| *** ricolin_ has joined #openstack-tc | 01:10 | |
| *** edmondsw has joined #openstack-tc | 01:13 | |
| *** edmondsw has quit IRC | 01:18 | |
| *** jaypipes has quit IRC | 02:06 | |
| *** jaypipes has joined #openstack-tc | 02:15 | |
| *** edmondsw has joined #openstack-tc | 03:02 | |
| *** edmondsw has quit IRC | 03:06 | |
| *** edmondsw has joined #openstack-tc | 04:51 | |
| *** edmondsw has quit IRC | 04:55 | |
| *** e0ne has joined #openstack-tc | 05:18 | |
| *** e0ne has quit IRC | 05:21 | |
| *** ianychoi has joined #openstack-tc | 05:34 | |
| *** jpich has joined #openstack-tc | 07:34 | |
| *** e0ne has joined #openstack-tc | 07:46 | |
| *** edmondsw has joined #openstack-tc | 08:28 | |
| *** edmondsw has quit IRC | 08:32 | |
| *** e0ne has quit IRC | 09:21 | |
| *** e0ne has joined #openstack-tc | 09:23 | |
| *** dtantsur|afk is now known as dtantsur | 09:33 | |
| *** spsurya_ has quit IRC | 09:49 | |
| *** ricolin_ has quit IRC | 09:57 | |
| *** ricolin has joined #openstack-tc | 09:57 | |
| *** ricolin_ has joined #openstack-tc | 09:58 | |
| *** ricolin_ has quit IRC | 09:58 | |
| *** e0ne has quit IRC | 10:28 | |
| *** gcb has joined #openstack-tc | 10:58 | |
| *** gcb has quit IRC | 11:13 | |
| *** ricolin has quit IRC | 11:14 | |
| * jroll sees lots of conversation outside of office hours and applauds | 11:19 | |
| *** edmondsw has joined #openstack-tc | 11:29 | |
| *** e0ne has joined #openstack-tc | 12:07 | |
| *** edmondsw has quit IRC | 12:07 | |
| *** rosmaita has joined #openstack-tc | 12:12 | |
| *** edmondsw has joined #openstack-tc | 12:13 | |
| *** edmondsw_ has joined #openstack-tc | 12:16 | |
| *** edmondsw has quit IRC | 12:19 | |
| *** rosmaita has quit IRC | 12:24 | |
| *** rosmaita has joined #openstack-tc | 12:30 | |
| *** mriedem has joined #openstack-tc | 12:31 | |
| *** e0ne has quit IRC | 12:48 | |
| *** cdent has joined #openstack-tc | 13:15 | |
| *** ricolin has joined #openstack-tc | 13:22 | |
| *** zhipeng has joined #openstack-tc | 13:48 | |
| *** e0ne has joined #openstack-tc | 13:59 | |
| dims | o/ | 14:33 |
|---|---|---|
| * cdent passes dims a cold beverage | 14:33 | |
| dims | thanks cdent | 14:34 |
| dims | cdent : i was thinking about kevin's email about "rewrite all services using CRD's to get rid of databases" ... what does it buy us over and beyond running openstack services in containers like folks already do. the biggest con is re-writing everything we already have in go (and sqeezing rpc-style operations into CRDs). i just can't wrap my head around it | 14:40 |
| cdent | dims: I'm not certain, but think he's saying that is a way to be more tightly integrated with kubernetes (and some of it's capability to manage itself) but he's not been very clear about that | 14:43 |
| cdent | but the gist of much of what he says does appear to be: yes, really, bite the bullett, rewrite everything | 14:43 |
| cdent | which I think even he acknowledges is unlikely | 14:44 |
| smcginnis | I was a little confused on that too. It seemed like he was suggesting OpenStack should go all in on just being a part of k8s? | 14:48 |
| cdent | yes, that is one of the things he said | 14:49 |
| smcginnis | Seems a very k8s centric view, but I guess that's what he's advocating. | 14:50 |
| cdent | he's saying: if the end game is cloudy compute resources then ____ | 14:51 |
| * mnaser will be in and out of todays office hours | 14:51 | |
| cdent | where I think the discussion falls down is if that is in fact the end game | 14:51 |
| * mnaser is traveling to the bay area and happened to run into this https://www.meetup.com/openstack/events/251802241/ | 14:51 | |
| mnaser | given ill be here, ill try to attend and see what its like | 14:51 |
| dhellmann | what is a CRD? | 14:52 |
| smcginnis | k8s object basically | 14:53 |
| smcginnis | https://kubernetes.io/docs/concepts/extend-kubernetes/api-extension/custom-resources/ | 14:53 |
| dhellmann | so we wouldn't get rid of the database, we'd just use whatever kubernetes is using to store those | 14:53 |
| cdent | mnaser: do a trip report please :) | 14:54 |
| smcginnis | dhellmann: I guess so? | 14:54 |
| smcginnis | It is taking both the stance that OpenStack is only for cloudy compute resources and that k8s is the ultimate final goal for running those cloudy things. | 14:55 |
| dhellmann | that seems like an extremely disruptive change | 14:55 |
| cdent | I think we need to read these email messages not as descriptions of things that should be done, but as things from which we can cherry pick useful bits | 14:55 |
