 Hi everyone who came to my core conversation on community stuff, so what are we here to talk about today? Yes, there is We're very casual in the core conversation area So I wanted to talk today about Stuff that goes on the community related to the size of the community and the growth that we've been undergoing the mentoring system Processes all that related. I think that's something that we really should Analyze much more than we currently do because fundamentally Drupal has grown so much and will continue hopefully to keep growing But there's some signs. I think of some unhealthy things that might be going on And we should really take a look at that on a regular basis year by year and see if there's ways that we can improve and you know Things that we care about should we care about them as much as we do? Should we be caring about other things? You know, are these some of these things good or they bad or whatever? So a little bit about me, I'm David Hernandez I work for FFW the manager of learning and contributions there So my role specifically is All about internal learning education training mentoring I'm also community liaison Ideal with contributions everything. So this is all very relevant to me and something that I'm actually Looking at more and more as time goes on because it's directly related to my role of the company Again, I work for FFW which is the world's largest Drupal agency Go team and again, this is a core conversation So I should not be the only one talking I want to hear from everyone I'm going to go over a lot of different things, but this content is not only about my presentation You all have to participate So anything that you want to complain about go for it If you have ideas if you want to interrupt me while in the middle of slides go for it use the mic shout it out All of those are fine. I definitely need you to participate Because I don't have enough slides to go for an hour So audience participation is mandatory All right, so three thousand strong This is the number that obviously I put in my description and something I want to talk about because it gets Communicated a lot as part of the marketing for Drupal of hey Drupal 8 as we've gone through development We've had over 3,000 contributors We even made a really big deal about this when we got to that number I remember exactly when it happened during GovCon last year and it was like who is going to be the 3,000th person and it was a really big deal and we even gone past that so I think by release that number was up to 3,300 and it's like 3,500 now And that's gigantic Drupal is one of the largest open-source communities if you go by people participate people contribute And if you actually just go by that number it may in fact be the largest one Just by number of people who actively contributing so We're allowed to interrupt. Yes, we have to be careful when we say we had 3,000 contributors We had 3,000 people with commit credit. Yes We had a lot more contributors than that. Yes, officially credited To an issue that got committed to Drupal 8 As opposed to someone who might have just say edited documentation on Drupal dot org or ran you heard me Or ran an event spoke at Drupal con like I'm doing now Or just speaks at Drupal con Just gives a key note So let's talk about that number a little bit and how we got here So there's a lot of things that we actively do which is really great and things that we're doing now and throughout the Entire development process for Drupal 8 that we haven't done before one of them is having this actual official mentoring team and mentoring initiatives So if you're not familiar we have these weekly office hours and IRC where people come by you can get Mentored a lot of this happens at events. So this Friday There is mentored sprints that are going on and we always get lots of people that show up I think technically we might even get a couple hundred people that show up for first-time contributors at every Drupal con I know that at every camp that I'm at I help mentor and at the sprint rooms We're always seeking out new people to get them to participate and it's a really big deal You get your first can contribute credits your commit credits and we do a lot of that Which we've doing more than I think we've ever done before. I think there is some cultural change. So a lot of people who Even the same individuals who've been contributing in the past who you can tell I think in the issue queues that they've been much more Welcoming to new contributors than they were like five six years ago Because people are really been understanding the idea that hey, these are new people who are coming into the community They want to actively help It's a good idea if you like be nice to them and right like not yell at them and act like a complete jerk to them I'm like, oh, yeah, then they come back and they actually help Yes, and there are plenty of open source communities who are really bad at it That I've even participated in actually had to get new arguments of people Like why are you being a jerk to that person and like they argue like it's effective being a jerk like No, it's not Like this is how we weed out people who can't hack it. Okay Good for you And greater communication. So yeah, we definitely promote a lot more So I stole some charts. I mean well, I made one chart and I stole a chart If you've never seen this before this is the famous Drupal long tail that we you know people have heard of If you've never heard of before a long tail is this graph where you can see obviously this is Ranking of contributors and the number of commit credits they have so You can see it's significantly the tail extends out for people who only have like one or two actual commit credits And most of the work is actually just being done by a small number of people So I made this one so I must have just completely copied everything that you did I'm not saying that but You'll see in another slide that I completely stole that one accredited XJM So it's too lazy to redo it for triple eight and you'll notice that hey, whoa, look at that It's like the exact same thing. No matter, you know, we did that in triple seven and in triple eight We still have this long tail where we have so many people who are just Mostly doing one or two thing. It's even longer Yes, we have a smaller Well, they're not both to scale you They are they they are not So the numbers aren't the same and either acts the numbers are longer on both axes but at the same time also the area of like regular Repeat the work done by repeat consistent contributors is a smaller part of the whole compared to the long tail Yeah, but some of that is also I think because there were so many more Commits on eight. I think that might be part of what's demonstrating that But I have some other numbers to talk about So I just wanted to show you these two things just to Demonstrate the fact that it hasn't really changed that much But I did look at the numbers for both use triple seven and triple eight and there were some interesting Not just the numbers, but if you actually look through