14:07:49 <spotz> #startmeeting divsersity-wg
14:07:49 <opendevmeet> Meeting started Tue Feb 14 14:07:49 2023 UTC and is due to finish in 60 minutes.  The chair is spotz. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:07:49 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:07:49 <opendevmeet> The meeting name has been set to 'divsersity_wg'
14:07:56 <spotz> #chair fungi
14:07:56 <opendevmeet> Current chairs: fungi spotz
14:08:03 <spotz> #topic ROll Call
14:08:30 <spotz> Checking to see if anyone from any of he projects are here
14:09:00 <fungi> openstack's community manager is present and accounted for
14:09:26 <spotz> Whoohoo. Any other projects?
14:10:02 <fungi> sometimes ildikov appears during these meetings, but i don't know whether she's travelling at the moment
14:10:29 <ildikov> o/
14:10:46 <ildikov> at a call also, but should be able to follow the convo here
14:10:48 <fungi> so we've got the starlingx community manager too
14:11:06 <spotz> Sweet, I was typing I wasn't sure if you'd just stayed in Europe after FOSDEM or not
14:11:56 <ildikov> spotz: I stayed, not on my own will though, but that doesn't really matter :)
14:12:15 <spotz> ildikov: oh no:(
14:12:17 <ildikov> I wonder if PagliaccisCloud[m] is around too?
14:12:50 <spotz> #chair ildikov
14:12:50 <opendevmeet> Current chairs: fungi ildikov spotz
14:12:56 <ildikov> spotz: it's fine, working on a resolution right now
14:13:37 <spotz> Started an agenda area on the etherpad but I think we have 2 main topics Summit and Survey
14:14:13 <spotz> s we discussed them in that order last time let's do the same
14:14:22 <spotz> #topic Summit Planning
14:14:44 <spotz> fungi you're typing in last month:)
14:15:01 * fungi lives in the past
14:15:56 <spotz> As fungi is typing he completed his action item of sending the list of things we've done in the past and would like to see again in Vancouver. I don't think there was anything new except maybe the distance badge?
14:17:18 <fungi> i can put the prosified recommendations into the pad too, for posterity
14:17:26 <ildikov> I haven't been in Berlin, so I don't have the comparison
14:18:00 <ildikov> *wasn't in Berlin
14:18:23 <spotz> #link https://etherpad.opendev.org/p/diversity-wg-agenda
14:18:42 <ildikov> But the has a list of items that are familiar, I didn't see anything completely unexpected
14:19:15 <spotz> Berlin might have had the distance thing because it was the first post-COVID thing for us
14:19:50 <spotz> We did do a CHAOSS Event badge for Berlin which was new
14:20:44 <spotz> And we recommended applying again even without being able to offer child care
14:21:52 <spotz> ildikov anything else you'd like to see added? Otherwise that item I think is on hold for now until we get a reply
14:22:18 <ildikov> spotz: there's nothing else I can think of from the top of my head
14:22:37 <ildikov> I think the email is a very comprehensive list
14:23:16 <spotz> Good and the hope is that it becomes just another checklist of we need to get these done for the event
14:23:33 <spotz> #topic 2023 D&I Survey
14:24:11 <spotz> So fungi cleared off his other action item by emailing Allison but she's traveling this week so we're a little on hold.