| smcginnis | There are certainly good points in there, even if I don't agree with a lot of the main premises. | 14:56 |
| mnaser | cdent: i kinda want to do that because it's a lot of cross communities | 14:56 |
| mnaser | so i'm a bit curious on what things look like | 14:56 |
| jroll | tbf, I think kevin does realize this won't happen, he's just saying what he personally would do right now with unlimited time and resources | 14:56 |
| smcginnis | The tricky part is always extracting the useful bits that can be done with very limited time and resources. | 14:57 |
| jroll | the even trickier part is agreeing on which bits are useful :) | 14:57 |
| cdent | jroll: quite | 14:57 |
| smcginnis | Heh, probably the trickiest part. | 14:57 |
| cdent | but without some concrete stuff to look at, we can't even pick and choose, so I'm grateful to kevin for laying out his fantasy | 14:58 |
| jroll | dhellmann: as an aside, thanks for the love on ironic's release notes :) | 14:58 |
| *** AlanClark has joined #openstack-tc | 14:58 | |
| dhellmann | jroll : I like the use of the prelude section in particular. Just what I had in mind. :-) | 14:58 |
| jroll | \o/ | 14:58 |
| *** zhipeng has quit IRC | 14:59 | |
| cdent | hey tc-members, it's office hours time | 15:00 |
| cdent | I believe we agreed to not startmeeting? | 15:00 |
| smcginnis | I think that's still an open question. | 15:00 |
| dims | +1 to not do start-meeting | 15:01 |
| smcginnis | ++ | 15:01 |
| fungi | i'm good skipping the meetbot | 15:01 |
| * mnaser is in and out | 15:03 | |
| pabelanger | o/ | 15:03 |
| mnaser | Wild idea: who’d be open to having at least 1 meeting between the TC once a week | 15:04 |
| mnaser | I enjoy that we discuss things outside a period of time but I feel like as someone with many other responsibilities I follow up with the channel and it’s a bit too late to chime in on a discussion | 15:05 |
| ttx | o/ | 15:05 |
| mnaser | And just to keep up in general action items and progress on where were at in some of the things we’re doing (individually or in pairs) | 15:05 |
| cdent | mnaser: what's the difference between a meeting and an office-hour? | 15:05 |
| cmurphy | o/ | 15:05 |
| smcginnis | mnaser: This timeslot does kind of end up being that "meeting" due to usual turn out, but I'd hesitate bringing that back as an official thing. | 15:05 |
| cdent | mnaser: is it a sense of obligation? | 15:05 |
| mnaser | I think office hours are meant to be for those who want to reach the tc (imho) | 15:06 |
| mnaser | And meetings is the TC getting together (with anyone from the community welcome) | 15:07 |
| smcginnis | To be fair, I think we were supposed to move those meeting discussions between the TC to the mailing list when we stopped having the meeting. | 15:07 |
| ttx | What do people think of my suggestion there ? | 15:07 |
| smcginnis | I don't think we've really been good at that. | 15:08 |
| *** hongbin has joined #openstack-tc | 15:08 | |
| mnaser | Given we don’t really have much people other than the TC show up, might as well announce that “anyone is welcome anytime in #openstack-tc” and we can find slots where we can sit and figure out and catch up what we’re doing/working | 15:08 |
| smcginnis | ttx: Sorry, which suggestion? | 15:08 |
| ttx | http://lists.openstack.org/pipermail/openstack-dev/2018-July/132003.html | 15:08 |
| cmurphy | ttx: I'm a fan | 15:08 |
| ttx | (add a feature to meetbot that it would allow taking notes without a meeting going on) | 15:08 |
| ttx | that way we get the benefits without the drawbacks | 15:08 |
| dhellmann | so that would just be time-boxed to a UTC day instead of ~1 hr? | 15:09 |
| dhellmann | I thought the benefit of meetbot was having a link to the beginning of the discussion | 15:09 |
| ttx | no | 15:09 |
| smcginnis | I think between dhellmann's and cdent's summaries, we kind of already do get the documentation. | 15:09 |
| dhellmann | I can just get that link by looking at eavesdrop logs | 15:09 |
| ttx | err yes | 15:10 |
| dhellmann | it's a little more work for me, but I can probably script something to look for "it's time for office hours" or some similar phrase to find the links | 15:10 |
| ttx | dhellmann: it would let us do #info or #agreed in the middle of the channel and have a page that lists them | 15:10 |
| ttx | I see no benefit in finding "the start" | 15:10 |
| dhellmann | that might be nice. we don't really use those in hours, now, though. | 15:11 |
| cdent | ttx one way to make that happen is to start and end a meeting at midnight each day. as in: that's the quickest path to what you describe | 15:11 |
