PyCon Community Backstage: A Decade of Camaraderie, Growth, and Lessons Learned

Speaker 1 [00:06]

Yeah, so I think this year, like usually I used to do tech talks, but I think this year I really had the feeling it's important to do a community talk and talk about the Python community and some things I have learned and some things I need to complain about and some things I want to motivate you to contribute. So, if you don't know me, I'm Alex, so I'm involved in multiple organisations. This organisation, the Python software is a board member, I'm also currently a director of the Europe Python Society. You will learn how this happened, this accident happened in a few minutes. So I obviously think you can see I enjoy also doing this, although this is not my work. Or not yet my full work yet. So I also launched a nonprofit earlier this year. It's called Pioneers Hub, and I'm going to tell you a little bit more about this and why the Pioneers has a P-I and not a P-Y as well. So I think it's always best to learn, hey, what was my journey into the community? How did this happen? And where did it all, like, start? And there were so many things. So, actually, like, 1920, 2010 plus, there was a lot of stuff happening. Only in my life I decided I want to work with different software. I need to evolve from the tools I was working with. So, I switched to Python as a programming language. I discovered data science and everything that came up. And this was also like a time where this huge explosion happened where you could learn data science for free on Coursera. I mean, I spent, I think, two years, I spent all my weekends on Coursera learning about cloud data science at a very early stage. And then, of course, once I learned about the tech and many things, I also, I don't know how. discovered PyCon DE in Cologne in 2013, which was also already a unique experience. And then somehow I discovered actually EuroPython 2014. So I was an attendee ten years ago in the same building, and I know Valerio, who is for the lightning talks, I was his session chair, and he was talking about algorithms on this stage. was his session chair just as far as now session chair for me and I got to know people and I got to know this very open community which is not obvious if you don't, if you're not a part of it it's not obvious how welcoming and easy it is to meet people to exchange knowledge, to learn to contribute as well and so I really enjoyed the conference and then there was a meeting of the EuroPython Society because at that year EuroPython was still run by local chapters. So, actually, PyCon.de, the same organization, was also running EuroPython that year. So, I joined the EuroPython Society just as an auditor of books. So, not much to do because nobody else was up for it. And, yeah, later on, I happened, like, half a year later, I happened to be a program chair of EuroPython and Bilbao, co-chairing with somebody I've never met. and I was organizing a conference with strangers over email and I could not really imagine how good this is. This cannot work, yeah? Like a lot of strangers working on emails on a 1,200-people conference at the really nice conference center in Bilbao. So it was like, it's a decent conference at a really big conference center. And I think why I'm telling you this, I think it's always like very important to remember. So also, like, how did I discover PyData? PyData was a satellite event on the weekend. On the Saturday and the Sunday, there was, like, three PyData tracks with talks. And so I stayed the whole weekend, so I attended PyData as well. So two conferences in a row. I met Leah, who just gave the keynote. And, yeah, people, we just talked about stuff and community. And so in total, like, this ended up in this. So, this is actually my community CV, if I may call it like that. So, I became part of Europython. I get really involved and dedicated to Europython. I also was a board member. Then Sebastian, who was also one of the organizers still of this conference, asked me at Europython, hey, what about organizing PyCon DE in Karlsruhe? And I live in Mannheim, close by. He said, oh, yeah, sure. So, this is how I ended up in this team. And then Valeria just announced me program chair of EuroSciPy without asking me. So this is how I got involved in more and more organizations. You see, I think I'm still enjoying it. But I also learned what community, what we need. Because the community also, it grows. The conferences get larger. We are really good in onboarding new people. The whole Python ecosystem is evolving. and we see there is a huge need also like the professional things in the back end because only if things run smoothly and are well organized and planned in the back, we are able to deliver something like this. For example, this is the closing session of EuroPython in Prague just last year. And you see EuroPython is very similar to this conference. It depends a lot on volunteer contributions and yeah, it looks, it's fun. As you see it like this. And since I couldn't wait for like the yearly conferences to happen, I also decided to launch PyDataSuitVest. And I'm really happy my co-organizer Uwe is also here in the audience. So PyDataSuitVest, yeah, I also like to foster like a local community because it's really important to share the knowledge, to connect with other people, and also take the breaks and just talk. And so, and all like this. and stuff like this happens. So, anyway, what I think is important for community ISA, the importance of community organising