Django Governance: Growing Sustainable Open-Source Communities Together

Hi

Speaker 1: everyone, it's great to see so many people interested in governance and growing sustainable open source communities together. So let's start with some introductions. My name's Tim Schilling. I'm on the Django Steering Council for the 6X series. I'm an admin for DjangoNot Space. I'm an admin for Django Commons. Software engineer for Spyre Edu, but outside of software, um some of the things I like to try to do is I enjoy trying to perfect my breakfast. I've come to the conclusion it starts with homemade bread. I try to maintain several hobbies, one of which is gardening, which did not go so well this year. And then I enjoy playing fetch with Roland, who is a cat. So yeah.

Speaker 2: Well, I wish my cats would play fetch with me. Um hello everyone, I'm Rachel. I'm a um I'm a Django Girls longtime organizer and trustee. I'm an admin and co-founder of Django Knot Space. I'm also a DjangoCon US organizer. And when I'm not tending to my Django Gardens, Let's say you could find me practicing my guitar or going for a jog. I try not to take myself too seriously, governance excluded, obviously. All right. So why are Tim and I up here talking to you about governance? Well, honestly, we've been working alongside our organizations as they evolve. And we've been learning. And this is our attempt to

Speaker 2: share what we've learned about governance while we've been working on those organizations. So our approach here is prescriptive, it's exploratory and descriptive. We're learning as we go and sharing that with you and hoping to build a toolkit and not a rulebook. So who is this talk really for? You. Well, the people here last year too. Okay. This talk is for you, the jail community, whether you're an organizer, maintainer, volunteer, contributor, community member. conference attendee. So if you're a current or future organizer or maintainer, we want you to learn. About good governance and take that and help plant seeds of good governance across all of your communities.

Speaker 2: Avoid the pitfalls that many of us stumble into and hopefully walk away with some concrete governance you can use. But more than that, let's try as a Django community to find some common standards and approaches and good governance that we can set going forward so we don't all have to reinvent the wheel.

Speaker 1: And if you're a contributor or a community member, we want you to uh empathize with maintainers and community organizers. We want you to see that these people are doing some invisible work and they can use your support. We want to inspire you to be more impactful in your communities. And with greater awareness, you can be more effective in your contributions because by understanding what the community is doing and having an idea of what's all possible, you can identify those gaps and then help tend the garden.

Speaker 2: Great. Okay. So what exactly is governance? Let me read you something from the CNCF Governance Guide. That's the Cloud Native Computing Foundation, which hosts projects like Kubernetes and has extensive open source governance experience. They really capture it well with, at its heart, open source project and community governance is about people, their rights and responsibilities as part of a project, and the expectations others have for them. So for us, what is governance? It's the processes and structures that define how an organization operates. It's the meetings, the async conversations, decisions, documentation, emails. Lots of things that take a lot of time but often go unnoticed. So it's like watering your garden, weeding, inspecting for pests and prevention.

Speaker 2: Essentially, it's your how we work together agreement. And here's the thing, it's often completely unwritten until something goes wrong. Good governance is like pruning. Well done, it's barely noticed, but neglected quickly leads to chaos. And let me tell you about chaos. How many of you have walked down on a sidewalk and thought to yourself, wow, these trees are perfectly pruned? No. Have you walked down the sidewalk and gotten hit in the face with a tree branch? And then you're like, oh wow, they should prune these. Same in governance, okay? Alright, so organic verse, intentional growth. Even when we think there's no governance, there actually is. It's just informal and unwritten. Even informal gardens have governance, right? Who decides what to plant? Who waters when?

Speaker 2: Who responds to pests? Even if it's not written down, it's still there. For example, right now in this conference, there's an unwritten governance happening. I mean, of course, we have a code of conduct, but um We all have an understanding of how this talk will go. Tim and I will stand up here, give an amazing talk. You all will listen very attentively, laugh hysterically at our cheesy jokes, and clap loudly at the end. We all understood that and agreed to that walking in here. Right? So I mean we could ask you to stand up, run around the room screaming, uh, but I guess not if we want to be invented uh invited back next year, right? All right. So let us give you some personal examples. Both Tim and I are involved in some non-tech

Speaker 2: organizations and communities, and we've seen them struggling with the same gaps in governance that we oftentimes see in tech spaces. For example, uh I'm I was part of a running club that had great energy, but no structure. So when the main organizer got burnt out and left, the club couldn't continue. It just kind of died. The club had great energy, but there was no framework to sustain it when life happened to the key person. Tim?