| dhellmann | ttx: links to the start of the discussions was specifically the thing we were asked for | 15:11 |
| smcginnis | Can we still use those tags and just have them as a search target when looking through normal channel logs? | 15:11 |
| ttx | cdent: yes -- figured iyt would not be hard to automate it in the bot itself though | 15:11 |
| dhellmann | smcginnis : sure | 15:11 |
| fungi | there was a similar feature request to have something like statusbot's thanks and success features but more as a per-project note taking mechanism | 15:11 |
| dhellmann | if someone wants to work on the bot, that's fine with me, I'm just suggesting we may not need it | 15:11 |
| smcginnis | A notebot does sound like it could come in handy, but I agree I'm not sure we really need it. | 15:12 |
| ttx | I feel like it could provide a nice summary of "what is being discussed" with links to further logs | 15:12 |
| fungi | i'm cool with the lighter-weight log parser idea | 15:12 |
| dhellmann | I mean technically it should be possible to generate the links to the hour start times mechanically, right? if we're all punctual? ;-) | 15:12 |
| ttx | I think we are not talking about the same thing | 15:13 |
| jroll | it seems like dhellmann's summaries capture things we would use #info and #agreed for | 15:13 |
| smcginnis | jroll: ++ | 15:13 |
| dhellmann | if we actually used #info and #agreed, I would use those for my summaries | 15:13 |
| dhellmann | we just haven't been using those | 15:13 |
| jroll | heh, yeah my next question is if they'd be useful for that | 15:13 |
| zaneb | o/ | 15:13 |
| dhellmann | the thing rockyg and others asked for was to get links to the logs from when office hours start | 15:13 |
| dhellmann | and that's why I thought we were using meetbot | 15:13 |
| dhellmann | since these things aren't meetings, it felt right to not use #info and #agreed and #vote as part of them | 15:14 |
| dhellmann | although I would have liked #info :-) | 15:14 |
| ttx | If that is the goal, I'd argue that's an antifeature | 15:14 |
| smcginnis | My opinion, but I think having the summaries are good for high level and channel logs are good if someone actually cares to dig into little details. | 15:15 |
| dhellmann | it's not really clear to me that anyone cares about those links, so I'm content to just include the general day's log link instead | 15:15 |
| smcginnis | So honestly, talking about spending time on a bot or changing process seems like a waste of limited resources to me. | 15:15 |
| cdent | smcginnis++ | 15:15 |
| jroll | ++ | 15:15 |
| dhellmann | the objections I heard to using meetbot was that its features made it easy for the general discussions to turn into meetings and that's what we wanted to avoid | 15:15 |
| cdent | it's not hard to find the logs | 15:15 |
| ttx | sure. I was just suggesting we might want something between raw logs and terse summaries | 15:16 |
| fungi | more like use #note or #highlight | 15:16 |
| ttx | something between 500 lines and 3 lines | 15:16 |
| jroll | if we switched to slack we could star messages | 15:16 |
| * jroll ducks | 15:16 | |
| ttx | (real example_ | 15:16 |
| ttx | ) | 15:17 |
| fungi | to annotate logs so they can be more easily parsed for potentially interesting content | 15:17 |
| dims | LOL jroll | 15:17 |
| fungi | or maybe #summary with a summarization of some discussion? | 15:17 |
| cmurphy | it seems like we're unclear on what the original ask was, so discussing the solution doesn't feel as productive as it should be | 15:17 |
| dhellmann | I have been focusing on the other tracking artfiacts for most of my summary content, and maybe that's what ttx is referring to in the relative line counts | 15:18 |
| fungi | it's been brought to my attention that the public cloud working group meeting underway right now is discussing a new effort strikingly similar to poppy | 15:18 |
| ttx | cmurphy: also condensing two separate discussions into one | 15:18 |
| ttx | 1/ making the office hour more like a meeting considered harmful | 15:18 |
| fungi | for integrating cloudflare global load-balancing with openstack | 15:18 |
| ttx | 2/ potential other tools we could use | 15:18 |
| dhellmann | maybe I should be asking for volunteers to summarize office hour discussions separately from the weekly status updates | 15:19 |
| ttx | Basically I'm saying that we should summarize the channel activity/discussions, not just the office hours | 15:20 |
| dhellmann | I had been treating them as not-a-meeting so I didn't emphasize them in the summaries | 15:20 |
