lots and lots of these trivial patches can be annoying, and eat up resources, but I also don't see a reason we need 2 reviewers to approve them | 13:33 |
| dhellmann | and if modifying readme.rst runs our full test suite, that tells me our test job configuration is wrong | 13:33 |
| cdent | +2 | 13:34 |
| lbragstad | wasn't there a push to make documentation only jobs only runs documentation builds (as opposed to the entire suite) | 13:35 |
| lbragstad | i wonder if we can find ways to keep doing things like that? | 13:35 |
| dhellmann | I'm sure there are still ways we could tell jobs to ignore certain files | 13:36 |
| cmurphy | that's what this does http://git.openstack.org/cgit/openstack/keystone/tree/.zuul.yaml#n21 | 13:37 |
| dhellmann | that seems like it might be a useful effort with the current CI resource limitations | 13:37 |
| dhellmann | cmurphy : yep, I think we could probably do more of that on some other jobs | 13:37 |
| lbragstad | aha - yeah | 13:37 |
| cdent | what I've heard was that irrelevant files was being hard to generalize due to lack of path consistency and if we could improve that we'd have goodnes | 13:38 |
| zhipeng | Not every padder is new contributor :) | 13:38 |
| dhellmann | a lot of these trivial patches touch the readme.rst, and that's already standard | 13:38 |
| lbragstad | if projects keep their zuul config in tree, they can define irrelevance however they need to, right? | 13:38 |
| dhellmann | zhipeng : I don't understand why I need to care about someone padding stats | 13:39 |
| dhellmann | lbragstad : yes, although we could also update the job definitions themselves to have a global effect | 13:39 |
| lbragstad | true | 13:39 |
| zhipeng | It is true that not everyone cares about it, so just leave it to the projects :P | 13:40 |
| dhellmann | zhipeng : in this case, having projects be aggressive toward contributors gives the community a bad reputation, so I would like to understand why anyone is so concerned | 13:41 |
| zhipeng | The CI meltdown at the end of 2017 ? | 13:43 |
| cmurphy | dhellmann: I would think back to the presentation you gave to the board in yvr about how big a difference one contributor can make and look at that from an outsiders perspective; all of us here know who is really doing the work, but someone like a board member or the media might look at http://stackalytics.com/?metric=commits and assume that melissaml is doing pretty much the same amount of work as | 13:44 |
| cmurphy | mriedem, if they have nothing else to go on | 13:44 |
| zhipeng | cmurphy that is a great one | 13:45 |
| dhellmann | do we think people really look at stackalytics to see who is leading the community or doing work? | 13:45 |
| dhellmann | I'm trying to get past our own annoyance over this to understand if there's really a problem. I hadn't considered that anyone would take stackalytics seriously, so that may be a concern but I don't know how real it is. | 13:47 |
| persia | I have been told "get that company to appear on stackalytics, and we'll speak to their sales team. Until then, don't bother asking." | 13:48 |
| dhellmann | ugh | 13:48 |
| persia | I've also been told "can we have commits with this domain name? We're trying to improve stackalytics". | 13:48 |
| zhipeng | And bonus for each patch | 13:48 |
| dhellmann | I don't care about anyone giving bonuses for patches. The sales thing is more worrying. | 13:49 |
| persia | zhipeng: Yep. I've also seen incentive schemes that give bonus for positive reviews (including +1s) when the patch merges. | 13:49 |
| dhellmann | someone is going to go out of business doing that, so I don't care about that either | 13:49 |
| dtroyer | dhellmann: yes, they still look at stackalytics, even after I've told them for years to treat it like the OpenStack Enquirer | 13:49 |
| mriedem | global irrelevant files doesn't really work fwiw | 13:50 |
| jroll | I would like a bonus per patch :P | 13:50 |
| pabelanger | dhellmann: lbragstad: we've also discussed making doc jobs use less resources in the gate, eg: 1vcpu/1gb vs 8vcpu/8gb but most providers just cap at total number of VMs, so that itself won't help to land things faster | 13:50 |
| zhipeng | It is really mind boggling | 13:50 |
| mriedem | lbragstad: e.g. https://github.com/openstack/nova/blob/master/.zuul.yaml#L14 write once, re-use in all nova dsvm jobs | 13:50 |
| persia | dhellmann: Broadly put, as far as I can tell, the wider audience believes stackalytics is an accurate answer to the question "who controls openstack". | 13:50 |
| mriedem | via reference https://github.com/openstack/nova/blob/master/.zuul.yaml#L41 | 13:50 |
| lbragstad | pabelanger sounds like y'all need some unified limits action | 13:50 |
| cmurphy | i think marketing and sales need some kind of external publishment to point to and as long as we're not giving them a qualitative alternative then stackalytics is the only thing they have to go to | 13:51 |
| pabelanger | lbragstad: I think the current idea is containers, running some place for things like docs / linters. But that is still a while out I'd guess. | 13:51 |
| mriedem | dhellmann: "do we think people really look at stackalytics to see who is leading the community or doing work?" - my understanding is it's a thing in china | 13:51 |
| mriedem | zhipeng: ^ correct? | 13:51 |
| lbragstad | pabelanger sure - that makes sense | 13:52 |
| zhipeng | Yep | 13:52 |
| dtroyer | mriedem: it's a thing in Portland too | 13:52 |