in the Python community is actually the success factor, to put it a little bit into words the rest of the world maybe understand as well, is success factor is a large and active welcoming community. So the culture is one of the building blocks of this society. And because we are very authentic, also like the program, we have a program committee, people apply to give the talks, we always have a mixture of newcomers providing new talks, expert talks on really top-notch topics, but what we don't do is we don't really sell shelf space, like sales conferences. We don't say, oh, you're a sponsor, we are a talk, whatever. Even we have sponsored talks, but all the sponsored talks also go through the program committee. And we don't have just like sales pitches and we don't want to steal your time. And I think we have enough exciting content to share. And so it's really important. So if you go here, you can really get authentic content and you can really learn something. And I also like to phrase it, the conferences are run by experts for other experts or footnote for newcomers, experts to be. And we have a really strong focus on the extended experience and what the attendees, and we are also attending other conferences, what we really need to get things done to make it a memorable, pleasant event, but also learn and share knowledge. So that's why we also have this very strong focus on diversity and inclusion. and it's a lot of work. Everybody has to work together to make this happen. So Python community, the prototype of Python community conference works. There's volunteer committees. They organize multiple parts of the program, like a program, a diversity program. I think everybody was in the opening. We introduced all the committees, you see. We distribute the work in different committees. How do the committees form? We have an open call, so of course we like to have people stay on because they know the processes, they know how things work, because running such a huge event and handling like 500 calls for proposals is a lot of work, and also we need automation, we need processes, we need guidance, how do we do things, but it's still an open call. Every year we make an open call, new people can join in, bring in their ideas, their perspectives, and this is what we do on multiple things. Program, diversity, communication, website, design, conference chairs. We have their different governance models, how the conference is navigated. And also, for example, the video recordings. The video recordings are very important because we share knowledge. we share here in this room we can still share it with the rest of the world after the conference and as I mentioned in the opening session I actually met people on the train from Africa that said oh yes I'm a machine learning engineer now in Germany and I watched talks from your conference and they taught me machine learning and now I have a really cool job and it was just like a coincidence so this was still the best and most emotional proof I ever had the thing oh we share and other people consume works. So, very important is we have some people who carry the torch, who actually organize things and take some responsibility. And, of course, there's a lot of workload also to be done and responsibility because we work in teams and committees. Also, we have cross-committee communication, which is sometimes, of course, committees need to coordinate things. For example, the program is ready, communication teams need to publish that, other people know that, coordinate with other monitors, monitor managed milestones, also have an eye on what if somebody just disappears or goes. That happens from time to time. But I think also what's not on slide, I think it's very important to have a healthy environment. So people really enjoy contributing, and I think it's very important to have a healthy environment to contribute. It's not too stressful. It's not always short on time. It's always not like, oh, when should we have this done? Oh, yesterday. This is very human, but, of course, the thing with volunteer organizations or workers, of course, it's volunteer. Oh, maybe next week. Oh, no, I'm too busy. Oh, something else at work came up. So it's probably delayed a little bit more often, but we also need people to monitor and watch and step in and just like friendly ping or sometimes we also need sometimes fallback plan. Somebody just steps in and takes over in case somebody just goes. I'm really happy. I don't think we had one ghosting this conference. So it's really great, but we have to prepare because, for example, like if you organize a conference, what's the big difference to software? Any ideas? Except from the delivery. It's really easy. Okay. We delay the software release for two weeks. The thing with an event like this, the door opens no matter what. So there is no delay. You cannot say, oh, can you please come three weeks later? We have some delays. So this is quite another challenge to keep things on time. And also, apart from the organizers, volunteers. And volunteers can help on specific paths. And I think for my experience, it's very important to onboard people, also to find new organizers. And I think my best advice is not try to find the person who solves your big problem. Rather, find a person who solves or takes over like a small task, which is well-defined, which is also footnote work. You need like a Google Doc or some description. This is your task. This is the expected time. This is the time frame when you have to deliver it. Because