14:24:27 <fungi> yeah, it's on her radar now at least
14:24:43 <fungi> and she had an alternate suggestion which might work out better
14:24:46 <spotz> She did recommend only one survey with sections for the different projects
14:24:57 <PagliaccisCloud[m]> o/ i'm here
14:25:07 <spotz> Hey PagliaccisCloud[m]
14:25:09 <fungi> basically have the sections "hidden" unless the respondent indicates they participate in that project
14:25:20 <spotz> #chair PagliaccisCloud[m]
14:25:20 <opendevmeet> Current chairs: PagliaccisCloud[m] fungi ildikov spotz
14:25:35 <fungi> that way we get one response with separate answers per project, rather than multiple separate responses from the same person
14:26:24 <ildikov> PagliaccisCloud[m]: hi! We just finished talking about D&I related actions for the upcoming OpenInfra Summit, and switched over to the D&I survey that I mentioned to OpenInfra community managers already
14:26:32 <spotz> The part I'm going to probably going to need to see is if we're going to repeat questions or work out some way to ask a question only once but have it mean something for multiple
14:27:20 <fungi> i've also previously added a bunch of notes to the survey update brainstorming pad
14:27:24 <spotz> I think I get and like what you're saying fungi, but I'm visual so have to see it to know if it's what's in my head:)
14:27:52 <fungi> #link https://etherpad.opendev.org/p/diversity-survey-2023
14:29:17 <spotz> fungi looking through that, line 29 is actually where the Git and Gerrit Lunch and Learn came from
14:30:02 <spotz> We might need to break it out into 2 questions, but I think knowing folks from different areas of the world might need help with tools is important
14:30:37 <fungi> yes, i still think it's useful to separate "contributing to the project is hard" from "contributing to the project is hard for underrepresented groups but not for others"
14:31:14 <fungi> it's possible for people of all races, genders, ethnicities, nationalities, et cetera to struggle with git
14:31:48 <fungi> using gerrit vs using github doesn't seem to be a diversity-oriented problem, it's just a general contributor problem
14:32:34 <spotz> I can see that cause I rock with gerrit and hate doing PRs:)
14:32:58 <fungi> i don't mind us helping tackle general contributor problems and getting more people involved, i just wouldn't want to mislead people by equating that with diversity problems
14:34:08 <fungi> and by putting wrong or vague questions on a diversity survey, we risk drawing unhelpful conclusions from the answers we receive
14:34:40 <spotz> Ok so let's mark 29 for rewording for sure
14:35:12 <fungi> but also a number of other questions in there smack of americentrism or occidentalism
14:36:10 <spotz> A lot of the questions were created by the first group which was very gobal. One reason we don't ask ethnicity is it was very american
14:36:47 <fungi> getting meaningful answers from a globally distributed community will work better if we can avoid trying to shoehorn respondents into our preconceived categorizations
14:37:03 <spotz> Education we can definitely update for more global
14:38:18 <spotz> Yeah 94 didn't make it into the survey last go round
14:39:10 <fungi> at line 38 i'm curious what options we have to deal with there being people who are offended by seeing taiwan in a list of countries and other people who are insulted by taiwan not being listed (or ukraine for that matter)
14:39:15 <PagliaccisCloud[m]> can i make a suggestion on line 54 & 59 to change "racial/ethnic minority" to "member of an underrepresented racial or ethnic group"?
14:39:29 <ildikov> big +1 from separating tools from diversity challenges
14:39:44 <spotz> PagliaccisCloud[m]: please do!
14:39:58 <ildikov> I always struggle with GH too...
14:41:12 <fungi> PagliaccisCloud[m]: that's an excellent point. also we should make sure questions like that are scoped to our communities. for example, people of mexican heritage may be an underrepresented ethnic group in the usa but are a majority representation in mexico
14:41:47 <fungi> do we want to know if people are underrepresented in the country where they reside, or underrepresented within our open source community demographics?
14:41:56 <PagliaccisCloud[m]> ^
14:42:22 <spotz> Keep in mind the original survey was liberty/mitaka time fram and then the last one was before the first Berlin? So things can definitely be reworded we're just trying to get enough consistency to have comparisons
14:42:24 <ildikov> fungi: for line 38, if there's the option to have a field that gives suggestions, then I would be good with that. Free-form responses might have a lot of typos.
14:42:48 <spotz> I thought we asked 2, underrepresented where you reside and another for in the community
14:43:23 <fungi> yeah, we do separate them
14:43:45 <fungi> just reminding that we need to keep them as separate questions
14:44:05 <spotz> The idea behind that is you could be in the majority in where you live but be a minority in OpenStack/Kata/etc
14:46:17 <ildikov> what do we use that information?