| mriedem | oh intel... | 13:52 |
| mriedem | took me a while there :) | 13:52 |
| mriedem | you mean HILLSBORO?! | 13:52 |
| cmurphy | was gonna say ^ | 13:53 |
| dims_ | LOL | 13:53 |
| persia | cmurphy: It's not just outbound though: that the outbound marketing/sales use stackalytics has caused many purchasing departments to rely on it. | 13:53 |
| dtroyer | I only go there a couple times a year and still can't drive through downtown without making a wrong turn, I don't know the geo | 13:53 |
| zhipeng | It has quantifiable numbers that could not be faked easily, so sales like it | 13:54 |
| persia | zhipeng: That's not fair: sales teams don't care that much about the ability to fake things. They want quantifiable numbers that make them look good (and ideally ones that look hard to fake). | 13:54 |
| dtroyer | all that said, we've seen those same one-off fixes in starlingx repos, they are just hitting every repo in Gerrit. but if they would do the whole repo at once maybe rather than one at a time they might get a bit more attention… | 13:55 |
| mriedem | i definitely know who does it in nova now, and click on the topic branch to see they are pushing the same changes across dozens of repos | 13:55 |
| mriedem | like s/http/https/ in docs | 13:56 |
| mriedem | that was a fun one in rocky | 13:56 |
| TheJulia | jroll: ++ to bonus per patch. :) | 13:56 |
| TheJulia | I do see value in some activities like that, the motivator not as much. In my mind this all goes back to the question of "does it (as in the patch) improve things?" | 13:59 |
| TheJulia | The volume or act of the patch may not be helpful in the moment, but later on those small things do build up | 14:00 |
| lbragstad | if someone does have good intentions, and they are parsing docs converting http -> https (for example), I'd rather suggest they use that time to generate coverage reports and propose more test coverage | 14:01 |
| lbragstad | if they really are looking to get involved with a project | 14:02 |
| mriedem | i've seen changes where someone runs spell check on a single module and then corrects one or two words and pushes that up | 14:02 |
| mriedem | like, in code comments | 14:02 |
| lbragstad | yeah - we've seen that too | 14:02 |
| mriedem | in that case they are definitely not just trying to make things better | 14:02 |
| cmurphy | since there are more people here i'll mention https://etherpad.openstack.org/p/fc-sig-contributor-outreach-message again, looking for how to positively redirect all this energy into more constructive work | 14:04 |
| mriedem | if the contributor, or their company, isn't interested in making positive changes to openstack, that won't do anything | 14:05 |
| mriedem | and in that case, i'm not sure what to do | 14:05 |
| cmurphy | sure, this is trying to assume the best of intentions | 14:05 |
| mriedem | zhipeng: how does one influence negative guanxi for an entire corporation? :) | 14:05 |
| * dhellmann hesitates to suggest publishing stats of useless patches | 14:07 | |
| zhipeng | mriedem like what I've been doing ? lol | 14:13 |
| cdent | If we have unlimited CI resources would people be more happy to receive typo (and other trivial) fixes | 14:13 |
| cdent | s/have/had/ | 14:13 |
| persia | mriedem: How does an organisation new to openstack differentiate between "positive change" and "change"? | 14:13 |
| cdent | Because at some point we'd like those typos to go away, yes? | 14:14 |
| cmurphy | CI is not the only scarce resource | 14:14 |
| cdent | sure | 14:14 |
| cdent | but I'm using the term "trivial" on purpose | 14:14 |
| cmurphy | what is most frustrating to me are the "trivial" changes that are actually wrong, so you either spend several patchsets on it or it sits in the queue untouched after the author abandoned it | 14:15 |
| cmurphy | i'm happy to single core approve correct trivial patches | 14:16 |
| lbragstad | to me - figuring out positive change versus just change is documented in the teams priority list | 14:16 |
| lbragstad | if an organization wants to be influential in a specific project, they'll need to interface with the community either synchronously or asynchronously | 14:17 |
| persia | lbragstad: To me, that implies that only those who have "proven themselves" are permitted to decide what to do. Also, it's trivial to game those lists | 14:17 |
| lbragstad | i might have misunderstood the question | 14:17 |
| persia | There exists another community that was going to do "cloud" that required new contributors to spend six months working on "team priorities" before being permitted to suggest new features. It wasn't particularly successful at attracting contributors from those who weren't metal sponsors. | 14:18 |
| mriedem | cdent: we clearly don't have unlimited CI resources | 14:19 |
| mriedem | as can be seen from this week and last | 14:19 |
| mriedem | 600+ changes in the check queue yesterday afternoon | 14:19 |
| cdent | mriedem: it's a thought experiment | 14:19 |
| cdent | to try and tease out what people's perspectives are | 14:19 |
| cdent | i'm well aware we don't have infinite ci | 14:19 |
| cmurphy | i don't think all organizations have the end game of being influential in a specific openstack project, more like they have an end game of being influential in the market | 14:19 |
| mriedem | cmurphy: yes i think that's exactly correct | 14:20 |
| cmurphy | openstack is more or less mature enough that you can passively consume it and have a pretty good cloud | 14:20 |