many have seen this not only in the Python community, also in other organizations, volunteer organizations. People are hesitant to commit if it's like, oh, write a blank check. But if you say, hey, this is this small work package, and are you able to handle it? They say, oh, yeah, I can handle this. I have the time. And people want to be reliable, so they rather pick up a smaller task. and sometimes if you reach out to people you can also motivate them to if you need to take more tasks and yes Alina is smiling thank you yeah well so you see it's also like culture, organization, project, management in a way you see we need also like structures it's not just like anarchy or magic happening so the more structure we give around and a nice setting the more we can accomplish together as well. It's not but still this is very open to new ideas and still going. So there's multiple organizations like Python Software Foundation, NumFocus Python Software Verband, legal entities running conferences, supporting communities, looking for sponsors because sometimes like a big sponsor signing up a big sponsor can be really a lot of work because there's a lot of paperwork to be done. We also of course need to take care of filing taxes, reimbursements, financial stuff, a lot of boring stuff. Also a lot of stuff which for my career actually, I know legal, I know tax and run a company but I learned, okay that's not so common. Data scientists sometimes say, yeah, so there's a big gap. People have never dealt with filing taxes on a business level or different tax vats across european countries and other tax laws so we need organizations who basically cover that because this is not really adding this is an operation we need to run stuff but it's not really adding value in in content or our educational mission so we just need things to run okay have a nice back office run things and Let us do the fun stuff to make the best of it. So the challenges on opportunities, of course, like very often we work in a decentralized nature. We don't have a common office. We work on Slack or Discord or many other platforms. And, of course, this is also like a challenge when you need a driver. Because sometimes somebody posts a message, Okay, people are probably not all the time on slack and the other organizer reads it the next week and then you see okay it's really slow and sometimes feet things just like fade out because Nobody's driving it. Sometimes we need you need like a structure to say, okay, this is a meetup We want to do this like in the next four weeks and and also send some reminders and very often It's not just like it's not like neglect. It's actually just yeah, you're just not on a channel You just don't see the message and of course, there's a lot of information every day, task every day, work and stuff like that. And also, very often, one of the obstacles is newcomers reinventing the wheel. So, I think it's very good to have structures that can give guidance and mentorship to say, yes, we've tried this. This was our experience. I think it's very important to say, yeah, we've tried this. This was our experience. But still, it's your call to make because you're in charge now. So, it's not like, you know, just like spreading wisdom and saying this will never work, as sometimes people did, which was very unhealthy. So, hey, if you want to try again, feel free, but take this into account, and good luck with that. And because we want to avoid mistakes and retrying things, and also, like, because this takes energy, and it's a redundant setup. So individual workload is a very important part as well because we have to balance individual workloads as well. Some people are more free. So I was always like self-employed or running my own company. So I'm the boss of my schedule. So if I want to do my work for my clients, if I want to do it during nighttime and I make them happy, it's fine. And I can attend stuff for the Python community during the daytime. But many people who are contributing have a day job. They can only attend at night. And sometimes also you have, like, yeah, you become parents. And then this nice spot between 6 and 8, yeah, it's occupied, oh, children want to go to bed and stuff. And we also have to account for that. So we have to distribute well. We have to also, like, manage ideas. So, of course, who has worked in volunteer organizations before? Yeah, great. Many. So, have you ever had the experience, your lack of people who want to join just to give you ideas? Yes. So, and people don't do this intentionally, but very often it's a struggle. You have like a bunch of people who are just like, oh, I have this idea. We should do that. We should that. And it's very often good ideas or ideas which are sometimes hard to argue against because it's for a good reason and stuff. But we always have to question ourselves, do we have the capacity? Can we really make this happen? And I think this needs also like strong leadership to do that. So it's almost like, do we have the capacity? Or just ask the question back, will you take over? Will you take the lead on this? Because we never lack of ideas. we lack of execution of ideas and making things better and so prioritizing is one of the very important parts but still what I also learned is if somebody comes up with an idea you probably don't get in the first time or you probably don't give it the priority or the weight I think it's sometimes also good to let it sink in because sometimes people come up with great ideas you won't see it in the