| dhellmann | yeah, I was hoping discussions would lead to other artifacts like mailing list threads or tracker page updates | 15:20 |
| cdent | I tend to summarize anything that is actually different/unusual in my weekly reports, and when I do I read the entire week. And there really isn't that much that we say. | 15:21 |
| ttx | like I said in my email, office hours were meant to encourage gathering around specific times (1) to increase the odds of reaching critical mass necessary for discussion, and (2) to ensure presence for outsiders wanting to reach out to the TC. | 15:21 |
| dhellmann | (1) seems to be working in some slots | 15:21 |
| fungi | yeah, i think that having designated office hours where larger number of us try to be around is good, avoiding making decisions in irc and pushing those sorts of tc-centric discussions to mailing lists is good, but also annotating what discussion goes on in channel to make it easier to summarize later also good | 15:21 |
| ttx | And I feel like using meetbot is making (2) more difficult | 15:21 |
| dhellmann | (2) is unclear | 15:21 |
| ttx | anyway, i developed my argument in that email, I won't paste all of it here :) | 15:22 |
| dhellmann | ttx, if you offered developer time for meetbot, I think I would ask you to spend it on storyboard instead | 15:22 |
| fungi | and i too think annotation of irc discussion shouldn't be limited to what gets discussed during office hours as we have some good conversations in here outside office hours as well | 15:23 |
| jroll | fungi: thanks for mentioning the global load balancer thing, that's quite interesting | 15:24 |
| fungi | if the goal is to better collect, highlight and summarize things which are talked about in here, log parsing would work just as well as an active bot integration | 15:24 |
| dhellmann | is that the goal? | 15:24 |
| zaneb | ttx: seems like all we need to implement you idea is a bot that types '#endmeeting\n#startmeeting' once a day? | 15:24 |
| ttx | dhellmann: I don't know. If the goal of using meetbot was just to timestamp logs... maybe not | 15:24 |
| dhellmann | *my* goal is for all of us to be talking about things, but then actually *doing* things that would show up in the summary | 15:25 |
| dhellmann | that's why I have been trying to emphasize linking to patches or mailing list threads | 15:26 |
| ttx | zaneb: there are alternate solutions ! Let's write a spec. :P | 15:26 |
| ttx | dhellmann: amen to that | 15:26 |
| ttx | To talk about something more productive... I did some namedropping on the Adjutant review to encourage the few members that haven't commented yet to say something, if only that they are neutral | 15:28 |
| fungi | jroll: i also passed along a link to https://review.openstack.org/273756 (poppy's application) for historical reference | 15:28 |
| jroll | ++ | 15:28 |
| dhellmann | fungi : thanks | 15:28 |
| ttx | On the diversitytagectomy, some appear confused with the goals of the health tracking | 15:29 |
| dhellmann | your description was good, and I tried to reply to that email thread, too | 15:29 |
| fungi | what confusion was there besides the one reply to the ml thread offering to be a liaison to the tc? | 15:29 |
| ttx | fungi: see on the revoew | 15:29 |
| fungi | ahh, thanks | 15:30 |
| dhellmann | that was the same person, wasn't it? | 15:30 |
| cdent | yes | 15:30 |
| ttx | yes indeed | 15:30 |
| cdent | I think I'm partly to blame for that | 15:30 |
| ttx | I blame cdent | 15:30 |
| ttx | and flaper87 | 15:30 |
| cdent | 3rd paragraph on https://anticdent.org/tc-report-18-27.html | 15:30 |
| cdent | and I actually think what I said is true, despite our protestations | 15:30 |
| dhellmann | https://review.openstack.org/#/c/579870/ is the review | 15:31 |
| ttx | cdent: ah, "aspect" | 15:31 |
| ttx | it definitely is an aspect | 15:31 |
| cdent | we are intending to take a more human oriented approach to evaluating if projects are in good shape and doing well | 15:31 |
| ttx | cdent: except the tags were not really consumed by the TC itself | 15:31 |
| fungi | hongbin does seem to be concerned that the tc is passing judgement rather than trying to help | 15:31 |
| fungi | i can see how it would come across that way too | 15:31 |
| hongbin | o/ | 15:32 |
| cdent | ttx: that's just not even remote true and you know it. We react to the data processing that build tags all the time | 15:32 |
| cdent | it's a thing we talk about all the time | 15:32 |
| fungi | hongbin: in reference to your points on the diversity tag removal change and your reply to the tc summary ml thread | 15:32 |
| fungi | (for context) | 15:32 |