| mriedem | i've actually been pleasantly surprised lately by more non-huawei chinese devs contributing good fixes and reviews to nova | 14:20 |
| persia | cmurphy: I think that is trye for organisations. I think it's less true for specific managers tasked with activity (although I think there is a strong argument they were "tasked wrong"). | 14:20 |
| mriedem | and i've seen those same devs calling out the "garbage" changes | 14:20 |
| zhipeng | hiahia | 14:20 |
| smcginnis | cmurphy: I think "being influentual in the market" hits on exactly what I've seen. | 14:25 |
| dhellmann | maybe we should be doing a better job of publishing the stats for the things we actually care about, instead of leaving it up to the market to look at the raw stats published by a tool/group over which we apparently have limited influence | 14:27 |
| lbragstad | like - completed specification or bug fix ratio? | 14:27 |
| cmurphy | dhellmann: ++ | 14:28 |
| persia | dhellmann: That would be lovely. | 14:28 |
| dhellmann | the goal-tools repo has tools for pulling contribution info on filtered queries. so we could focus on the patches we care about because they applied to a specific topic, or set of files, or whatever | 14:28 |
| dhellmann | the tools are rough, and produce csv which I then imported into google apps to make charts, so they would need some work | 14:28 |
| smcginnis | I think that was kind of the idea with trying to get more a narative account from member companies about how they are contributing. | 14:29 |
| smcginnis | Because looking at stackalytics is a very very skewed view of the world. | 14:29 |
| dhellmann | yeah, that would be the other side of it | 14:29 |
| persia | Another approach is post-hoc: flag "significant" patches in some way, and create a post-hoc burndown that shows patches landed over a cycle, with different colors for "significant" and "insignificant", and reports by affiliation or something. | 14:29 |
| dhellmann | "here's what we think was valuable, what did you think was valuable?" | 14:29 |
| lbragstad | who's going to be answering that question? | 14:30 |
| dhellmann | we asked the board to ask member companies to respond as part of a survey | 14:30 |
| dhellmann | the answers we got back were all over the place, because the question wasn't very focused | 14:30 |
| zhipeng | Also user survey ? | 14:30 |
| dhellmann | they went into the last user survey, yes | 14:31 |
| persia | Another interesting tidbit: I occasionally see comments like "so I hope to contribute more closely to $project" in my IRC feeds. This gives me the impression that some folk are told "go work on $project", rather than soemthing I think more positive "go help make $feature work" or "go ensure $deployment is reliable". | 14:31 |
| dhellmann | I think ultimately we'll have more luck influencing if we start celebrating valuable stuff more visibly | 14:32 |
| TheJulia | cmurphy: Reading the fc sig outreach message, I can't help but wonder if something much shorter to engage in discussion might be more fruitful since we don't know their intentions, and a lot of text and links to try and redirect seems like a lot for a non-native english speaker to digest. | 14:32 |
| cdent | (celebrating valuable stuff more visibly)++ | 14:32 |
| smcginnis | dhellmann: ++ | 14:32 |
| dhellmann | we're a bit early for office hours, but since we're already talking about it: | 14:32 |
| TheJulia | Could we do burndown data based upon storyboard so we drive its use even more? | 14:33 |
| TheJulia | And we can do freeform tagging... and that common reporting output pattern would actually drive more consistent use (maybe) | 14:33 |
| cmurphy | TheJulia: that is good feedback | 14:34 |
| dhellmann | I talked with jbryce, ttx, aprice, and lauren at the ptg about doing more of that celebration during keynotes. I had in mind an individual contributor award (to be named) but we could probably fit in some thank yous for companies that contributed significantly to work that teams had prioritized, too | 14:34 |
| dhellmann | I would personally rather focus on individuals, though | 14:35 |
| TheJulia | dhellmann: I like | 14:35 |
| TheJulia | Both would be good, in a sense "thanks to x person from y company for their work on z thing" | 14:35 |
| cdent | I think we should focus on "z thing" that happened to be done by "x person from y company" | 14:35 |
| dhellmann | we also talked about recognizing all of the recipients of the current contributor awards on stage in some form (probably just a slide with names/photos due to the number) | 14:35 |
| TheJulia | cdent: ++ | 14:35 |
| TheJulia | cdent: I was just tossing text out as my brain processed, that does seem much better | 14:36 |
| cdent | because it is almost always the case that "z thing" is the product of many people, and while "x person" was a huge factor, the thing that end-people care about is "z thing" and how it matters to them | 14:36 |
| cdent | _we_ care about "x person" | 14:36 |
| TheJulia | cmurphy: I only thought of shorter sentences because of training I received last week. \o/ | 14:37 |
| dhellmann | if we want to control which behaviors are incentivized, we need to give positive as well as negative feedback | 14:37 |
| cmurphy | TheJulia: my thinking was if they're in this situation (and assuming they have the best intentions) that maybe they don't even know what to ask in order to get started on something productive, so i thought a list of things might be helpful as a jumping off point | 14:39 |