Speaker 1: Yeah, for my home brewing club, it was a little bit different. So we had set roles and expectations, but people just weren't interested in doing it. They were happy to show up to meetings, they were happy to bring beer, they were happy to share it and drink it, but I couldn't find anybody to help me with it. I was doing most of the work. But I couldn't find somebody to even take notes during a meeting. And so eventually I had to step away. And it wasn't just from the club, it was from homebrewing altogether. So the club still operates today, but it's with that same format. There's one person doing 90% of the work, and it's clear, at least to me, that it's burning them out too.

Speaker 2: All right. So let's get to it. Why does governance matter? I mean we talked uh previous slides similarly, but um let's dive into the concrete things. So, good governance protects people and communities. It prevents burnout by setting clear expectations and reasonable workloads. It's like having the right tools in your garden shed. They good governance provides protective measures. Term limits prevent exhaustion. Role definitions prevent scope creep scope creep. And clear processes prevent people from feeling lost and overwhelmed. Next, good governance prevents crises Or tries to. It provides frameworks to handle conflicts constructively before they become community splitting. And we've seen this in other organizations. It's not about avoiding disagreement, it's about having structures to navigate disagreements productively.

Speaker 2: And next, good governance and powers know. What does this mean? This means being explicit about things like which roles are you interested in? People last longer if they're enjoying the role they're in. How much time can you realistically commit? And then happily accepting and encouraging no as an answer. It helps members set boundaries without guilt. And when there are clear processes and expectations in place, stepping back becomes a natural part of the organizational rhythm and feels less like a personal failure. This prevents burnout by normalizing healthy limits. And next, good governance enables growth. With clear onboarding, supportive mentorship, and inclusive practices, organizations can scale sustainably. It creates pathways for new contributors to eventually grow into leadership roles.

Speaker 2: Tim.

Speaker 1: So now we know why governance matters, but how do we recognize when it's time to tend the garden and update things? So our shortest answer here is when you're experiencing difficulty managing your community, it's probably time to revisit your governance. And these difficulties can crop up in a number of ways. Get it? Crop up? I did get a laugh. Yes. Alright. So for example, maybe people are interested in contributing, but they don't seem to follow through. And then you notice this kind of becomes a pattern. Or maybe you want to bring someone onto your team, but you're unable to give them a full list of the expectations for the role. Another sign is when discussions initially start out really engaging. But they lack any sort of clear conclusions

Speaker 1: and then they fizzle out. And our last example is when decisions are occurring in private channels rather than out on team channels. So if your community is facing one or more of these, it's a good idea to revisit your governance. But keep in mind, more governance does not create more activity. What it will do is empower your community members to bet better operate within the community. It will give them the ability to make better decisions and take action. And that's not to say that these changes would be easy either. All those difficulties, those require cultural changes, and those kinds of changes are incredibly hard. It requires patience and perseverance to see them through. So start small and celebrate your wins as you go.

Speaker 2: Great. And you know what, Tim? Something we do really well in the Django community is celebrating wins. We celebrate wins of all kinds because all contributions matter. I think that deserves a high five. Uh so this but this talk isn't just a feel-good talk. This is actually critical for good governance. And let me show you what I mean. First, we have here's a photo of Django Girls Plus Baira. Django Girls Plus provides opportunities for thousands learning to code through free workshops worldwide. Teaching and community building are critical contributions. Here is a photo of the organizers of DjangoCon US 2024. Conference organizing is governance in action. These folks create spaces where our community connects, learns, and grows.

Speaker 2: So thank you for providing this space and this platform. And in this photo, we see Natalia, a Django fellow, teaching how to contribute during a sprint. Knowledge sharing and mentorship in action. This onboarding work is essential. And here's a photo collage of some members of DjangoNot Space. Django Not Space is an open source contributing mentorship program. And what we've learned there is that the real work isn't just code contributions. It's checklist, scheduling, emotional support, community building, doing um code review events, test a thons, newsletters. Um we value all these contributions equally.