| dhellmann | hongbin : did the replies from ttx on the review and from me on the mailing list help address your concerns? | 15:33 |
| hongbin | dhellmann: not quite | 15:33 |
| dhellmann | would you like to discuss it more here (I don't know what time it is where you are) or on the mailing list? | 15:33 |
| hongbin | because somehow, the report published to the wiki has some impacts on the project visibility | 15:33 |
| ttx | cdent: I'm pretty sure nobody can tell me here without looking if karbor has diverse-affiliation. I'd agree that when a project ends up single-vendor. it raises a fragility flag. | 15:34 |
| cdent | I was speaking of tags in general | 15:34 |
| dhellmann | hongbin: the notes in the wiki are intended to help the tc identify issues. Are you worried that others will read that wiki page and come to conclusions based on it? Or that other people will not find the wiki page at all? | 15:35 |
| hongbin | dhellmann: I can continue the discussion in here or ML, either way is fine with me | 15:35 |
| dims | hongbin : you mean folks using the wiki health report for internal company decisions? | 15:35 |
| ttx | hongbin: would it be better if notes were documented in a more anonymous etherpad ? We are trying to balance collaborating in public and not making it "official" | 15:35 |
| hongbin | dhellmann: yes, sort of, everyone can read the wiki and judge if the project is mature or not | 15:35 |
| dhellmann | hongbin : is that bad? | 15:35 |
| hongbin | ttx: using etherpad might be better because it is less formal | 15:36 |
| ttx | I wish nobody would consider teh wiki as formal, but i get your point | 15:36 |
| cdent | etherpads are _really_ hard to manage for anything other than real time notetaking | 15:37 |
| dhellmann | I can work on adding some text to the wiki page to explain that these are notes and not conclusions | 15:37 |
| fungi | also etherpads are fragile and hard to recover when they become corrupted (which is increasingly likely the longer a single pad gets reused) | 15:37 |
| dhellmann | but I need us to use a tool that notifies me when we make changes because otherwise I have to mentally diff an etherpad every week to produce the status report | 15:38 |
| dims | ++ dhellmann | 15:38 |
| ttx | dhellmann: maybe we should remove the traffic lights as they make jumping to conclusions easier | 15:38 |
| dhellmann | ttx: yes, that may be a good idea, too | 15:38 |
| fungi | that was my primary initial concern with the traffic lights (or any badge/icon really) | 15:38 |
| dhellmann | hongbin : what do you think of those suggestions? | 15:38 |
| ttx | I liked that it helped me focus my attention on projects that need more help, though | 15:38 |
| cmurphy | ++ to removing the traffic lights | 15:38 |
| * hongbin is reading the logs | 15:38 | |
| hongbin | sorry, slow english reader | 15:39 |
| dims | no rush :) | 15:39 |
| ttx | so if we remove traffic lights, it would be good to get some amount of reporting back, for example on this channel | 15:39 |
| ttx | I for one would like to know if a project team needs urgent help (the red light) | 15:40 |
| hongbin | dhellmann: so the suggestion is to remove the traffic light from the wiki, anything else? | 15:40 |
| ttx | hongbin: and add some warning text that these are notes | 15:40 |
| hongbin | ttx: ack | 15:40 |
| dhellmann | I added a little text to the top of https://wiki.openstack.org/wiki/OpenStack_health_tracker | 15:40 |
| smcginnis | dhellmann: Looks good. | 15:41 |
| zaneb | +1 to removing traffic lights | 15:41 |
| * ttx sobs | 15:41 | |
| cdent | don't worry ttx, there will be another chance | 15:42 |
| ttx | I'll find a way to reuse my art in another wiki page | 15:42 |
| hongbin | ttx: from my point of view, it is better for me to directly involved on making the final judgement on my project | 15:42 |
| *** e0ne has quit IRC | 15:42 | |
| smcginnis | Hopefully the notes there are from your feedback though. | 15:42 |
| dhellmann | and I removed the traffic lights | 15:42 |
| cdent | hongbin: I don't think judgement is what is really happening here, but if it were, then people who within something can never be accurate judges of their own situation. | 15:42 |
| ricolin | ttx put it in your twitter will works:) | 15:42 |
| dhellmann | hongbin : your input will be included. The point of this exercise is for the TC members to *learn* about the state of your project, not to judge it. | 15:43 |
| hongbin | cdent: ack | 15:43 |
| dims | thanks for bringing up the concerns hongbin | 15:43 |
