 Alrighty, folks. We're gonna go ahead and get started. This is the Let's Talk Diversity in our community. And I'm gonna go close the door and turn it over to our panelists. So we're gonna really have a dialogue here more than conveying information. We really wanna share information and gain more knowledge as a result of that. So hopefully everybody can go ahead and join in the conversation with us today. Thank you, Carol. Yeah, thank you. For those that don't know, Carol, she's doing a lot of work on our diversity working group. So we'll start with introductions. I am Agnes Sigler. I work for Ack Space. I am also on the board of directors for the OpenStack Foundation. And I am one of the people that are on the diversity working group. And I care about diversity because I think we don't have enough women in the OpenStack community. I really would want to work with more of them and see more of them. Now that there is anything wrong with men, it's just the proportions are so bad. Yeah, so bad. Yeah, hi. I'm Imad Susu. I'm with Intel. I'm also on the OpenStack Foundation board and part of the diversity working group. Actually, the three of us started the diversity working group in the OpenStack Foundation. I mean, I believe diversity is important beyond that it is right. And this is, you know, Agnes talked eloquently about this. I think also just, you know, just bring aside all of that from a strict company and organizational perspective. I think diverse organizations, whether they're companies or open source projects, or whichever type of institution just makes those institutions just that much better. There's actually a lot of research on this topic and we can talk about it more. There's research with a lot of numbers, but there's also a lot of research that was done by, I believe, at Stanford where it shows that a diverse group have a collective IQ much better than a non-diverse group. And this is a lot of studies has been done in this area. And I think, you know, as an organization, whether from an Intel perspective or from an OpenStack perspective, I have the vested interest that, you know, our organization, that we are collectively smarter. So anyway, go ahead. Hi, I'm Kavit Munshi. I work for a company called Aptira. I'm also on the OpenStack Foundation board. And diversity for me means inclusiveness, right? I want more voices from everywhere to be heard. And I want the community to become better for it. And it's really difficult to see a certain pattern repeating every year and not having any change. And I thought something needs to be done about that. And I guess once I got on the board, diversity was something I pursued, and here we are. So hopefully we can use this opportunity to get something meaningful. And yeah, let's get started. Right. So everybody that is an OpenStack Foundation member has to register to become one. And when registering, there's a couple of questions. And one of them are male or female. And I think this data is slightly old because the foundation is growing so fast. And I think now we have closer to 30,000 members rather than 20,000. And oh, it's 20,000 that have responded because this question has been asked, was not asked initially. So only 9% of our community are women or identify themselves as women. So these numbers are really not very good at all. They're actually worse than industry average, at least in the US. And I'm sure it's worse than the rest of the world. When it comes to geographic diversity from the member registration, the United States is 36%. India is 14% and China is almost 13%. However, when we look at the participation in the community, when we look at PTLs, court reviewers, even the board, the numbers are very skewed more towards US rather than these other countries that are well represented in the community. A couple of weeks ago, we had sent out a diversity survey that was geared together more information about our community beyond what has been asked in the registration form. And we did not get a very great response. However, we have received a lot of great comments in the survey that were really interesting and in some ways surprising. 538 people responded out of 32,000 that have received the surveilling or that were sent out. Now, in some cases, we heard, I never got a survey. Okay, well, do you get other emails? Yeah, it could have been lost because if you are part of OpenStack community, there are lots of mailing lists and sometimes the amount of emails is quite large. I think I have received maybe, I sent out a follow-up reminder to the community and saying, hey, if you don't have the link, let me know. I'll provide it for you. And I think only a couple of people reached out to me and say, hey, I can't find it. Can you send it to me again? So we don't know whether it got lost or people don't care. I think someone also mentioned in the working group that 2% is a very good response rate. Well, I don't know about that, but I was hoping for a little larger one. But then again, I care about these things. But I do think 538 people is a good sample. Yes, we see some idea of where... It's a very good cross-section of what the community looks like. Yes, and overall there are 9% women in the community and 13% of the survey takers identified themselves as women. So it was not too far off, but a good number of women responded. We had also asked people if they faced any barriers to participate in OpenSAC community. We did not ask whether it was their... They were facing diversity-related barriers. It was just an open-ended question whether they're facing any barriers. And so only about half of the people said they have no barriers. The other ones were from few barriers to several to many. And we had received a lot of comments on what kind of barriers they face. And this