first site. And of course, what we absolutely have to avoid is last-minute pileups, because people will probably solve it, but then we'll leave, because, say, okay, this was too much. We pulled through, but I'm out. So yeah, we basically covered ideas and expectations already. So ideas are important. Community is very dedicated, but we also have to navigate this. I just covered that. And of course, very often overlooked is a lot of work in getting funds, taxes, everything set up, like consistent processes, also like for volunteers, also like to empower them. It's not just like, yeah, do something. So you need to basically navigate people contributing, also like to make their life easier, because if you say this is your task and please do it, this is our process, it helps and relieves everybody. Of course, and running a huge conference like this, you also have a budget to navigate. It's a huge conference like this. It's also easy to mess up. So, yeah, stuff like that. So, I think we can skip that a little bit on time. And I think the skills I learned, which are most crucial, is many talents are welcome and many skills are needed. And I think we have a lot of technical expertise, but we also need to communicate to others, also like to people outside, who are not part of the community, people who have probably never heard of Python or open source, to tell them this is the mission, this is why we do it, and translate what we do for them so they can understand what we are all about. Because very often people would like to support this as well, but we have to get the message across. I learned, working with many committees, committees work well when you have also like a team lead. So from my experience, it's not just like everything, it's not like, I would say, like a consensus on everything. It's some people have to become team leaders and also like lead the team somewhere. That's super important. And you have to find people who are for it because from my experience, most people are not up for leadership roles. They would like to contribute, but are not so much up for leadership, although my experience also is sometimes it's also okay to push people sometimes a bit because sometimes they might not see that they are capable, but I think sometimes I do. I think they are capable of doing more things, so I'm also trying to hopefully friendly push a bit in a direction. Yeah. So committees, you've seen this in the morning and a little final thoughts and future perspective. The community has grown a lot and is still growing. We need to optimize further to help others thrive. So also like the culture you could see here in the Python community could enable other tech communities as well. Because I never heard of many other tech conferences. There are a few other communities very similar to me around, but usually tech conferences are not as friendly or as welcoming as here. They have other drivers. They never, for reasons unknown, they never were able to build this family of Pythonisters up which just carry this culture on from generation to generation. But what we need actually to help others and help others thrive is we need stability, reliability, mid and long-term strategy, and not just like the next event to run. And also resources to thrive. Because planning, if you can plan or if you have a budget, and I also know sometimes it's a tiny budget, 50 euros, 100 euros can really help to make many things happen to thrive. And so I think we need to optimize and handle growth also on a level of have a back office with administration, project management, local requirements, like for tax reason, legal. Also, I think we also have a real need on governance and documentation because we probably don't know, and I probably should not say this in public, but a lot of these things are run on Gmail accounts, on private Gmail accounts. and we have talks about ethics and data protection and I think we really should fulfill what we preach at the conferences so we never had like a breach of that but I think we really also need better governance especially if we also handle private data but a volunteer organization just like a random group will not take care of it we need an organization because this is probably boring this probably needs also experience and stuff And so it's really important also to build stuff like that and have people to take care of that and are proficient in that. Yeah, and also mentoring. Because many people, when I talk to people at conferences, you often see, oh, yeah, what do you think about this? Or just like connecting people or just like say, okay, look, this is a playbook. This is how we run a meetup. This is what we hand out to sponsors. just share this knowledge, and help to get things going, because there's many people, if you start from scratch, you have to reinvent the wheel very often. So actually to enable future organisers, also independent of Python, I decided to launch or help launch Pioneers Hub. And it's Pioneers, I have to point out it's an I, not a Y, so it's not for Python. So it's a non-profit, it's a Gemeinnützige GmbH in Germany, and its mission is, or our mission is a new non-profit for Germany to help in maintaining thriving communities and expert in technology and research, to share knowledge and collaborate in a way together. And it covers many things I mentioned before, and I also learned last year I would like to have it also available for non-Python. So you probably, because last year I was faced with a question, for example, hey, what about