| ttx | dhellmann: how do you propose that we raise awareness amongst TC members as to projects that experience difficulties ? Should we talk about it here ? Discuss it in a thread ? | 15:43 |
| hongbin | dhellmann: ack, get that, i would like to involve on what is printed on the wiki | 15:44 |
| ttx | or just ask members to regularly read that wiki page and estimate based on the notes | 15:44 |
| dhellmann | ttx: that's a good question for us to answer, now that we have some data to look at | 15:44 |
| hongbin | dhellmann: you can ask me to give a draft of the report, and have the TCs to review it, if it is fine, i write it to the wiki, i think this is a better way | 15:45 |
| dhellmann | hongbin : no, that is not ok | 15:45 |
| dhellmann | I want to be absolutely clear on this | 15:45 |
| hongbin | :( | 15:45 |
| dhellmann | this is a TC activity, and I want TC members doing it | 15:45 |
| dhellmann | they should talk to team members, but we want an outside perspective about what is going on with each team | 15:45 |
| dhellmann | so your input can be included, but I do not want you to write the notes | 15:46 |
| dhellmann | has that been unclear to anyone else? | 15:46 |
| dims | dhellmann : i agree. pretty clear to me | 15:46 |
| ttx | dhellmann: personally I used that page in two ways: quickly spot red lights to know which projects are in trouble, and when I have questions on a specific project. read the latest update. I'm mostly concerned about missing red flags, so maybe a policy of raising a discussion for every team we found in trouble would be enough. | 15:46 |
| hongbin | yes, it is clear, but i would have a bit concerns on that | 15:46 |
| dhellmann | hongbin : I am concerned that you are so concerned about participating in this. | 15:47 |
| zaneb | dhellmann: so it's probably wrong that we encouraged TC members to sign up to be liaisons for projects we were already involved with? | 15:47 |
| cdent | hongbin: can you explain in more detail what you are worried about? | 15:47 |
| dhellmann | zaneb : I think we only have 1 where both members are? But yeah, I thought about that and decided that we should go through the exercise at least once. | 15:47 |
| ttx | It's a bit frustrating that we would not even be discussing this if the wiki page was somehow only accessible to TC-members (or a document on Google docs) | 15:48 |
| dhellmann | it was more important to get everyone to agree to do it at all than that we do it perfectly the first time | 15:48 |
| hongbin | cdent: the real problem is what printed on the wiki will have impact on the visibility of individual project | 15:48 |
| hongbin | (IMO) | 15:48 |
| dhellmann | hongbin : I don't understand how that will happen. Can you give a more specific example of what you are worried about with "visibility"? | 15:48 |
| cdent | hongbin: do you think that people in the TC will not be impartial? | 15:48 |
| smcginnis | hongbin: That the visibility won't be good for some projects? | 15:48 |
| zaneb | ttx: maybe we should have our own private slack channel? :P | 15:48 |
| smcginnis | Someone ban zaneb. :P | 15:49 |
| dims | also if you do spot something that you don't like, you could raise it here and we could figure out what to do then? | 15:49 |
| zaneb | lol | 15:49 |
| ttx | zaneb: #slackjokes is the new #dadjokes | 15:49 |
| dhellmann | hongbin : you are talking about zun, right? | 15:50 |
| hongbin | dhellmann: for example, based on stats, some projects is single vendor, but it can be presented in different ways | 15:50 |
| hongbin | dhellmann: you can say a project is single vendor, or you can say the trend of the project is becoming mature and growing its diversity | 15:50 |
| dhellmann | yes, well, that's a good reason for dropping simple numbers and having a conversation about diversity instead | 15:50 |
| hongbin | dhellmann: it is totally depends on what aspect you looks at the stats | 15:51 |
| dhellmann | and I think pointing to a trend or anticipated contributions would be a good thing to have in these notes | 15:51 |
| hongbin | what printed on the wiki would be the same, it is based on collected inputs, but can be presented in a different way | 15:52 |
| dhellmann | zaneb , ttx : you are signed up to cover zun. Do you have any idea when you might be ready to talk to that team? | 15:52 |
| ttx | I can prioritize it up, but won't be before the ~17th | 15:52 |
| ttx | happy to let zaneb do it though | 15:53 |
| zaneb | having just learned about the ZTE situation, I'm inclined to push Zun up to the top of my list | 15:53 |
| dhellmann | does zte affect zun, too? | 15:53 |