| TheJulia | I suspect it is going to be very situational | 14:42 |
| cmurphy | that's true | 14:43 |
| TheJulia | dhellmann: ++ | 14:43 |
| *** evrardjp has quit IRC | 14:45 | |
| *** evrardjp has joined #openstack-tc | 14:47 | |
| *** annabelleB has joined #openstack-tc | 14:58 | |
| *** evrardjp has quit IRC | 14:58 | |
| cdent | tc-members hours | 15:01 |
| smcginnis | Beat me to it. | 15:01 |
| cdent | <- gunslinger (?) | 15:01 |
| * cmurphy wonders if she is still a tc member | 15:02 | |
| cdent | few more hours? | 15:02 |
| smcginnis | :) | 15:02 |
| smcginnis | Yep, technically the election is still in progress. | 15:03 |
| dhellmann | o/ | 15:03 |
| dims_ | o/ | 15:03 |
| smcginnis | Not that that should change anything. | 15:03 |
| dhellmann | cmurphy : regardless, you're always welcome to join | 15:03 |
| cmurphy | :) | 15:03 |
| dtroyer | there are still comfy seats here in the TC-emeritus section | 15:06 |
| * dtroyer wonders if that is the correct use of latin? | 15:06 | |
| * cmurphy reserves a seat | 15:06 | |
| dhellmann | I've been considering the feedback Alan gave us at the PTG about the joint leadership meeting. I wonder if anyone else has suggestions for topics we might want to cover? | 15:06 |
| ttx | o/ | 15:07 |
| *** evrardjp has joined #openstack-tc | 15:07 | |
| scas | o/ | 15:08 |
| * dhellmann thinks maybe we used up all our energy in the previous half hour | 15:09 | |
| zaneb | o/ | 15:10 |
| ttx | yes in a consensus-driven body, who actually casts the vote is not that much important | 15:10 |
| ttx | It's more of a safety valve thing | 15:10 |
| TheJulia | dhellmann: I don't know, it felt moderately energizing | 15:11 |
| dhellmann | TheJulia : that was more a comment on the lack of response to my question :-) | 15:11 |
| TheJulia | Suggestions wise, rewarding, raising visibility of peoples efforts.... I feel like we should at least discuss the board's perspective and maybe willingness to aid in such? | 15:13 |
| dhellmann | what would we be asking them to do? | 15:14 |
| cdent | I think it was pretty clear from what alan was saying that job 1 at the meeting is to convince the board members that we've done relevant stuff | 15:15 |
| dhellmann | I wonder if maybe talking about valuable areas of contributions would be more like what he was expecting | 15:16 |
| TheJulia | dhellmann: Well, I feel like it is pretty close to driving media perception as well | 15:16 |
| dhellmann | is this an area where we should be asking, or telling? | 15:16 |
| dhellmann | IOW, do we need their help, or do we need to just do it? | 15:17 |
| TheJulia | dhellmann: telling that we will do it maybe | 15:17 |
| dhellmann | I'd rather we're able to say we have done it, but yeah :-) | 15:17 |
| TheJulia | cdent: that is a good point | 15:18 |
| TheJulia | Are there any community wide graphs of velocity measurements cycle to cycle? | 15:19 |
| dhellmann | ttx has some tools to do that sort of thing | 15:20 |
| TheJulia | because everything we do is at a meta level, and can only really be measured via direct response and aggregate end result (and 99.95% of that is the long play sort of impacts) | 15:20 |
| *** annabelleB has quit IRC | 15:21 | |
| ttx | It's just hard to look at "community wide" since we are adding things and removing things all the time | 15:21 |
| *** jamesmcarthur has joined #openstack-tc | 15:21 | |
| mnaser | i think we can look at the technical point of view | 15:22 |
| ttx | Like in my "number of repos" vs. "number of commits" over cycles graph, I actually edited out the deb repos because they were slanting the data so much | 15:22 |
| mnaser | more like "this is what openstack allows you to do" | 15:22 |
| dhellmann | so, talk about new features? | 15:23 |
| ttx | If we start selecting exciting goals (which was hard before, but we seem to have more community buy-in those days), it will facilitate reusing them in such comms | 15:24 |
| ttx | exceiting/user-beneficial | 15:24 |
| dhellmann | yes, that's true, so we could talk about some of the goals being considered | 15:24 |
| mnaser | dhellmann: yes, but rather than new features, maybe focus on the use cases of those features and the value they provide | 15:24 |
| cdent | mnaser++ | 15:24 |
| mnaser | "nova has vgpu support" is cool, but "nova has vgpu support which means you can virtualize usage of your gpus just like other resources, saving your money, and making ai/ml use cases much more accessible without investing a ton of money" is way more exciting | 15:25 |
| mnaser | (imho) | 15:25 |
| dhellmann | that makes sense | 15:26 |
| dhellmann | we should start making a list of the things we think are worth mentioning at that level | 15:26 |
| TheJulia | I have lots of mixed feelings regarding raising up a single project, but I do agree... and that is the way much of that is going to be perceived as returning value. | 15:27 |
| TheJulia | dhellmann: ++ | 15:27 |
| dhellmann | we could start by going through the cycle highlights from rocky | 15:28 |
| dhellmann | and then pick a small number of things from several projects | 15:28 |
| TheJulia | But conveying the value of such is important because they are not going to be focused on the low level technical details. They are going to be reporting up their chains to indicate value that OpenStack is bringing them. | 15:28 |
| cdent | we could follow that list of "things done" with a list of "things nearly done, please help" | 15:28 |