is where it becomes really interesting. As you see, some of these barriers are not really diversity-related. They're technology-related, they're process-related. And some of them were diversity-related. Some of them were like, I don't have time to work on this. So yes, it's a barrier, but it's not something we can fix. Not right now anyways. But the process and IRC was a big one. IRC, people were saying, I am in China, it's blocked. Or my company is blocking, I cannot participate. Or they were saying, I don't use IRC because people are not responsive. Or I ask a question and nobody ever responds to me. Or things related more towards the community. Process, it was also similar. The documentation is hard to find or is nonexistent. Or they were saying it's not up to date. Or everything that is documentation-related that could potentially get fixed. But it is a lot of work. Hard to get the code upstream. This was also a big portion of the commons. Yes. The process is too complicated. It's antiquated. There were a lot of complaints about... People thought there was a cabal of a clique. They thought that there were certain patches that wouldn't get through. Just because there was an organized existing kind of elitism going on. And it was very hard to hear about... Right. It's the large projects that we all know. And we probably know people working on these. And we're like, wow, I had no idea that it was so hard for newcomers. Now, in some cases, it might be reasonable to think that for a newcomer, it's going to be hard to contribute to NOVA because of the scope of the project and the complexity. Now, I don't know. It might not be the case. Yes. Okay. So yes, it's a huge barrier. Yeah, but I'm not saying the complaints might have been valid or invalid. This is about the perception out there. People were complaining about the review process and they were complaining about not being heard. So if their patch was rejected, they felt like they weren't given a decent enough reason. I don't know. I think this leaving diversity aside, this actually highlights the need for all the projects, TC, anyone to do feedback. Like we need feedback from the community, from the developers as to what is going on. I mean, this was not just a surprise to us. This was a surprise to the TC. We know that more like almost 50% of the people were facing barriers. If you were a traditional software company, this would be very, very bad. I mean, this would be considered catastrophic. So I think leave diversity aside, but there needs to be some process for feedback, some process for change and improvement. Right. The next common barrier was time zones that most meetings are US and Europe time zone centered. Now, a lot of large projects do run meetings in US centric zone and Asia Pacific centric zone. NOVA is one of them. And I did talk to the NOVA PTL and say, hey, what do you guys do? And he said, yeah, we have great participation in both time zones. So I don't know if it's really like, even though people do have the two different time zone meetings, it can still be inconvenient for quite a few people. So it's hard to have a time that works for everybody. But I think what happens there is exactly what's happening with the diverse working group where people, the vast majority of people tend to join the US Europe and end up with a very, very small meeting for that. Right. So at least in NOVA case, they have about equal participation. NOVA is probably an anomaly, though. I mean, given the size of the project. Yeah. It's probably true. And NOVA has a former PTL that is actually based in Australia. But yes, I'm sure that's an anomaly for diversity working group. We don't get a lot of participation in the Asia Pacific time zone. For deaf core, we also try to run it into different time zones. And in both time zones, we were getting the same people, except in the late evening meeting, we were getting fewer people and almost none or sometimes one from Asia Pacific region. If you're not familiar, Diversity Working Group has a charter. Please take a look at it. On the wiki, it has been approved by the board. And our main goal is to foster an inclusive and welcoming place for all people. And by diversity, we mean all kinds of diversity. However, we do have a work plan on how to increase diversity in phases because diversity can include so many different aspects. In phase one, we focus only on gender identity and expression, sexual orientation, geography and culture. And follow in next two phases, race, ethnicity, religion, education. In phase three, age and disabilities. For our plan, we have to establish a data-driven baseline and already talked about some of that data. And we sent out the survey a little over two weeks ago and we received the results back in the middle of last week. So we didn't really have a lot of time to slice the data in different ways and really go through it. I think we are now going to work on analyzing the data a bit more and then establishing some goals and some results. So we can then reach out to the board, to the TC. Because we've already spoken to the TC and we have volunteers who are willing to help us. But we need goals, we need metrics that we can achieve and we can, targets we can hit. And I think over the coming weeks, we'll dissect the data a bit more and take it from there. Yes, that's one of the slices that we want to take a look at. So as I mentioned, we received only 534 responses total. The first one, that is something we should probably take a look at. So now we would really like to open the panel for discussion and we have some topics to discuss so we can either discuss them or you can try to ask us questions. We do have a mic and since it's being recorded, it would be great if everyone use the microphone before speaking. Carol