that list of people, there's some really interesting things that you can take from it One thing of so here the numbers so over 3,000 over 3,300 now for triple eight and it was exactly 954 people for triple seven and one interesting thing I noticed that both of those Lines start to flatten out around the 200th person And so I started looking at all the individual people because one of the things we've always discussed it is a burnout and turnover of the people who are contributing so I wanted to see what that was actually like and I started looking through the list of say the top 10 20 30 50 things like that and I thought it was actually kind of interesting that It didn't seem like the turnover was as big as people think it is So for example of the top 25 people 10 of them for triple eight are still in the top 25 from triple seven and other remaining 15 eight are still in top 100 and 12 are still in the top 200 There were three people who were in the top 25 for triple seven who were not in the top 200 for triple eight, and I think all three of them are still like in the top like 400 or something To a degree yes, there certainly is that but I would I would encourage anybody who's really interested to actually go look at those lists and Even if you scan the top hundred it's still extremely familiar groups So an example clarification I think what Kathy is getting at is that son for example was extremely active in triple seven extremely active in the first half like the first two years of Drupalite development Fell off the face of the earth as far as Drupal's concerned like he basically left Drupal He continues to get issue credits to the state In in eight one eight two, etc. Because he commented on issues five six years ago so there's there's an extent to which like the Contributors aren't actually the people who are actively working on issues during that release cycle. It's it's like that Because of the way we credit issues and because of the length of time it resolves problems There was a period of time where I wasn't contributing the perpetuals at all And I continued to get more and more issue credits and that happens for a lot of people So there's that there's to some extent that those numbers sure I mean we're gonna have variation when a project takes five years to complete And that's a problem in itself if you want to accurately understand who was doing what over what time frame let's not take Five years to have a release but the the bigger problem that I think in even discussing that three thousand number is the fact that Yes, 70% of the people on Drupal eight had only one or two commit credits Which is actually a greater number than even for triple seven So let's discuss some challenges that I think exist on problems that currently exist I know that it's it's been discussed a lot with people about Drupal that work itself and the patch process and the tool set That's there That it's a major barrier for people to contribute Although I think to some extent it's not as significant as some people think as far as getting people to contribute very Actively, I don't think that's actually the number one bearer. I think it's a significant barrier to get people who are less technical involved So if you want people working on documentation UX changes and simple like CSS changes, that's yet some major barrier for somebody who's not a serious developer I don't think that's necessarily the barrier for somebody who's already a high-level contributor Continuing to be a high-level contributor when they most of their time is not spent dealing with the issue queue and patches and things like that You know that the thing that's stopping you from making that 500 kilobyte patch was not the fact that you had to make a patch and upload it It was like it was the fact that you had to write the code to do it And the same thing issue to workflow I'm really state stating here more of a project management related issue of like how How we deal with the issue queues themselves? Statuses things like that. That's something that's completely outside the tool set I've talked to a lot of new contributors who still have confusions about that and don't Understand not necessarily just what the statuses might say say in documentation like you said it to this but I talk to people all the time say I don't understand why a person said it to that and What it is that they're actually saying to me like is it getting sent back to me like they think just says Uploading a patch and then someone reviews it sends it back to me to work They're like, oh are they saying they're not going to accept the patch like no They're just sending it back to you that you might have to do something or there might need to be a discussion or something. I Think we need to have a like we need to talk about this status I think that might be helpful because it's not just needs work or needs review. It's like This needs more planning or justification. You know we have some some statuses for that but There's definitely it seems like people who are not culturally in the contribution community themselves We're new get confused around a lot of that sort of process and workflow things that are just understood by other people I was having conversations with people about that directly yesterday They were just like we're off issues and they were just like well because they set it back to these work Or they did something else to it like well So like yeah go talk to them have a conversation like oh, I can't do that I Think governance we have some new governance policies around contributing directly to course specifically about being able to escalate Conflicts between people and their ideas and like up to component maintainers and the core commit team stuff like that I don't know how well it's actually actively being used. I wish there was a better mechanism for doing it other than setting tags You know like how do we actually get someone's attention and especially for new contributors like if someone's Posting something and someone rejects it like how does this new person know what the process is to get someone's attention to get a second opinion? Essentially, like there's nothing so those issues just died. They don't come back to it So that's a bit of a problem that ties into conflict resolution How are people dealing with these basically about all volunteers and many of them have no idea who the other people are Right if you're a completely new person and someone shows up and rejects Something that you're doing or says it's not a good idea. You don't even know who these experienced contributors are And then we have a human component, which I think is actually one of the most important pieces to this Yeah, for a lot of people there are language barriers, right? So how do they contribute when maybe English isn't their first language? And there's cultural barriers so even people who speak English we you know You might have to deal with issues where they're from another country and there's just all kinds of different cultures people come from different places They're used to working