| ttx | or "we want to do that, but nobody signed up yet" | 15:30 |
| smcginnis | I like that - here are the things we were able to do, here's where we could use some help. | 15:30 |
| smcginnis | Rather than just - we need help. | 15:31 |
| scas | ++ | 15:31 |
| dhellmann | that feels like falling back on the pattern that we've had before, and I understood Alan's message to say we need to dial that down | 15:31 |
| cdent | alan's message was "keep people excited" | 15:31 |
| lbragstad | i'd be curious if associating the "we need help in $area" to actual usecases will help drive attention to it? | 15:31 |
| cmurphy | i think he said dial it down but that it's also okay to present issues if they are framed in a way that sounds solveable | 15:32 |
| cdent | and what I'm trying to suggest is that from our side we need leverage people's excitment to draw in resources | 15:32 |
| TheJulia | alan's prior message was to try and influence our own hype cycle | 15:32 |
| TheJulia | or perception there of | 15:32 |
| TheJulia | I do think the "please help" is falling back on a not great pattern if we don't convey what the value of the end result would be | 15:32 |
| TheJulia | help them make a business case | 15:32 |
| smcginnis | Yeah, it's all about how it's framed I think. | 15:32 |
| TheJulia | exactly | 15:33 |
| ttx | it's also a quantity thing | 15:33 |
| cmurphy | s"we want to do X, Y, and Z so that it will be possible to A, B, and C" is better than "we need help everything is doomed" | 15:33 |
| zaneb | I missed most of that session but my understanding was that we need to (1) decide in concrete terms what we want from the board, and (2) communicate those in a way that is meaningful/persuasive to *them* rather than to us | 15:33 |
| cmurphy | it's hard to enumerate those specific asks though | 15:33 |
| ttx | I think saying "we did X, Y and Z because we listen to our users! Oh and we'd like to also do A but are still looking for volunteers there" is ok, quantity wise | 15:34 |
| cmurphy | ttx: ++ | 15:34 |
| lbragstad | e.g., "we need help with incorporating system scope and default roles across services so that you can securely offer more functionality to your users without incurring massive operator involvement" | 15:34 |
| TheJulia | ttx: ++ | 15:34 |
| ttx | It doesn't have to be all-singing. But currently it's all-OMG-we-are-sinking-help-us | 15:34 |
| ttx | It was natural for us to raise /issues/ in that forum, since that is the only one we have | 15:35 |
| ttx | But I understand Alan's point. The audience is more than just board members | 15:35 |
| ttx | and our words resonate more than we think | 15:36 |
| evrardjp | that's true | 15:36 |
| TheJulia | exactly, we need to be conscious of how things are perceived | 15:37 |
| dhellmann | If we assume that this meeting is not the only way for us to ask the board for help, how would that change what we want to talk about? | 15:37 |
| TheJulia | future planning | 15:37 |
| ttx | It could change dramatically -- more into a progress report than a quaterly list of grievances | 15:38 |
| TheJulia | where do we (collectively) want to see things | 15:38 |
| scas | i think pushing the narrative of "we need help" does hasten the decline of the hype. sure, everyone needs help, but i'm not expecting people to pick up what it is i do just because they know how to implement it. it'd help for continuity, but it's not an expectation i place on people | 15:38 |
| TheJulia | if we're focusing on grievances, then we're not being constructive | 15:38 |
| ttx | It's not as if pushing the "we need help" narrative was a super success. The help most wanted list was more used by developers than their managers | 15:38 |
| *** annabelleB has joined #openstack-tc | 15:39 | |
| dhellmann | TheJulia : I don't think we're going to have much time for strategic planning in these meetings with the growing number of people who are going to need to talk. So I think we're going to have to address planning separately, too. | 15:39 |
| ttx | The only thing it helped with is going past the "but you don;t tell us where help is needed" objection, but that one was probably spurious | 15:39 |
| TheJulia | dhellmann: then what does the purpose of the meeting really become for us as an established project | 15:40 |
| TheJulia | s/established project/established community and ecosystem" | 15:40 |
| dhellmann | that's exactly what I'm trying to work out | 15:40 |
| cdent | TheJulia: remind people that the lights are on, and the electricity bill needs to be paid | 15:40 |
| TheJulia | cdent: good point | 15:40 |
| ttx | I still think the help-most-needed list is useful, btw. It helps us advocating consistently for the same things | 15:40 |
| cdent | dog and pony | 15:40 |
| dhellmann | as ttx says, a report on what's going on seems like it might actually be the most valuable thing | 15:40 |
| * TheJulia feels like we're reaching consensus on that | 15:41 | |
| ttx | I really liked the reports from the K8sSIG for example detailing efforts there | 15:41 |
| ttx | hogepodge did a great progress report on K8s integration / cloud provider work | 15:41 |
| ttx | that was board material | 15:42 |
| ttx | we need to do a bit more of that, for OpenStack proper | 15:42 |
| evrardjp | ttx: you mind sharing said work? I haven't seen it, but I am not surprised of what you're saying. | 15:42 |
| ttx | I think dims and mrhillsman were involved as well | 15:43 |
| ttx | (giving credit where due) | 15:43 |