has graciously agreed to help us. You know, I think one of the areas that companies are addressing is doing unconscious bias training across companies because that really unhurt a lot of hidden biases and barriers that people may put up without even knowing. Have you thought about doing something like that, at least for the leadership of OpenStack so that they can be more inclusive? You know, I think it's a great idea. No, we have not thought of it, but I think it would be great if BTLs and Quarry viewers would go through it. I don't know how reasonable it is. We can make it. It's a one-hour training available on the Facebook page. I can send the link to you guys. Which one is that, Ruchi? I attended a Grace Hopper's conference last week, a couple of weeks ago, and Cheryl Sandberg talked about it. She sent out the link and I can send it. Yeah, Mori and I were at the conference also a couple of weeks ago and there was an amazing speaker who talked about that and I think actually it would help. Maybe we can start at least with the joint board to see meeting next time and just go from there, scale it from there. And R is not unreasonable. Right, and I don't think it would be unreasonable to ask BTLs and Quarry viewers to go through it, especially if it's only one hour. But that's a great suggestion. Speaking about the unconscious bias, I mean, in that conference that we'll talk about, which was a professor I cannot remember whether it's UCLA or UC Berkeley, that they did this study and they had this one resume that the resume was exactly the same. I mean, exactly the same. But with one difference. One resume used a name that sounds African American, Latisha or so on, and the other one used a white American sounding name, Lauren or something like that. And this was distributed fairly widely among companies and the response rate from managers were 50% more positive towards the resume of the person with the name Lauren, which is the exactly same resume. And this was done like a fairly broad study, so it's not like an anomaly or something like that. So the unconscious, what Nithya talked about, the unconscious bias, it's actually a very real bias that ends up really hurting us because you're ending up not including people based on their technical merit and so on, because obviously the 50% called in Lauren because of technical merit. So you end up missing out on a lot of talent when you do that. I know this has been discussed somewhat on the IRC channel, but I think our code of conduct could be more specific towards sexism and racism. Right now it's geared towards people who are, I think, on the OpenStack Foundation, things like don't take bribes, and there's many open source projects that have already adapted and expanded their code of conduct, so if I face some sexism or racism, I have nowhere to report it to. So I think there is a way to report it. It might not be obvious how to do it, but I did talk to Lauren from the Foundation and she said that those complaints do go to Jonathan Bryce and they get about one complaint or less per year. Now, I don't know if it's because it's not obvious how to report it or whether it's not happening. Now, I think it's probably the reporting issue. I think just because we are getting a question from the audience that they don't probably know about the mechanism being there, I would say it should be made well known. It should be publicized. Right, and that's why I'm saying that it's not obvious how to report. I've actually seen the code of conduct and it doesn't say anything about facing sexism. So it's not included in there. So even if I did encounter someone being sexist against me, the code of conduct doesn't state that that's against it. It just says, like, respect people, but you can't really go and report it specifically. Yeah, I'm curious. Just a question to you. Do you have a code of conduct that you like? Actually, I did. No, I'm asking that seriously because before we invest time, like I'll give you an example. We've been discussing the DCO and we just decided to use the Linux Foundation DCO because it was, you know, so it would be really good if there is, if somebody knows of something that is, it would be great to just adopt it. So I did talk to the TC board about this for working the Django code of conduct, which is a really good code of conduct. It has a reporting guide. It has an enforcement manual. So if someone, like it doesn't, your code of conduct doesn't work unless someone is enforcing it, right? If I have a way to communicate with someone and someone actually is, you know, talk to you or, you know, there's a way for it to become change. And it also has, like, a change log. And I have brought this up to the TC, but... Can you send it to one of us? Yeah, and I know we talked about some of it during the meetings, but I think other things took priority, so we definitely can bring it back up. But I don't think with those kind of changes. Yeah, and so there is, I think, general code of conduct and then there is code of conduct for each summit and the summit one, I think, is slightly different. It addresses different things. I think the Linux Foundation did update recently their code of conduct to include something along what you described. So we should just take a look at that as well. Yeah. What did the Linux Foundation do? We are in the business of... Oh, you did? Oh, really? Yeah, just send it to us. That would be great then. Oh, you did? Awesome, thank you. Hey, thanks. Yes, for those of you who don't know, we have an IRC channel. Are you in China? Or is this a company issue or a country issue? Any other questions or comments? So yes, we have a few prepared discussion topics. How do we attract more diverse people to join our community? And by diversity, right now we're