with people in different ways And you know you're you have an entire amalgamation of all these different people for coming from different places that are Volunteering giving their time trying to work with each other doing it Online doing it in IRC doing an issue queues doing all over the place and it's you know, it's not always an easy process And of course time management which you know, how do you? How can you devote time most of this is volunteering so? It's very difficult for people to get enough time to actually do their contributions And I think that's part of what leads to burnout and some of these other problems that we have where people You know are being asked to devote so much to themselves outside the normal work hours if they're not being paid to do it And you know that's becomes really difficult over an extended period of time And that leads to burnout and a lot of these other problems And growing pains I talked to people about this I think that's something that dribble definitely experiences that we don't talk about enough Especially with things like tools and processes Or a triple start out as being a very small community You know you you might have ten people in a room who are all working on something and it becomes very profound And it got contributed to dribble and this was back when it was like triple one and two and three and four But then you reached a point where you're not dealing with ten people in a room Anywhere you start dealing with dozens of people and then hundreds of people and thousands of people and right now We have thousands of contributors the processes and the way things were done and even the culture around how things were done When there are ten people in a room don't work And I see people try to force those same standards and those same processes on this much larger group and it just doesn't work that way And you know some people might say things like oh well, you know the way it used to be and I just missed that The way it used to be and it's like well when you're at a point where you've got 5,000 people trying to work on a project It's not going to be the way it was you have to change you have to adapt to that size And our tools have to adapt to that size our processes have to adapt to that size And of course that all increases expectations like yeah, we have we're much bigger projects We're dealing with corporate interests now We have bigger agencies you have a entire triple agencies who was like their entire livelihood is devoted and dependent on this project And it's success and giant corporations that are part of it And that just raises the expectations for everyone involved and that increases the pressure on everyone who's trying to contribute and trying to be part of this So what can we do? What are some proposals? I'd like to hear some ideas from people. I have some ideas of what I think is really important and what needs to be done Where my focus is, but I'm really interested in what everybody else has to say Personally, I think We need to focus on the human component not on just the tools and things like the numbers and the growth I think that Our biggest problem, especially when it comes to burnouts and some of these other issues like getting a release done is that we don't have enough People that are at that higher level of understanding and technical skill You know, we had so much difficulty after beta getting released just because we're burning through criticals But certain criticals there were only like so many people in the world who literally knew how to do that thing So everyone else is like sitting around like hey, I love contributing to Drupal. It's great It's awesome. They're like, oh, we can't get a release. So this guy critically gets done I'm like, well, there's only two people who know how to do that thing So we're all just gonna sit around and wait till they do it and find the time to do it because that's just the way It is I think if we can take that middle group of people Who we know are already committed and interested and self motivated to contribute and can help mentor them to be That upper echelon of skill level and expand that top group So it's not these five people who know how to handle those criticals We have 50 people who can do it instead I think we're much better off doing that than focusing on trying to get another thousand people Involved when those thousand people are only going to commit one One issue and then you're never going to hear from them or see from them again And I think a good way to do this is with the local events smaller events It's been my experience that the small sprints and small events are actually much more productive per person I think than the large ones and often require less effort to actually hold them So little sprints where we've had like five 10 15 people are always just really concentrated and really productive And they're much better learning experiences that's showing up at an event with like 500 people And half those people are trying to mentor half those people don't get anything done Everybody's just talking with their friends and then by the end of the day you might actually get more work done But per person you don't actually get more work done So if we can put more effort into that maybe more sponsorship for those local events More support around them and mentorship for them I think that would be great and just more things in general that are local. I'm really big on local events and local resources. I think that solves some of the human element when it comes to Language barriers and communications problems. I think it's you're much better off having mentors that are regional for those people They're much better at mentoring them than having random strangers that you don't know You've never heard of who are in another time zone in another part of the world Who was just trying to tell you what to do and there's so much effort There's so much overhead and efficiency that goes into just trying to manage those relationships I think if people can concentrate on keeping that local I think it's much more efficient and much more beneficial to those people You're gonna say something You're just gonna say thanks Oh So i'm listening and to So my my challenging question that I have is Looking at that number that long tail and looking at the efforts we put into like mentoring resources and these events and stuff like that If we really look at it and ask ourselves honest questions Is some of it actually worth the effort for the for those people and for the mentoring programs? And i'm not suggesting stopping I'm just suggesting maybe there's different ways to do it Um, I'm a proponent of doing one-on-one mentoring I think that would be much more beneficial for actually growing that middle group of people If we can develop a one-on-one mentoring program. That's more long term and has greater commitment from the mentors And My suggestion is maybe we don't stop things like the giant friday sprints But maybe we also don't try to kill ourselves doing it Um, so I I'd like to do a couple of things. I'd