Speaker 1: And we have one more celebration over Django's 20 years. Happy birthday, Django. Uh we've gone through each of the three governance models we're gonna cover. At least I think we did the Django core team I wasn't here for, so I'm gonna count it as appointment, but Yeah. So the three models are founder leader, appointment, and election. Each of them have their own advantages and disadvantages. The founder leader or benevolent dictator for life is common for new projects. It provides a great deal of flexibility because there's only one person that has to make a decision, which then means it also has very clear lines of authority. Like you know exactly who to go to. In this structure, it's very popular with new organizations because it enables them to go through rapid growth. Long term though, there could be some side effects.

Speaker 1: Community grows and the number of contributors increases, they become all dependent on a single person. This in turn can cause those community members to feel powerless to make real change. And lastly, if a founder leader decides to step away from a project, there is significant risk that the project may disappear altogether. So obviously, to counteract this, you should add people to the decision maker group. And one way to do this is to appoint others, hence the appointment model. We use this governance model quite frequently in the Django community. Django Not Space, Django Commons, most of the working groups, most of the technical teams, they use this model. Having the team select its own members. Gives them consistency as the team changes. They're also allowed to choose to scale the size of the team based on their actual needs in the moment.

Speaker 1: However, this model doesn't always encourage outside contributors to be active. People may sit around and wait to be chosen rather than to just outright participate. Taking um the other downside, sorry, is that these groups may lack some diversity. This is because uh They select people from within their own networks, and so you just kind of keep selecting people within your network. It's kind of hard to find people outside of that. Taking it further though, it's possible that a homogenous team doesn't reflect the true diversity of a community. And may struggle to consider all perspectives when making decisions. Which then brings us to the election model. So this provides a pathway for the decision-maker group to reflect the community as the community changes.

Speaker 1: It provides greater transparency and can usher in greater diversity. The regular turnover it causes helps bring fresh ideas into the decision maker group. However, elections themselves don't create effective teams. Without explicit roles and procedures, elected individuals can feel lost. Compounding on that. If a role isn't well defined, it's difficult to know if someone, or yourself included, are gonna thrive in that role. Then from a broader perspective, elections are costly to a community. It requires a significant amount of effort to hold them, and then there's an opportunity cost the team pays when you transition from one election uh one elected group to the next. So, obviously, all these models have disadvantages, but these shouldn't serve as a reason not to try one.

Speaker 1: Instead, they're a reminder that we need to focus on supporting individuals so that they can be more effective.

Speaker 2: And supporting effective supporting individuals effectively means intentional organizational design, right? So let's discuss how to do that. A good rule is to start small and iterate often. Now this is kind of more for if you're just getting started or you're a small group. You'll want to write things down. This could be in a Google Doc or README. Oftentimes there's meetings and meetings of how to do something. But just do it instead of talking about it. That's the best thing you can do. You'll want to make it findable and put it in an organized place so it's not buried in a chat history or a private message. And you should review it regularly. You can even set calendar reminders to have um quarterly or yearly reviews. And going forward, um You'll want to leave change for versioning, so you can uh

Speaker 2: you'll want to leave change for leave room for change so you can version your documents. And then going forward, you want to keep governance in mind. when you're discussing new changes to the organization. And the reality is every organization has how they work and then how how their documents define their their procedures. And the goal is to try to minimose minimize those differences where possible. And so let's look at a minimal viable governance stock, right? So you can you can get started today if you'd like. This is this could be a Google Doc or readme in um GitHub. It says how we work version 1. 0. For decisions you can write, we make our decisions on weekly calls and we document those in our we meeting notes. And for new contributors, you can say if you want to contribute, open an issue, and we'll help you get started.

Speaker 2: And for questions, you can say if you have a question, ask in the general chat on Discord or wherever you chat. And for major changes, you can say big changes require a one-week discussion period, and after that, the team will vote. That's it. That's your governance doc. It doesn't need to be a large, complicated document. You just need to write down how you actually work. And then it can grow and refine things from there. So your governance garden infrastructure. So what's the essential infrastructure of your governance garden? First, rich soil. This is your decision-making process. From this, everything else grows. For example, at DjangoNot Space, we require 50% admin approval for decisions. Second, irrigation. These are clear role definitions that channel responsibility where they need to go.