focusing on the gender geography and... Gender, geography, and sexual orientation. Yes. So how do we attract that? Because right now we supposedly have a pretty good geographical diversity when it comes to members, but they're not really participating or it's not being represented. Much like pipeline discussions around gender focus on pipeline and not retention. I mean, is that a secondary piece to this question or is that something that we're not discussing? I'm trying to figure out what we're focused on. I think the pipeline question is a great question and it's definitely something we need to address. When we did a survey, a lot of the results that we saw was one of the things that we saw was that OpenStack community is not really no longer a developer-centric community anymore. We now have people participating in many, many different ways. So yes, when it comes to development or technical side, we need to have a good pipeline. But from other side, it's probably people that work at the companies and they get involved, I suppose. But how do we get them to speak up in community forums or mailing lists or IRC? Well, and to stick around. And stick around. Certainly in the technical track, you're not going to become a core reviewer if you don't stick around. You might not even ever get a patchlanded if you don't stick around. And I think that kind of stickiness aspect. And so the tools that come up as major barriers and I think that's something we really need to talk about in our diversity working group and bring it to the board and ask, can we get more online training on all of these different tools? Can we make it more obvious where to find documentation if you don't know how to use Garrett or IRC? How to get started? I think it's all of it is there, but you have to really search. I think we also talked about recognition for non-developer tasks. And I think that needs something, there needs to be more recognition for maybe doing online videos and tutorials. I think the more content that is out there that draws people in, the better it is, and the community needs to recognize that. And I think the second we start doing that, people are going to put effort into it. I guess the other question I had was, what do you guys reckon we should do to get more participants? People keep them here, because people join all well and good, but then we've seen a tapering off effect. Do you have any concrete ideas for us? I mean, what do you reckon? Hi, so I don't know if this necessarily will bring in a diverse type of people, but it will certainly attract people in general. And this is something that I saw in the Mozilla group. Not too long ago, I saw a website called WhatCanIDoForMozilla.org, I believe, and it basically just was a questionnaire. It was like, what do you think is interesting? It doesn't necessarily take you down a particular path until you drill down into, oh, I'm a technical writer. And then it says, well, are you, do you like things about compute? Are you a networking person? And then at the end of them asking you those questions, they say, well, here are like five projects that you can get involved in, and here are some people that you can reach out to on those projects to get you started. And so having something like that where someone just green that, you know, they've never been in the organization before, but they've heard all about OpenStack and they want to get involved and they're not exactly sure where to start. Having something like that certainly drew me into even, you know, just using the website like two or three times to see, okay, I'm just going to answer the question differently just to see what projects it's going to, right? So there's a level of sort of engagement that, you know, that one little website was able to do, and I ended up like, okay, well, this does sound a little interesting and I actually did get involved for a while. So having something like that that's just, you know, sort of like the front door that's very easy for folks to get into as opposed to sort of overwhelming you with, you know, wiki pages and all this kind of stuff, something very easy that someone can, you know, very quickly get in. And not only that, but once I got into the project that I was working on a couple years back, as soon as I got there they noticed that I was a new person. They introduced themselves to me and they said, well, you know, how'd you end up here? And then they immediately kind of engaged me in conversation and we began talking about what they were doing. So that was sort of a soft introduction into working with that group. So that made me something that we could maybe adapt. Right, that sounds actually really cool. I think we have a wiki page where it talks about how you can get involved. I don't know how easy it is to find or whether it's hard to find for new beginners. But I think I've seen it. I don't think I'm making this up in my mind. And I know we have a few... But this is one topic where, unlike the IRC, I think it would probably be appropriate to ditch the wiki. Yes. I think that makes a lot of sense. Thanks for the idea, by the way. That was good, yeah. I know we have a few marketing people from the foundation. So maybe they know better about this. Something more appropriate than a wiki for the... Yeah. Choose your own adventure kind of a thing. There's actually a session happening right now that's about documenting kind of the culture and the practices of the community. It's in the Design Summit building. And one of the things that I think kind of needs to meet there is about managing expectations around contribution. So a lot of people are not used to code review in their jobs. A lot of people are not used to participating necessarily in a community environment where they're not going to