let's do something in MLOps? I said, oh, that's great, let's do a MLCops conference. And then we thought about it at the Python Software Verband, said, oh, yeah, we all love the idea, but all bylaws say the Python Software Verband is about Python. So we are not allowed to do MLOps or have to build like a really shaky bridge and slippery slope, and I can look what our book auditors would think about it, but I saw, hey, we need something which is probably tech-neutral, but I would love to introduce the Python culture and everything I learned in the Python community here, and to help others thrive and innovate together, and so the first partners of Pioneers Hub, Pioneers Hub is running the office of the Python Software Verband, PyData Midwest, also helped building this conference, and so there's like an initiative, so what about if you had like a conference or like a community gathering, probably not so many people more often in your area. So that's why we launched an initiative to collect people who are interested, to contribute, to do run a local chapter to do something and also footnote if you have any other tech, I see Jens from Rosen has other technology and really cool technology by the way and yeah so to really to accomplish the following, like team connect with others and give guidance how to set it up but also like first steps, know how, how approach sponsors probably also point to sponsors via the network and also provide other resources. It can be like a free account for network drive, email account, whatever. And also like content, like really supporting, finding speakers, getting things going. A few weeks ago, we had a PyData leadership meeting in Eindhoven, and we basically all agreed, yeah, we could do a lot more local meetups if we had more speakers available or like if we know who is up for speaking because there's a huge gap um and especially if you start something new that's um it's hard and also like apart from meetups and local things also like what i really like when the in the pub like me being a board member of software for bunt my colleagues or like people who are close to present software suddenly barcamps have disappeared after corona have appeared after corona again sorry So, there's now three Barcamps this year again. So, it's also like a different format. It's open. It's a very good fit to the community and everything. And so, yeah, it's good. And I think it's really important to support that. And as I said, tiny budgets can help. Mentoring can help. Some tools and guidance can really help to make things happen. And, yeah, get something great started. And to build a bridge to the open again. I wasn't there, unfortunately, in person. But imagine, some people meet at EuroPython 2014, and something like Jungle Girls comes out of that. And I think there's so much more potential for many other technologies, and maybe that's a personal note. I like to like my work. That's why I think I'm an interested person. I think many people here are interested and curious people who want to learn new things. This is probably why many also are drawn to tech. Because there's also always something new. It's probably not so much routine. We can really help building communities and enable them and make things better. Yeah. That's actually the end of my talk. If you want to learn more or just connect, then there's a QR code. You can just go there and fill out a form. I'm also really happy, so I'm around at the conference. The good thing is we have everybody on boarded. The opening session is done. Now my talk is even done, so I'm really available for the rest of the conference. I'll be around the Python software for Bantam Focus, Pioneers Hub. We have a combined booth. Just come by, have a chat, and, yeah, let's exchange some ideas and see what we can do can accomplish together. Thank you very much. So thank you so much for the talk. It was really great. I hope you all have enjoyed it. So we have some questions for you. First question is that how do you lead the team without making everything dependent on you as a single person? Oh, that's a really hard challenge, actually. And I'm actually honest. I'm still working on it. Sometimes it's difficult. It sometimes depends on how your peers and organization or how things evolve. Sometimes we have to improvise. For example, last year, in the opening session, we do an open call for committee members. We also called for a conference chair. But we also learned last year, conference chair, the job is so big, nobody was up for it. And then we said, okay, what do we do? So that's why we decided to do a steering committee for organizers who worked on the past conference before to say, okay, let's mentor. We focused on mentoring the committee members, give guidelines, or, for example, just throw a bunch of spaghetti code over to Florian over here, who is one of the program chairs, and say, okay, please build a real library. And he built a real library out of it. And so, of course, it's really hard work not to make it dependent on a few people, actually. And that's what I learned. We also have to see the last few years we still had to come back out of the crisis because suddenly no more in-person events, a lot of online. We have to sometimes also reinvent things. So, for example, the meetups, we are streaming the meetups, for example, now. So, of course, I think we are still in a transition to distribute work and responsibilities better. But still, I believe it's not just like a Google Doc or documentation how he thinks and you