like to um provide some Context both historical and current I I'd like to propose rhetorical Possible partial counterpoints to your how about we go one by one And no, but I these things are all related and also um, like Uh, talk a little bit about the things that we're already doing now that maybe That we're doing on a smaller scale now that maybe aren't reflected in the in in your experience as a mentor to this point Or actually to some extent are reflected in your experience because you were part of sprints like the new jersey sprint and so forth over the Drupal development cycle Uh, is that okay? So like this framing background information here. Did I did I introduce myself this time? I didn't I'm xjm I'm a Drupal 8 release manager now Um, and so I kind of accidentally started the core mentoring program sort of Um, the what Kathy and I were chatting about at the beginning of the session is like You stated that core mentoring was an official thing and we're like, is it an official thing? What's official? How did official come to the maintainers? The the aspects about it are that our officials we did put it in maintainers dot text a year or two ago And we do now have Drupal association support for the mentoring sprints that we run on Friday So those two aspects of it. I guess you kind of could call official um But so when when we originally started the program the intent wasn't We're going to get 500 600 700 people committed contributing to core on this friday sprint It was Let's have a sprint that people can join and The demand turned out to be much higher than we anticipated. So every subsequent event It like the first time we were like, oh, maybe we'll get like You know 40 people at this and then we had, you know an entire room full And then the next event it was like, whoa, we need to plan for that entire room full and it was more So it wasn't so much that the intent was to have one person mentoring five to 10 people It was that the demand for mentoring at the events Outpaced what what we were able to deliver and so we had to change what we were doing In order to scale it So I wouldn't necessarily say that that's like a design thing of the of the core mentoring initiative No, I wouldn't say it is I certainly understand the demand is there But you also have to look at the results that you get from that So the question is the question is is Like there's two separate goals is our goal to help other help people contribute to Drupal or is our goal to get stuff done And it's less efficient for getting stuff done Is it a better or worse experience for the individual contributors? Is it a do we Like is the is the experience that we give people in the mentoring sprints on friday Like sufficiently good enough to make for the lack of quality on like a tenth of the people Right, so if we could help we could help one tenth of the people really really well And they would have a great experience and love contributing to Drupal With the big friday sprints we had this hit or miss thing But we provide that opportunity to a lot more people so it's not even about getting the work done in Drupal No, it's not about getting the work done even specifically on that day, but To that point if 70 of the people involved did only do one or two things Are you even getting that result? Are you introducing anybody to the experience and it's up to Any real benefit to them because if they don't come back is that a success story or not So so what um one of the things that was always like part of what we Um just accepted Because Drupal is a is an open source project the duocracy. It's an opt-in you choose to do thing and in At least for the initial experience for a lot of people The philosophy I was like this this 110 one rule where you um you Give an opportunity to a hundred people to become exposed to something become engaged with it And they try it out and then of those 100 people maybe only actually 10 come back and And you know participate a second time maybe a third time But then out of those 100 people there's one person who it turns out is going to be super super amazing That that one time in office hours that Kathy's phase showed up in a holy crap, right? Like it turns out to be something or scott kotzer Like years later this this little accidental opportunity provided someone turns into something where where it's not it's something you do for like It's something that you do For everyone on the chance that that one person is going to benefit from it and then that that in turn like Is what provides the future of the project so that's like that's the question I want to ask is like Is it is it a bad is it a bad thing or not? Is it a bad thing? But Is it is it is it not not working? And do you have enough examples of people? Well, and see that's the question because it's so it's so subjective right like like when you when you look at the long tail graph I I was I was a little bit confused by your data there because it's like of the top A significant number of the top Contributors to triple a I thought were people who were completely New to the Drupal 8 cycle like a lot of the top top people were not heavily involved in triple seven core And so I like I I'm confused by the data at least in my impression of it was that There there are a lot of new people my my point in that was Really looking at how many of the people who are involved in triple seven were dropping off Not just how many people were new they're obviously I mean when you're more than three times as many Contributors as you had before obviously there are a lot of new people But no I'm saying of of like the top people right like is it actually a concern that people who were involved in triple seven Aren't involved in dribble. I don't think it necessarily is like I don't either That's that's why I think we always have these discussions about people who drop off and Burnout and things like that. And I don't think it's nest I burn out is a problem We don't want burnout because that's an abuse of your volunteers I don't think people leaving is a problem because You're talking about five years later or six years later if somebody wants to do something else They somebody wants to do something else tomorrow It shouldn't be seen as a negative I mean if you even if you have massive amounts of your community who are disappearing, you know for specific Reasons, you know, how are they being treated or you know, it's an indication of something really bad that's going on But I talked to people who say the number one problem in the triple community is People disappearing. I don't think it's the number one problem at all So that I wanted to the other so that the parts two and three there I wanted to talk about the the things that are currently changing and then I also have Just as a teaser we're doing a core conversation on thursday with threes that has the other part of the survey results Part of which are barriers to contribution and I have some data here that's I'm not supposed to share yet But I'm going to talk about it in any way at least potentially if you guys are