| smcginnis | hongbin: I think part of the goal of having these discussion based notes collected on the wiki is to get away from some of the concerns you are raising and getting rid of diversity tags and stackalytics numbers as the only way to check on the status of a project. | 15:53 |
| hongbin | a bit, because the PTL is from ZTE | 15:53 |
| ttx | dhellmann: primarily affects zun yes | 15:53 |
| dhellmann | murano's ptl is from zte, too | 15:53 |
| dhellmann | https://wiki.openstack.org/wiki/OpenStack_health_tracker#Murano | 15:54 |
| ttx | ack | 15:54 |
| hongbin | zun is a bit better, because we have contributors from other companies, just the speed of the reviews is slower than before | 15:54 |
| zaneb | they were ~25% of contributions in Queens I think | 15:55 |
| *** dtruong has joined #openstack-tc | 15:55 | |
| zaneb | so not fatal by any means, but probably painful to lose them | 15:55 |
| ttx | anyway, zun is a good example of why the tag doesn't capture the subtlety of the situation | 15:55 |
| dhellmann | zaneb: if you could try to reach out to fengshengqin next week some time, that would be good. | 15:55 |
| cdent | I guess I'm a bit confused here. I would think that showing on a wiki page that a project has less resources than it might otherwise need is a good thing: It's an invitation. | 15:55 |
| cdent | Is the concern that it will make people send it to the trash? | 15:56 |
| hongbin | smcginnis: ack | 15:56 |
| mnaser | I don’t think anyone actually seeks out or can easily find the health tracker page anyways | 15:57 |
| cmurphy | the point of the notes is so that we can help projects, not judge them | 15:57 |
| flaper87 | ttx++ for blaming me! I can't think of any other person to blame for things in life! | 15:57 |
| ttx | flaper87: I think you've been missing some blaming lately | 15:58 |
| dhellmann | it seems like the next step is to actually do the zun review, and then we can discuss the notes more concretely | 15:58 |
| * hongbin is heading to lunch | 16:00 | |
| fungi | perhaps calling it a "health tracker" also implies some sort of biased judgement | 16:00 |
| fungi | to some people stumbling across it | 16:01 |
| cdent | I don't think we should feel bad about using either the terms health or judgement | 16:01 |
| cdent | It is what we are measuring and doing. | 16:01 |
| ttx | health notes ? | 16:01 |
| fungi | yes, i expect having a good description of what it is and isn't addresses that concern | 16:01 |
| dhellmann | I would rather we argue about whether information is accurate after we collect it than worry that information we don't have written down yet will be seen in some sort of bad light. | 16:02 |
| scas | people will interpret things how they will, regardless of definition | 16:02 |
| dhellmann | I understand hongbin's concern, and I think we can address it | 16:02 |
| dhellmann | concerns/them | 16:02 |
| fungi | i concur on it being an understandable concern we can attempt to mitigate | 16:03 |
| dhellmann | if we have to hide these review notes to keep people happy, then I think that says an awful lot about how project teams see their place in the community | 16:03 |
| ttx | dhellmann: yes, I would focus on making sure people do not read too much into that page, rather than on ensuring fairness/accuracy in what will remain a subjective set of notes | 16:03 |
| ttx | Oh! we could rename the page every week to make sure nobody can permalink to it :) | 16:05 |
| scas | a project like chef can be interpreted several different ways, depending on one's own bias. no matter what you call the metric, someone will interpret it their own way | 16:05 |
| ttx | You can tell we reached beer o'clock here | 16:05 |
| dhellmann | in the mean time, thank you to all of the folks who have checked in with teams so far; we are collecting a lot of good information | 16:05 |
| dhellmann | I hope you're finding the exercise as enlightening as I am | 16:06 |
| scas | it coerced me out of my hole in the 'net, for better or worse | 16:07 |
| dhellmann | heh | 16:07 |
| jroll | what's "the ZTE situation"? | 16:08 |
| dhellmann | jroll : https://www.nbcnews.com/politics/congress/senate-blocks-zte-deal-rebuke-trump-deal-n882196 | 16:09 |
| jroll | another corp pulling dev? | 16:09 |
| scas | you could call the metric 'stability' and someone will still interpret it incorrectly | 16:09 |
| * jroll clicks | 16:09 | |
| dhellmann | that's not the most recent article on the topic | 16:09 |
| jroll | ah, this | 16:09 |
| jroll | thanks | 16:10 |
| * dhellmann heads off to find lunch | 16:11 | |
| *** tosky has joined #openstack-tc | 16:11 | |