put it somewhere in a Git because also I learned people don't really read all of that. I think it's really important to pass it on from humans to humans or like switch roles. But it's a hard job that is not dependent and I'm still working on it because, yeah. Or load off things to other, to an organization running things. Like, especially the boring part, probably. Okay, so we have one more question. Why it is important for you to support local tech community and how everybody can contribute to the mission? Because I'm really selfish. Because, actually, it's one really great example. So, before Corona, we had PyData Frankfurt. And what I learned is, hey, just go into breaks and complain about even the tiniest problem. And I tell you about, I was standing there in a break, and I said, oh, yeah, I think it's really so hard to manage condo environments. And I was just like, oh, yeah, it's probably you, it's okay to complain in our community or say, I have this tiny problem. And just talk about it. So it's not like, don't be shy about it. And, yeah, because sometimes people like imposter syndrome will probably push you, hey, go out and shine and look what a great expert you are. And I learned, oh, no, just be a human person and talk about human problems. And so, actually, I was standing there. And here, Uwe was standing next to me and said, oh, don't you know you can use miners from history there? And then it will only create environments for the packages you really installed and not all the sub-dependencies you get with conda and export. And I said, oh, that was awesome because this feature wasn't really documented. So, it helped me a lot. I think it was also one of Uwe's most popular tweets. So that's how these things work. And actually, yes, it's also something because it helps me to fulfill my work. Also being close to the community and talking to other experts also enables me in my job because I can tell my clients, okay, should we use this tech or should we use this library? And I can say, okay, go for this library. It's more stable. It has bigger support. Or, for example, no, this is probably this one bear is legacy. I would say it's probably legacy and still a good library. But probably for new project, you want to use the other bear. And I don't want to drop names, but people will probably know what I'm talking about. Yeah. So, yeah, it also helps me doing my job and do projects. And one more thing I'd like to add. Yes. And one more thing that how everybody can contribute to the mission. Again. Someone else also wants to contribute with you. How they can. Again? How everybody can contribute to their mission. Yeah, just go there, say what you're up to, what you like to do. And sometimes the job is taken already and we say, oh, sorry, what about doing something else? So, I mean, it's just like a human coordination thing. Of course, it's important. So we don't need a program committee of 50 members, I think. Yeah, there's like 10 core members and, of course, many reviewers, as you saw in the opening session. But, of course, I would say, I think always like, yeah, like the Apple team with few people solves this better than the Microsoft team back in the 90s, of course, today they do everything with 300 people. So, of course, too many people can also slow things down. And also, like, it's very important also to agree, how do we want to make decisions? and I see especially there's a shift between generations because usually my generation would say, okay, it's a democracy we are five people, three people are for it two are against it, okay free win, which is not bad but very often I see young generation, you have to see consensus on everything and consensus on everything I wouldn't say it's necessarily bad but you don't also have really bad compromises And if you look at German coalitions, we have a very good example of that. So my call with that, it's totally okay. And it's also like I sometimes think it's emotionally disruptive if not everybody's agreeing and I just want to spread the message, no, I totally agree, not to agree from time to time because you're still a team and your opinion is still valued. So it's totally fine just sometimes to say so, yeah. So I'm very often overvoted by my fellows in the steering committee. So sometimes they see something differently and I said, hey, we're fine. That's why we are a team. Yeah, to bring in different perspectives. And yeah, here you go. Okay. So we have another question. How not to make an event dependent on you? How do you off-board yourself when you have organized and even for a few years? Again, I didn't get the question. It's like how not to make an event dependent on you. Yeah, that's quite the same. Yeah, I mean, it's distributed. And also, I wouldn't say it maybe looks from the outside, but I would say if I dropped that this morning, this would have still have happened. Maybe the opening wouldn't have been as cheerful, but it's... No, I mean, it's also like a different perspective sometimes. I mean, you see, we try to share presence in the opening, but sometimes people don't want to be on stage in the headlights so sometimes I say hey I want you to go there and speak and then they no Alex please rather you do it so also there's a bias and I think this year we really worked very distributed and yes so it might just look from the outside so we are done with the questions if you still have any questions you can reach out to him Sure, I'm around. Out the conference. And thank you so much for listening. Thank you.