interested These are all like a number of them are things that were up as bullet points on your slides as to reasons like what are the things that are Contributions like maybe I'll come back to that when we talk about it more since I've been talking a whole bunch here Um, but so there's two things that are going on right now currently that I think Address some parts of the the problems that you raised and I just want to make sure that everyone's aware of them One is that um with the with the changes to the release cycle this this extremely You know having six month minor releases. We're never going to have a year-long beta again where there's complicated things You can't understand there's everything is blocked on a handful of issues that most people don't know how to solve And you it's like why should I participate because I like that we've changed the release process So that does not ever happen anymore because when we like instead of We open the development branch For the next release as soon as the current release goes into beta So there's no actual freeze. You're just work just goes to the next branch. You can continue progressing You get still get feedback on your issues. You still get to commit as soon as your patch is ready You still get reviewers. There's just as much value in working on it Regardless of whether or not you happen to be in that beta phase. So that's something that hopefully is is It's something that was a consequence of the durable eight cycle that is hopefully never going to happen again Because it was in awful. It was awful for contributors. It was very difficult for the project and so forth So i'm hoping that that that kind of thing is a mostly solved problem already We will have new problems in its place. And then the other thing that we've started doing just recently is So We've started in terms of like assessing You know what in terms of like making sure that issues have like the next step in place that actually is What's relevant to the the component maintainers for that issue or the people who know the right thing Actually core mentoring started out. It was supposed to be the beginning just issue triage That was all that was all it initially was and it turned out there were there were a lot more needs people hand from it Then just like oh get these issues updated But what we started doing now is we're doing Regular major triage with the subsystem maintainers for just the major issues in each of the different components So at least the the things that someone has filed is I think this is a really major bug And it's broken and it's growing up my site Get a get a response. So how often are you doing that triage? I have we are have been we have had a triage session almost every two weeks since the beginning of the year Now the scope is obviously enormous because there are 1200 major issues Um, so what we're doing the other side of that Advertisement is on friday We're going to try to do as part of the new contributor sprint have an area that is specifically about The first part of major triage, which is verifying whether or not an issue still exists Which is a great learning opportunity for someone as a new contributor because instead of having to understand All of these complicated release policies all they need to do it is learn basic get skills about okay Can I follow the steps to reproduce this issue if I can prove that it still exists at a point in the past? Um, and it doesn't exist now Why did that happen and so that they it's it's about learning to navigate the issue to and at the same time Filtering out some of the noise in the major issue queue so that the issues that are relevant to people are more easy to find So those are like two things that are going on right now But I think are kind of aligned with with fixing the the experience for like the the huge Both for issues and for contributors the huge Waste land of of there being too many things And I can talk about the survey results, but someone else should ask This is chris so, um helping with I mean I think that The goal of the mentoring program has to be completely to support your ball, you know, it's In my view, it's it's nice if people have a good time at these things and maybe enhances their Visited to triple con. I mean the live the live sprints, but If they're not if they're not serving the purpose of forwarding the project then it has the purpose I think they do so And if and along the same lines if you mentioned the 110 one concept if If the only goal the mentoring program is to find unicorns there might be a more effective way of doing that You know there might be because we all of us spend a lot of time on this and it may be There's you know, there's easier ways of finding those people and if we don't look at that sort of scientifically Then this is like a cargo cult of mentoring cargo cult Cargo cult of mentoring google it cargo cult Yeah, it's we just keep going through motions hoping that they have the effect that they might not be So yeah, and that's and that's I'm just sort of echoing the second part what just said was that No, I mean in the event that we have a one of these big sprints and people can be immediately useful towards the project Well, that helps right, you know, if we can get a handful people to go through a bunch of these issues and triage them They just did a bunch of really positive work, you know, for example, it's not that hard and you know But I think that whatever we do we have to At least somehow validate whether these things are really working for a group Or not and if they're not why are we doing that? That's all And you know that because Scientific method So there's a couple of things and one is I think We had when we look at the resources that are devoted to the Mentoring program. Yeah, it puts in some money to make Fridays at cons happen, but they don't put Resources into other parts of the mentoring program. So they're done because people Get their mentoring sponsored through their work or they're done because they do it as a volunteer And they come from other places and so I don't think we have a responsibility to Evaluate whether or not the mentoring program benefits the project I think it doesn't matter Whether or not it benefits the project as long as it benefits the people involved So I have to disagree on the purpose but agree with the Possibility that it does both so it may not matter that we argue about whether or not we're right about the purpose I think Well, just to that point I just want to say one thing though is that When I when I think about it what always concerns me and why I brought this up is I think we do have an obligation though to make sure that If those people are devoting their resources their time their energy that We are certain that they're not being abused in any way And that if they are giving too much of themselves, which people will tend to do very easily We need to be certain that there's even a purpose to what they're doing Right, but I agree with that So we need to make sure people aren't injuring themselves or other people But I and