| *** jpich has quit IRC | 16:18 | |
| *** gcb has joined #openstack-tc | 16:28 | |
| *** annabelleB has joined #openstack-tc | 16:35 | |
| *** gcb has quit IRC | 16:36 | |
| ttx | mnaser: Blazar should appear on project navigator later today, sorry for the error | 16:39 |
| *** ricolin has quit IRC | 16:43 | |
| *** annabelleB has quit IRC | 17:01 | |
| *** e0ne has joined #openstack-tc | 17:03 | |
| *** diablo_rojo has joined #openstack-tc | 17:03 | |
| *** dtantsur is now known as dtantsur|afk | 17:26 | |
| mnaser | ttx: np, yay, thanks, ill let the team now | 17:35 |
| *** e0ne has quit IRC | 17:36 | |
| *** AlanClark has quit IRC | 17:53 | |
| *** e0ne has joined #openstack-tc | 18:01 | |
| *** e0ne_ has joined #openstack-tc | 18:03 | |
| *** e0ne__ has joined #openstack-tc | 18:05 | |
| *** e0ne has quit IRC | 18:06 | |
| *** e0ne_ has quit IRC | 18:08 | |
| openstackgerrit | Doug Hellmann proposed openstack/governance master: add check_review_status.py https://review.openstack.org/579953 | 18:09 |
| openstackgerrit | Merged openstack/governance master: Octavia asserts supports-upgrade https://review.openstack.org/577967 | 18:19 |
| openstackgerrit | Merged openstack/governance master: Octavia asserts supports-accessible-upgrade https://review.openstack.org/577970 | 18:23 |
| *** purplerbot has quit IRC | 18:48 | |
| *** hongbin has quit IRC | 18:48 | |
| *** hongbin has joined #openstack-tc | 18:48 | |
| *** purplerbot has joined #openstack-tc | 18:48 | |
| *** cdent has quit IRC | 18:53 | |
| *** flwang1 has quit IRC | 20:03 | |
| zaneb | lol "It’s hard to do anything [at Google] because for any idea you propose, three or four teams will run in shouting that they own that, you can’t touch it, and oh by the way, they’re not working on it for a few years. Microsoft calls this “cookie-licking”" | 20:16 |
| jaypipes | heh | 20:19 |
| openstackgerrit | Doug Hellmann proposed openstack/governance master: update the pti to use tox for building docs and releasenotes https://review.openstack.org/580495 | 20:21 |
| *** gouthamr has quit IRC | 20:27 | |
| smcginnis | Awesome term to use. :D | 20:28 |
| openstackgerrit | Doug Hellmann proposed openstack/governance master: update the pti to use tox for building docs https://review.openstack.org/580495 | 20:34 |
| *** cdent has joined #openstack-tc | 20:58 | |
| *** flwang1 has joined #openstack-tc | 21:06 | |
| tosky | dhellmann: I commented on 580495, but I'm not sure why the switch to bindep.txt at the same time; most projects have a tox environment working right now with doc/requirements.txt | 21:09 |
| dhellmann | tosky : bindep is not a new change. it's how you install things that might be needed to get your python dependencies to install. Like python. :-) | 21:10 |
| tosky | dhellmann: I know that it's not new, but why change now? Is there a reason why this would not work with python3? http://git.openstack.org/cgit/openstack/sahara/tree/tox.ini#n58 | 21:11 |
| tosky | it's working at least locally | 21:11 |
| dhellmann | tosky : if you look carefully at the diff, it is not a new thing. | 21:11 |
| tosky | dhellmann: the diff removes it | 21:12 |
| tosky | removes doc/requirements.txt | 21:12 |
| tosky | and the tons of patches that were sent around blindly used just one change | 21:12 |
| dhellmann | do we specify the other requirements files for other tox environments? | 21:13 |
| tosky | for tests | 21:13 |
| dhellmann | ok, I can restore that part | 21:13 |
| dhellmann | actually, it's still there | 21:14 |
| dhellmann | look at line 53 | 21:14 |
| tosky | I see that, correct, sorry | 21:15 |
| tosky | ok, thanks for claryfing; I'm a bit scared every time there is a potential change which produces a rush of patches | 21:17 |
| tosky | sorry again | 21:17 |
| dhellmann | yeah, all of this is going to roll out as part of the python3 stuff next cycle, so we shouldn't need any changes right now | 21:18 |
| *** cdent has quit IRC | 21:33 | |
| *** gouthamr has joined #openstack-tc | 21:35 | |
| *** rosmaita has quit IRC | 21:36 | |
| *** e0ne__ has quit IRC | 21:41 | |
| *** e0ne has joined #openstack-tc | 22:00 | |
| *** annabelleB has joined #openstack-tc | 22:15 | |
| *** edmondsw_ has quit IRC | 22:40 | |
| *** edmondsw has joined #openstack-tc | 22:41 | |
| *** edmondsw has quit IRC | 22:45 | |
| *** e0ne has quit IRC | 22:50 | |
| *** hongbin has quit IRC | 22:50 | |
| *** tosky has quit IRC | 23:01 | |
| *** ianychoi_ has joined #openstack-tc | 23:01 | |
| *** ianychoi has quit IRC | 23:04 | |
| *** mriedem has quit IRC | 23:08 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!