| ttx | Let me see, that was in Dublin I think | 15:43 |
| evrardjp | they are in the SIG indeed, but I mean I'd like to see content here, as this could serve as inspiration | 15:43 |
| dhellmann | I don't have links to those presentations handy, unfortunately | 15:44 |
| ttx | That would be https://docs.google.com/presentation/d/1_N7xhzwk6HzCl0fMm_cWfQ2UFu8bOaZYD2OOAV4y5yQ/edit?usp=sharing but it required permission apparently | 15:45 |
| ttx | https://wiki.openstack.org/wiki/Governance/Foundation/26Feb2018BoardMeeting | 15:45 |
| ttx | maybe hogepodge can give us a readable link | 15:45 |
| mrhillsman | try now | 15:46 |
| ttx | yay! | 15:46 |
| ttx | evrardjp: here you go ^ | 15:46 |
| mrhillsman | i agree the flow of it i think speaks to what i believe is being discussed here | 15:47 |
| *** annabelleB has quit IRC | 15:47 | |
| mrhillsman | what we accomplished, what we're working on, what we think the future looks like, call to action | 15:48 |
| dhellmann | so the challenge will be identifying the items that are worth putting on that list when we look at the entire community | 15:50 |
| cdent | quite so | 15:51 |
| *** e0ne has quit IRC | 15:51 | |
| dhellmann | and without having it turn into a mash-up of all of the various project update presentations | 15:52 |
| ttx | indeed | 15:55 |
| evrardjp | focus on community goals achievements and progresses first? | 15:55 |
| hogepodge | hi | 15:55 |
| hogepodge | I'll create an etherpad of all the presentations I've given on OpenStack and K8s integrations, and make sure they are world readable | 15:56 |
| cdent | evrardjp: a problem with that is that often community goals have limited end user value and part of the "ask" here is to make it clear that openstack is making more awesome for people, not for itself | 15:56 |
| dhellmann | thanks hogepodge | 15:57 |
| evrardjp | cdent: well I hope that what we are working as community goals have also a value on users, maybe some may not be directly linked, but they are ... | 15:57 |
| evrardjp | else I am not sure why we are doing them. I am getting paid for delivering code to users, not to hypothetics :D | 15:58 |
| cdent | evrardjp: sure, yes, but I'm very conscious of what I heard from alan: keep board members excited | 15:58 |
| mnaser | im in and out but | 15:58 |
| evrardjp | but I digress | 15:58 |
| cdent | that, to me, is code for: make them believe this is worth their investment | 15:58 |
| mnaser | i agre with TheJulia that it shouldn't be focused about a specific project, but we can pick highlights from different projects and rotate projects | 15:59 |
| evrardjp | mnaser: rotation is not a bad idea. | 15:59 |
| ttx | even if the end result is a message of "business as usual" continuity | 15:59 |
| mnaser | and i think ttx made a really really good point about how we deal with all of this, we should be more conscious in our messaging | 15:59 |
| ttx | taht's not the worst image to have in our industry | 15:59 |
| evrardjp | mnaser: agreed | 16:00 |
| mnaser | ttx: you can even put that as excellent... alan said himself that he thinks thats a good message on its own | 16:00 |
| mnaser | we're stable. we're shipping stuff. people are consuming it. | 16:00 |
| ttx | yet another boring release | 16:00 |
| mnaser | and we're still doing it at super web-browser zuul status page breaking scale | 16:00 |
| mnaser | :P | 16:00 |
| ttx | At the same time we are doing a LOT of things, so we should communicate the value of that | 16:01 |
| evrardjp | well python3 is for me both not exciting for users, and exciting for the future -- I am pretty sure we can turn that out as a good "user" story | 16:01 |
| ttx | It's still tends of thousands of commits. So something is happening :) | 16:01 |
| mnaser | yeah. it's all about messaging | 16:01 |
| ttx | tens* | 16:01 |
| evrardjp | (just giving an example, of what is maybe not super fancy, but can become very interesting) | 16:01 |
| mrhillsman | staff provides a presentation and i think it makes sense to take for example the user stories they highlight/discuss and have project highlights that speak to those | 16:02 |
| mnaser | i dont know how late we are to the party for *major* projects | 16:02 |
| mnaser | but we could be one of the first major(i think?) to adopt python3 | 16:02 |
| mnaser | and not just say we support it, really go as far as have it fully tested in CI | 16:02 |
| ttx | mrhillsman: we need to avoid the "talk only about new projects" hype trap though | 16:02 |
| mrhillsman | not saying new projects | 16:02 |
| zaneb | ttx: there were a lot of typos ;) | 16:03 |
| mnaser | honestly | 16:03 |
| mnaser | i don't even think that's an issue | 16:03 |
| mnaser | a lot of those commits are adding consistency to our projects, documentation ,typo fixes, etc | 16:04 |
| mrhillsman | i also don't think it is hype if married with a user story the staff discusses; since they talk first | 16:04 |
| dhellmann | mnaser : oh, we're going to be one of the last by far | 16:04 |
| ttx | zaneb: and a lot of infra changes. But that's still talking to positive aspects, like quality | 16:04 |
| mnaser | dhellmann: well i guess we can say its just really big and it's a big effort :p | 16:05 |
| dhellmann | mnaser : yes, the success story is that our multi-year effort is finally paying off | 16:05 |
| ttx | sounds more productive than discussing how to avoid gratuitous/gaming typo fixes with board members | 16:05 |