that there should be a purpose behind it But I think it would be okay if the purpose was it doesn't hurt the project but it really helps enrich other people's lives because I think Contributing one time does in fact carry a long lasting Effect on people into their work environments and and other things that they do that may not affect the project So I agree we shouldn't Invest a lot of resources in something that's hurting people I Want to caution Referring to contributors as volunteering all the time Um because I I think we need to look at the percentage of the different areas of the tail the top contributors The ones who have a lot of commit credits are the ones that have just one commit credit and look at whether or not They were paid for their contribution because I if we Think about different theories about how people can get paid. Some people get paid to come to the con And so if they sprint on friday, they may have actually been paid and they may not have been volunteering and we get the same thing with um Repeat contributors. I think we need to be careful about not always assuming that they're volunteering their time uh, and we Are trying to collect data on that, but it's not publicly available in an easy way And I think the other part of that is when we look at how And especially for things that are not just commit credit, right? um When we look at why people might contribute once and then not again I think we need to actually just ask them So it would be nice It would be nice to know what the results of that survey question is that in that let's just I thought you just had the the drees survey of That's asking people about that was I did I don't remember questions about contributed Maybe I'm thinking of a different survey you're here talking so you're a contributor now So I was wondering uh Because you sound like you were looking for ideas. So this is sort of outside the box. Um, I didn't know like I'm signed up on dribble I don't even know what it is.org and I was wondering if there's a survey that you could give somebody that could that we could fill out to say where we might be a best fit And you know whether it could say, you know, like Maybe not directly like do you want to do this thing? But like that you could ask a question that you could gauge what we would be useful for So you kind of know and that would be like in a pocket for when you need people It would be maybe something that's associated with our profile and maybe this already exists. I don't know and Neil was any of that in the content re-architecture stuff About like because I know is that whole idea of like you could say you're more of a UX person and get sort of like tailored content to you It's So I was wondering website personalization Middle school, they would have a school out of survey and then they would say, oh, you should be a nurse because you show these qualities So I don't know if there's a way for you to make it generic enough to say they would be really good at testing You know, like I do I am a contributor in my site But I know my business process really well So I'm very good at giving them feedback because I can kind of see where it doesn't fit into the square peg So so is there a way in our profile that we could be poked back to come back and say hey You said you were interested and you only contributed once and maybe you'd want us to contribute again and then my last point would be um I don't know if this is possible, but I just registered like last week to come to this event So of course I paid a late fee, which you should but I don't know if you can say Well, we'll waive fees if you produce hours Or if you do that in advance because sometimes money motivates people who aren't Making money from selling software from their business or getting people to come to their company as a government employee But I don't know if it could be something where your hours Could be, you know come back. So I just suggest that as an idea If you can you mean actually paying people no no reducing fees based on The hours that you have done. Yeah Right, so if you've been doing core contribution this year you can get a free ticket to dribble down Yeah, obviously you have your expenses you have to cover, but I don't know if that might be So I'd like to answer like the your first question because it's much safer Um One of so there is something like this that's kind of a do-it-yourself version that sort of already exists We have the section of the dripple.org handbook called the contributor tasks handbook And so what it what it has is it has sections like Based on skills you may have like like um, I I speak english well or I speak a second language or or I I can do documentation or Like skill sets like that they not not topical areas, but like basic skills you have and then underneath that it has a can you can you Put could you put it up on the screen? the contributor task handbook um I think it's it's like Drupal dot org slash contributor dash tasks Um, and you can see under there is a list of tasks you can do each of those tasks has instructions And then it has a link to an issue queue that might have issues that have those tasks in it Obviously, it's it's really hard to like the problem is finding the actual issue that has that actual task you can do But at least you can read the instructions of it and say okay based on as as someone who's a non-english language speaker For example, if you click on non-english language speaker You can then see what the tasks are under that group go to that page And then this is like oh, this is something that I could do to contribute to the project The and the reason that we do this with like a documented task And then a link to a queue is that The the process of finding tasks for people to work on is actually one of the most Challenging things for the mentoring program in general has been finding ways because we there's you know, there's 5000 issues that are open There's all of these different initiatives and so forth So if if you're involved with like if you listen to jesus kino to be like wow content deployment I think I have great ideas about that. I'd love to help with that you can get involved with that team But if it's more like ah, I just kind of I don't really know where to start I I have this skill that I want to provide. It's much much harder to find Tasks that are related to that skill. So instead there's a lot of Um, we try to teach people how we find tasks ourselves. So this is like a starting point It's definitely not nearly as like Helpful as a survey. It's very do it yourself But that that resource is there and so it's maybe a starting point and maybe you maybe if you look at this You have you'll have ideas for us for how to make it into into that kind of an interaction If there's a way to take this and then do the thing that for you would be like, oh that would have made it So what I need what to do there? It does not it does not Well, I mean there's a lot of things we would have to give people when they say There's a lot of resources to try to do them. But I think one of