14:46:30 <ildikov> *for
14:47:03 <ildikov> I mean someone being in the majority where they reside while not in a global community
14:47:20 <fungi> that's one of the larger dilemmas for this survey. i think we've thrown a lot of questions in there in hopes that we may be able to draw conclusions from some of them, but without a clear idea of what we're actually trying to determine
14:48:00 <ildikov> do we have a more clear view on that by now, or is that still one of the exercises to complete?
14:48:14 <ildikov> or, are we ok with collecting more data than what we might need?
14:48:16 <fungi> if we were to recreate this from scratch, i would want to start with a set of challenges we're trying to solve, then figure out what questions we should ask in order to get the information we need to work out solutions for those problems
14:48:40 <spotz> It's more of an inclusion question. Keep in mind we can't go back and say 5 of these surveys were from X project in OpenStack do we need to see what's going on. But to get folks to answer any questions there's no way of knowing the 5 concerning responses came from X
14:50:11 <ildikov> spotz: I'm not sure I fully follow your last comment
14:50:54 <fungi> the main risks i see with having too many questions not all of which are necessarily relevant, are 1. people are less likely to complete longer surveys or it increases the self-selection bias by imbalancing for people who are more likely to complete long surveys, 2. some prying questions may unnerve or offend respondents leading them to abandon the survey, 3. we have to sort through
14:50:56 <fungi> the extra data even if we're not going to use those answers for anything meaningful
14:51:05 <ildikov> how would knowing the representation by location make a difference there?
14:51:24 <spotz> The survey is so anonymous it makes drawing deeper conclusions hard. Say 5 people felt discriminated in Masakari(chosen as it's inactive) but we can't use the survey results to know where the problem is only that there is a problem
14:51:54 <ildikov> ok, that I understand
14:52:35 <fungi> ildikov: one way i can see "underrepresented ethnic group where you reside" coming into play is if we equate that to challenges around lack of resources (lower income families, harder to get access to computers or education, ...)
14:52:36 <ildikov> but to fungi's point, if we feel there are problems, we can design the survey to try to collect information on those areas
14:53:27 <spotz> Yes which is why before each survey we try to update and refine the survey to get more useful answers
14:53:45 <ildikov> fungi: that could still be a regional problem as well as a problem that is specific to the individual
14:53:49 <fungi> however, it's also possibly incorrect to assume that just because an ethnic group is underrepresented in a particular locale, that they necessarily have less access to resources. in some cultires there are minority ruling classes or castes who have greater access to resources than the majority
14:54:48 <fungi> so we have to be careful where it comes to asking questions which we consider proxies for the information we want, as those proxies may not hold true globally
14:55:05 <ildikov> the underrepresented person could also visit a poor region, where the individual has more resources
14:55:31 <spotz> I think the reason for that initial questions is say yo are part of X race and Y social status where you live and you're the majority there, you might come to a project and be a minotrity. That was the initial reasoning behind it(and damn ithink I'm the only one still around from the first survey:()
14:55:46 <fungi> if we want to know whether a respondent has challenges with access to computers or higher education, then we should ask that rather than asking something we (perhaps incorrectly) believe will result in that
14:58:20 <fungi> spitballing... "do you struggle to get access to education or computer resources in the country where you reside?"
14:58:25 <spotz> Ok so why don't we work on something like that, we'll potentially lose the tie in to past data but I'm pretty sure we can grab a CHAOSS metric just for that and tweek if needed
15:00:29 <spotz> And we're at time
15:00:45 <fungi> thanks spotz!
15:01:46 <spotz> This was great having more input! Each time we do this the survey gets better. I'll tell the historical reasoning so we have it but we should always strive to make it better
15:01:52 <spotz> #endmeeting