| mnaser | i think also it's important that we realize that our words have strong value behind them. in #openstack-tc we can talk about problems and we all know "everything is not on fire" | 16:05 |
| mnaser | but when we talk on board how there are issues, that's all they see | 16:05 |
| dhellmann | yeah, we won't be talking about stats gaming with the board as long as I'm chair | 16:06 |
| mrhillsman | ++ | 16:06 |
| evrardjp | I have to go-- very interesting conversation, but holidays can't wait anymore :p See you laterz everyone! | 16:06 |
| ttx | Definitely Forum material :) | 16:06 |
| ttx | evrardjp: you're going for holidays the day of your TC nomination ? Didn't you know we don;t take vacation ? | 16:07 |
| scas | vacation, what language is that? :shifty-eyes: | 16:07 |
| evrardjp | ttx: better be fresh should I be elected? :p | 16:07 |
| evrardjp | scas: hahah same feeling. | 16:08 |
| ttx | scas: i think the French invented that concept | 16:08 |
| evrardjp | ttx: not surprised there :D | 16:08 |
| ttx | or if not it's like food, we perfected it | 16:08 |
| scas | ttx: ah, i think i picked up some vacationaise a while back, but it got lost in a relocation | 16:10 |
| *** jpich has quit IRC | 16:12 | |
| *** diablo_rojo has joined #openstack-tc | 16:17 | |
| cdent | can someone join me on https://demo.codimd.org/fqBziNc_Tc2k-fB0AYWatA and lemme know if it seems to be working as a pretty etherpad? is from https://github.com/hackmdio/codimd | 16:19 |
| cdent | agplv3 which might be problematic for some, but... | 16:19 |
| cdent | thanks whoever that is | 16:25 |
| *** cdent has quit IRC | 16:26 | |
| dhellmann | looks swanky | 16:26 |
| dhellmann | does it track author colors? | 16:27 |
| ttx | yeah was about to ask | 16:27 |
| dhellmann | or I guess does it track authorship somehow | 16:27 |
| ttx | might be better as a Wiki replacement than an etherpad replacement | 16:28 |
| ttx | also worth noting they are on Matrix, using Riot | 16:29 |
| dhellmann | the thing I like about the wiki is the notifications when pages change | 16:30 |
| dhellmann | and the authentication for editing them, although that seems weak if we're still dealing with spam | 16:31 |
| ttx | this supports the authentication part | 16:31 |
| dhellmann | ok, I didn't notice that | 16:31 |
| ttx | https://demo.codimd.org/features#Permissions | 16:31 |
| ttx | Their site is actually built using it, so that gives an idea of teh wiki-style usage | 16:32 |
| ttx | I like the UML diagrams helpers | 16:32 |
| ttx | I mean, I always wanted to include music sheets in my wiki pages. | 16:33 |
| ttx | dhellmann: in the menu at the top right there is a "history" that shows changes, but that's as far as I can see it doing attribution of changes | 16:35 |
| ttx | Promising i'd say, but a little light in tracking changes (for the wiki use case) or local attribution (for the etherpad use case) | 16:38 |
| hogepodge | ttx: dhellmann: https://etherpad.openstack.org/p/k8s-openstack-updates | 16:50 |
| *** e0ne has joined #openstack-tc | 16:52 | |
| *** cdent has joined #openstack-tc | 16:53 | |
| *** dtantsur is now known as dtantsur|afk | 16:54 | |
| *** e0ne has quit IRC | 16:54 | |
| *** diablo_rojo has quit IRC | 17:05 | |
| *** jamesmcarthur has quit IRC | 17:06 | |
| *** e0ne has joined #openstack-tc | 17:16 | |
| *** cdent has quit IRC | 17:16 | |
| *** annabelleB has joined #openstack-tc | 17:51 | |
| *** diablo_rojo has joined #openstack-tc | 17:56 | |
| *** dims_ is now known as dims | 18:05 | |
| fungi | wow, that was a rather massive amount of scrollback. remind me never to schedule long drives on a thursday | 18:15 |
| mriedem | anyone have the top release deployed from the user survey off the top of their head? | 19:00 |
| mriedem | ah https://www.openstack.org/analytics | 19:02 |
| mriedem | tied at mitaka/newton | 19:02 |
| mriedem | re this osc gaps goal proposal, | 19:02 |
| mriedem | if nova made it a goal to have at least ~mitaka level microversion parity, that would mean supporting up to at least microversion 2.25 in osc which was the max compute API microversion in mitaka | 19:03 |
| mriedem | that would help break that goal down into a more digestable chunk | 19:03 |
| mriedem | that likely wouldn't work as well for other projects though... | 19:04 |
| mriedem | given the gaps in osc between 2.1 and 2.25 https://etherpad.openstack.org/p/compute-api-microversion-gap-in-osc that wouldn't be insignificant | 19:04 |
| * TheJulia sees mention of vacation and wonders what sort of crazy talk that is :) | 19:26 | |
| * jroll considers turning all you not-vacation-people's internet off for a week | 19:27 | |
| TheJulia | mnaser: speaking of messaging highlights, I think it would also be good to point back to the foundation press release as well, since those are messaging highlights as well, and us echoing just gives them more power | 19:28 |
| zaneb | mriedem: note that that data is from last year | 19:28 |
| TheJulia | jroll: I was actually looking at airline tickets yesterday (and need to again today) to try and figure out the best way to take a vacation in europe | 19:28 |
| jroll | ++++++ | 19:28 |
| TheJulia | jroll: since flights are actually WAY more less expensive if I spend a couple more days in Germany | 19:29 |
| * TheJulia realizes that doesn't read properly and thinks coffee might be required | 19:30 | |
| zaneb | TheJulia: yeah, Friday flights were crazy expensive! | 19:30 |
| jroll | TheJulia: yeah, for me monday-monday is $800, monday-friday is $3k | 19:30 |