the things that DA has been doing really good with lately is The the newsletter that goes out That reaches a lot of people who just have an account and it kind of pokes people Every once in a while and the quality and information in the newsletter I find high like I actually will read it to figure out what's going on And so perhaps we can um Put in a Like a get involved section into the newsletter, which just occasionally dribbles out like look There's this resource you may not have known about and so It allows us to poke people using the tools that we have So we don't have to ask the DA to build any new tools So there's always some advantage in that So I think when you asked if we have tried poking people and we haven't But there might be a way we could do something like a newsletter That's already going out to a lot of people that might provide some of that function No, Mike Both of you Mike I think automating all that It's a hard problem computers aren't good at like you might show up to sprint and not know that really like Rerolling patches or you know what that is So but yeah, there's certainly opportunities Shorten some of the paths, but the computers aren't gonna tell us what to do Is there any would there be any resistance to pushing that information to new users as opposed to them seeking it out So like what we discussed before like someone signing up and filling out a survey and saying these are my skills And then actually have jupy.org pushing that in I think we have an an issue for that as a proposal to to like as to Adapt the new user sign-up process on jupy.org so that it gives them more information about the next steps And then a related Outstanding task that hasn't been cared for in the same way for a long time Is um improving the contribute section of a handbook so that it's actually something we can point them at that Is going to be useful to them. So like in Well, I think also part of the problem is being able to Better organize the work that we have like when someone shows up and says I'm a UX expert I can help on that stuff like give me something to work on We do have an answer for that now Which is that the usability team has started their own website that that scrapes issue data from Drupal.org using guvars rocket ship thing So that like if if if your skill is usability We've had a very difficult time for the duration of the Drupal 8 cycle connecting Usability experts designers and project managers, especially with the areas that that they're good at to some extent Like front-end developers. We were able to connect to issues And we did a really great job with with people with documentation skill and back-end developers And a great job with people who are just willing to test anything And so there's like depending on what the skill is like you'd ever have a better or worse experience at these these sprints that we've been talking about Okay, so so i'm going to talk like i'm going to talk just briefly about some of the survey results that came up in conversation But come if you're interested in this come to Dries's core conversation on Thursday afternoon It's the last one before the closing session And hopefully we'll actually manage to get some of this data segmented and in into the slides because We haven't done that yet because we're all very busy people and busy doing other things like giving the Dries note, for example That's Dries and then others of us helping with content in it So the in the survey results one of the questions that was asked was about If so so what is your involvement with with Drupal? Let's see the exact wording was Sorry, I had it right here in front of me and then I scrolled away from it for some stupid reason Okay, so what which the following best describe how you relate to Drupal is the exact question There were about 3 000 people who answered the survey and big big disclaimer that there is extreme participation bias here Because people who take the survey are people who are already interested in Drupal And to a disproportionate extent people who've probably had a good experience with Drupal and even more Disproportionately people who've been paid to contribute or who are already like have a successful like volunteer career if you will Yes, so but so in those in those survey results of the respondents of 3 000 people 24 percent say they contributed to Drupal in a volunteer capacity and then among those 24 percent 6 percent of the total So and actually 164 people I think as of like a week ago so that they're actually paid to contribute to Drupal So which is a bigger number than I was expecting now That's not necessarily paid full time that can be paid in a part-time capacity through like 20 percent time It's not specific on the survey But we do we do actually have you know some information that An increasing percentage of people are contributing as paid contributors But it's still it's still also a minority of the groups by a significant fraction At least from the people who cared enough about Drupal to spend 15 minutes filling out this difficult survey And then going on there's also a question on here about there were two questions about What it what are your biggest frustrations with Drupal itself? And then another one was what are your biggest frustrations with contributing to Drupal? And the one about but obviously the frustrations with Drupal itself are for some people a stopping point but what I thought was interesting is of Only 24 percent of people Said they contributed to Drupal But almost all of them answered the question about frustrations with Drupal or like like something something like Four fifths answered the question about what their frustrations are so this what we can sort of infer here is that There are a lot of people who tried contributing once and went away Because they had a frustration with it and one I'd like to look more like the the Like who answered which things but what what we can tell right now is that the top response was 34 percent of people were frustrated with the complex issue Q process Which supports what you were saying earlier about like They mark your needs work and I don't know what I'm supposed to do next and the issue doesn't tell me right Like I I have no idea where to go from here There's all of these requirements the requirements around the beta phase are another example There were there was all of this issue Q process that was a necessary evil But resulted in it especially for newer contributors of that experience um the second highest result was that Bugs and patches don't get reviewed and fixed fast enough that was 32 percent of respondents were concerned about the the rate of feedback on the issues And then the third highest response was that The the size of the Drupal code base itself was too overwhelming and too complex to figure to figure out how to get involved There are issues on there about language or cultural barriers Which was a very like one of the lowest percent interest. It was actually the lowest percentage response, but then there's the Participation bias thing there is if you've had language