land the patch the first seven or eight times. And helping them set those expectations for themselves so that they can not think poorly of themselves when that happens, because it will, is really crucial. I think maybe circling in with those folks who are writing up a lot of that documentation would be great. That sounds like a really good session. I wish the design sessions were recorded. Yeah. That's a good point. You're not? No. I'm afraid to go down there. So what I've seen a lot in the Outreach ERC channel, which is the internship program for underrepresented people to get started on OpenStack is that they don't read the Wiki. Even though we always point them to the Wiki, they just don't want to read. They want to contribute. They don't want to read like a book on how to contribute. So, yeah, it would be nice if we had alternatives to a wall of text on the Wiki. I think, yeah, I think at some point people will have to read. But yes, it's hard to make people read a lot and just stick to it. People have converted every book possible into audio books now. People will go to great lengths to avoid reading. Yeah, but I think her point that nobody wants to listen to two hours just to learn how to contribute. Like, what's the short version? Am I understanding your question? Hey, what's the short version? Give me the one quarter of a page that allows me to get started. Yeah. And I think that's probably a broader issue, you know, that the OpenStack.org slash start is probably where it's more appropriate to take a look at that. But that is, I haven't looked at it to be honest with you. Like, how do you get started with contributions? Because there's no chance that I'm going to be contributing any time soon. But you do contribute. See, you are contributing right now. Yeah, not in the way that... Not in the... But we should, somebody should look at it. Probably somebody like me who's absolutely not going to contribute or have not contributed in the past. So to look at it to see how hard it would be. That would be something interesting to do. Yeah, so I know, I'm trying to remember which summit it was, but myself, it just said there was no one else. We actually did a quick walkthrough on how to start with contributing and I'm sure it's recorded on YouTube somewhere. It was OpenStack Summit session. I think there was a slide deck that went out as well. With the steps of how to get a first patch in or something like that. Yeah, and I'm sure we were not the only ones doing it. So yeah, there's ways to get a quick overview. There's YouTube videos now. If you're in China, I think you're out of luck. But yes, there's definitely ways and yes? Yeah. Most people are Windows users so it can make things difficult for them to submit a patch along those lines, unfortunately. Right, and that's one of those things where if you are not working on Windows you don't even think about it. Guilty. I'm guilty of the same and I haven't used Windows in forever and I wouldn't even know how to help someone. We can fix that. Yeah. So, and I know for big corporations that's the default environment is the Windows machine and I honestly don't know what people do when they have to work in a Linux environment but they're issued with Windows. I assume they're either dual boot or they run a VM. Your own virtual box is what I do on my Windows system so that I can go ahead and get the tools and then I get myself away from the firewall so I can actually get things to work. Right, which is a lot of extra work. Yeah. Limits what you can do and when you can do it. Right. So, what else do you think we should do or know? I think we have only five minutes left. Does anyone know? Can I just ask, did anybody come into this session not knowing about the diversity working group? All right. Is everybody on the Diversity IRC channel? All right. Some head nods, some hand raises. For if you are using IRC please join us on OpenStack Dash Diversity Channel. We use foundation mailing list for all of our communications. It's not a very busy list so you won't be overwhelmed. It's not like the main list so please sign up and participate and send us your ideas. If you don't want to send it to the mailing list send it to one of us and or reach out to us. I am on IRC pretty much all the time. I'm also on email pretty much all the time. Yeah. And I guess I mean don't hold back, right? I mean that's we want ideas, we want because we're trying to do some we're trying to change the way our community is represented and it's not going to be an easy task and we need as much support as we can possibly get. We need all of you, we need the community. You know if you think this is worthwhile and tell your coworkers, your friends to get involved, participate if you have specific challenges that you think OpenStack, Community or Foundation are really terrible at let us know either publicly or privately because if we don't know it's a problem we can't help fix it. There was one last point that I wanted to make and I think someone here said it before which is if there are some simple projects and low hanging fruit that people can start with documentation being one, reporting a bug or doing a bug fix or something simple that each group can provide then that is a stepping stone towards more participation. Yeah and I think when for those of you who don't know and you want to start contributing each project after the bug is submitted you can go through triage and identify as trivial, medium hard the level of effort it would take and for beginners it's recommended they start with trivial and maybe someone else can talk to you whether that's working or not. Any OpenStack developers here? Alright that's another strategy, you submit a bug and then you fix it and create the bug. Thank you everyone for joining us.