| jroll | sunday-saturday also 3k | 19:30 |
| TheJulia | zaneb: Thursday, Friday, and Saturday for me.. | 19:30 |
| zaneb | TheJulia: my plan was to hang around until Sunday. maybe we should organise a TC tourism outing :) | 19:31 |
| cmurphy | vacation in europe is A+ would recommend | 19:31 |
| TheJulia | then again, returning on a weekend is tough because I end up hitting routes that have people going home from vacations | 19:31 |
| TheJulia | zaneb: I'm also looking at flying my soon to be mother in law over for a few days. (omg, so few days till the wedding...) | 19:31 |
| mnaser | i already know i'll be so out of it by the 16th | 19:37 |
| *** jamesmcarthur has joined #openstack-tc | 19:37 | |
| * TheJulia has a desire to ride trains suddenly | 19:38 | |
| *** e0ne has quit IRC | 19:42 | |
| *** e0ne has joined #openstack-tc | 19:47 | |
| *** e0ne has quit IRC | 19:50 | |
| *** e0ne has joined #openstack-tc | 20:08 | |
| *** diablo_rojo has quit IRC | 20:08 | |
| *** diablo_rojo has joined #openstack-tc | 20:11 | |
| *** annabelleB has quit IRC | 20:22 | |
| *** diablo_rojo has quit IRC | 20:26 | |
| *** jamesmcarthur has quit IRC | 20:47 | |
| *** e0ne has quit IRC | 21:14 | |
| fungi | yeah, i noticed fiddling start/end dates of my itinerary influenced airfare by up to a factor of 5x | 21:15 |
| fungi | for the cheapest coach/economy fares | 21:15 |
| dhellmann | yeah, I'll be arriving on the 10th because of that | 21:17 |
| dhellmann | so I guess a sunday of site-seeing, as usual | 21:17 |
| *** jamesmcarthur has joined #openstack-tc | 21:17 | |
| *** jamesmcarthur has quit IRC | 21:35 | |
| *** jamesmcarthur has joined #openstack-tc | 21:37 | |
| *** jamesmcarthur has quit IRC | 21:39 | |
| *** jamesmcarthur has joined #openstack-tc | 21:42 | |
| *** mriedem has quit IRC | 21:55 | |
| *** diablo_rojo has joined #openstack-tc | 21:57 | |
| *** annabelleB has joined #openstack-tc | 21:59 | |
| *** mriedem has joined #openstack-tc | 22:07 | |
| scas | there's a small chance i may be at the berlin summit, but flights for the size-challenged are crazy | 22:15 |
| scas | there's a zero chance i could make it work in coach | 22:16 |
| scas | cheapest i've found so far is in the thousands of dollars | 22:22 |
| diablo_rojo | A little over an hour left in the election! 403 of the 1636 have voted. Haven't had this high percentage of voter turnout since like..Liberty. | 22:29 |
| dhellmann | nice | 22:32 |
| diablo_rojo | Yeah pretty cool. | 22:33 |
| scas | neat | 22:34 |
| *** jamesmcarthur has quit IRC | 22:37 | |
| fungi | awesome! | 22:38 |
| *** jamesmcarthur has joined #openstack-tc | 22:53 | |
| *** tosky has quit IRC | 22:53 | |
| *** jamesmcarthur has quit IRC | 22:57 | |
| *** jamesmcarthur has joined #openstack-tc | 23:07 | |
| *** jamesmcarthur has quit IRC | 23:12 | |
| mriedem | jesus with those kinds of numbers we could have kept trump out of office! | 23:19 |
| *** mriedem has quit IRC | 23:26 | |
| persia | USA Presidential election November 2016 turnout: 59.2% of eligible voters for the office of the president registered a vote for the office of the president. OpenStack TC election September 2018 turnout: >24.6% of eligible voters registered a vote (won't have final numbers for a bit: USA preliminary numbers at a similar point were 53.5%) | 23:26 |
| persia | (but ballot centre lines tend to lead to more late votes in that class of election, due to how the election is managed) | 23:27 |
| mnaser | well | 23:49 |
| mnaser | dont keep us on our toes :) | 23:49 |
| *** jamesmcarthur has joined #openstack-tc | 23:52 | |
| persia | Turns out there were no final votes. 403 is the final count. | 23:52 |
| * mnaser forgot they can check results from the link | 23:53 | |
| mnaser | welcome back dhellmann, TheJulia and fungi. welcome evrardjp, lbragstad and gmann | 23:54 |
| smcginnis | Great group of people! | 23:54 |
| mnaser | so close but please, i'd love to see ricolin, zhipeng, knikolla, alameadows and scas come in for the next round. i appreciate how involved everyone has been recently | 23:56 |
| scas | we'll see. the winds do alter course unexpectedly | 23:56 |
| fungi | yes, all of you please do run again in... what is it this time, 7 months? | 23:57 |
| TheJulia | ++ | 23:57 |
| persia | Also, remember that these folk aren't TC members yet. | 23:57 |
| * fungi can go back to helping as an election official now at least ;) | 23:57 | |
| mnaser | persia: oh? sorry for jumping the gun then but why so? | 23:57 |
| persia | Changeover happens at Summit, or somewhere around there (I forget the details) | 23:57 |
| mnaser | oh | 23:57 |
| mnaser | TIL | 23:57 |
| mnaser | i thought they had to go through bootcamp and i escaped that, hah | 23:58 |
| fungi | date election concludes != effective date for change of office | 23:58 |
| persia | Basically, we have the election, and then we know who will be TC at changeover, so that folk have time to reorganise travel plans, etc. before they are expected to fulfill their duties. | 23:58 |
| TheJulia | mnaser: Oh! we somehow missed that. Well let me to ahead and get on the phone with the booking folks :) | 23:58 |
| persia | But this also means that folk who chose not to stand for reelection aren't magically off the hook just because the election concluded. | 23:58 |
| * TheJulia is joking | 23:58 | |
| knikolla | Congrats to all the elected :) | 23:59 |
| mnaser | cool, interesting | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!