or cultural barriers to participating That's probably also a reason that you the survey wasn't promoted to you, right? It was in English first of all and and so all all of those language and cultural barriers that helped you from Getting involved with Drupal are also barriers for all the things there But that that's so that's the data that we have so far it's it's I think that the safest thing to do is to start with the things that we know are Known problems for people who at least tried wants to contribute Because it's much more difficult to find the people that we haven't been able to reach it all yet And that's like an entirely separate problem So was the the whole patch workflow stuff? Was that one of the answers or is that roped into issue q workflow? Complex issue the the full text is complex issue q process and slow Census building was one possible response that had the highest number of respondents And the second was bugs and patches don't get refued and fixed fast enough um, so that that um, that doesn't Was there anything about like tools and technology and it doesn't distinguish between the the tools and the procedure Because for for a new contributor, those are actually kind of part of the same thing, right? There's this whole set of unknown things that you have to learn That are some of them are procedural and some of them are related to the tool itself But the fact of the matter is that procedure and tool are like two sides of the same thing Like our tools should explain what the procedure is that the tool should be what the procedure is And and our procedures on the other hand need to be Surfaced on our tools and not documented off somewhere else remotely So that's like a part of an internal part of the problem We did do a separate contribution barrier assessment Um, this was an exercise with a small group of people recently that I can also We we shared internally which is we went through and evaluated like a hundred different barriers to contribution and said Just is this from people who were dedicated driveless is this First of all, is this a problem that will take, uh Is is it a horrible problem? Is it a slight problem? Will it will it improve the experience a little bit or a lot and then compared that with Would this take a week to fix a month to fix a year to fix? I don't think we can ever fix it and and using those two things We came up with a short list of a couple of Both tool and process improvements that we can implement in the short term to try to Get the most benefit with the least amount of despair and frustration. Um, so I started there before The the difference is that they're being implemented right now Um, so so the the the top the the thing that we identified as the top priority from process perspective was this usability Problem that we were talking about it's too difficult to make Usability and design contributions too difficult to improve the user interface in the correct way through the core issue queue and so the solution that we're testing for that is this new usability team with its own separate site a new set of Of processes that hopefully are reflect more what designers needs are versus what a back end Developers framework managers needs are and then the one of the the sort of like we identified several top priorities in terms of the tools Things that had required changes to Drupal out art itself And then we actually shared that short list with the Drupal association and said, okay of these things We think these things could be doable in a short time. Which of these are hard or easy And like we were actually wrong about one for example Like one of the things that came up as something that is Could be implemented in in a short time scale It'll be really effective is just automatically checking if patches apply or not when their needs review And marking the needs work if they don't apply without a full test run Automatically so you don't have to come back after four weeks and find out the patch doesn't apply anymore And then as a second follow-up step to that is that you can then Automatically re-roll patches so that's no longer a requirement because that's something that you know Two-thirds of the time a computer can't actually do So there's so there's issues open for both of those things and so we are actually And maybe maybe we just need to communicate Communication Very recently from the past month That that we It was actually before the survey launched that we that we started doing these two things So that that's where the the like address the tools part of it and the process part of it in parallel because Fixing those two things has different requirements fixing process just requires us to come up with a better process That still meets the needs that the original process was created for But fixing the tools requires A lot of times design resources developer resources and so on so that that that's kind of like well Let's let's tackle both of those things at once so that they can all right Can you sum that up before we go any further? We're pretty much I did wasn't that done? Okay. Well, you kept talking Oh, no, I'm not. That's what I was saying. All right I want to hear them. Uh, this is peter willing and uh, just a couple things in terms of feedback more directly to david's thought about one-on-one Mentoring and something I haven't seen happen. I don't know could happen is collecting Contact information from people at especially the largest branch to maybe also as part of the people.org sign-up process And you know, I don't know what part of that is in big there Collecting, you know, where do you live and are you, you know, can we Share your contact information with people organizing events in your area Because I think that making that connection between people who have traveled to attention while already might not get reconnected with a more local group Is it seems to me like that could be a gap in kind of keeping those people engaged in a longer term And then another suggestion you sort of mentioned funding the money somewhere in there, but I think I haven't ever seen A push and I don't know who we come from exactly but it feels like Like it would be useful to start having a sort of conversation about If your company uses ruple, like maybe there's A sort of range of percentages of your revenue from ruple that you should be contributing back And you know, can we have people that are kind of stepping up and saying yes I am contributing this much back just for the association or sprints or You know fun people who are working for or like you have a range of ways that it could you know Contribute back money, but that you know, it is free software, but you know, you need you know money really to make these things Keep going Drum said yes, we can but we don't want it to be a multi-page thing And so we have to decide what's most important. He said yes, we can but josh said no You had a you had a point though. You wanted to ask a question the gentleman But just kept talking and I was trying to interrupt Dude excuse me You're welcome Track chair track chair