Speaker 2: Who runs meetings, who takes notes, who responds to emails. For example, at Django Knot Space, we have these roles defined, but we revisit them often. Third, seasons. These are natural cycles of renewal through term limits. At DjangoNot Space, we do annual admin reconfirmations. And finally, seeds for next year. This is secession planning. Life transitions happen, and we need to plan for smooth handoffs. This is the lottery factor, right? Can your garden survive with the key gardener running off and becoming a farmer to tend to their own garden?

Speaker 1: All right. So now that we have our general framework, let's talk about some of the tools you need to start gardening your governance. There are two key areas to focus on here, communication and people systems. For communication, you're gonna need some sort of synchronous communication style. You're also gonna need an asynchronous communication channel. Discord and Slack seem to be pretty popular. You're also going to want to have some document that defines your roles so that people understand what is expected of them. And then on people systems, You want to have checklists for onboarding and offboarding. And this can be as simple as add person to one password, remove person from one password. It doesn't need to be more complicated than that. Then Oh, I got that backwards. Roll definitions on this side. Um, yeah. And then you should have some

Speaker 1: type of support for people to be human. So with Django Not Space, we have a life channel, and then the Django Discord, we have a life pets and family channel, which by the way, if you are a fan of cat content, that is a really good place to go. But ultimately, you're going to start where you are, document what you're doing, and then tend to your governance garden regularly.

Speaker 2: All right, so speaking of tools, this is one of the tools we use at Jingana Space. You can hear us refer to that organization often because we're both part of it and a lot of work has gone in recently for it. So this is an example of one of our docs designed to help us be sustainable, onboard for folks more easily, and set clear expectations and clear expectations before folks decide and commit to these roles. So it's a great doc to kind of see what you're gonna get into It guides their mentors through an eight-week program with very specific actionable items. Like for week zero, we have attend orientation, join Discord, post an intro message on Discord. We have similar checklists for all the roles in the program, even our own, because we need to be replaceable. Some of the key benefits we found are clear expectations for everyone.

Speaker 2: Nothing falls through the cracks, or hopefully it doesn't. easily onboard new mentors and these become living documents so we can update them after each session based on lessons learned and feedback. All right. So everyone listening, this is the governance health check. Everyone listening here today and at home, I want you to think of an organization you're involved in as a As a contributor, organizer, mentor, volunteer, conference attendee, user, okay, have that in your head, and I want you to think of answers for all these questions with me. I'm gonna ask Tim to help me out up on stage. So Tim, I want you to think of an organization you're involved in and answer as honestly as you feel comfortable. You don't have to tell us what organization it is We can guess. Okay.

Speaker 2: Um okay, so first, uh transparency. Do people know how decisions are made? Thumbs up, yes. Uh accessibility. Can new gardeners join easily? That's a maybe. Next, clarity. Is it clear who does what? Yes, thumbs up. Sustainability. Can the organization survive if key gardeners leave? Thumbs up, yes. Inclusivity. Does everyone feel heard and included? Yes, as as far as he knows. All right. Um So it looks like, thanks Tim, it looks like uh Tim's organization is doing pretty well if I interpreted those answers correctly. And though I saw some concern around accessibility. That's normal. Every organization has one or two areas they need to work on, and that's exactly what this kind of governance health check is for,

Speaker 2: to identify where to focus intending your um to your governance garden. And I hope that you kind of reflect on that and maybe can think of some concrete ways or things that you need to focus on in your organizations. All right, let's talk individual actions. Okay, what if so we've looked at a lot of ways organizers and maintainers can improve their organizational governance, but what about the rest of us? What are some actions I can take to improve or help these organizations from an individual level? First you can ask how are decisions made here? Don't be afraid to ask this question when joining a new community. It shows you care and it can reveal gaps. Next, you can look for gaps and advocate for solutions. You don't have to be a maintainer to know if something's broken or not working.

Speaker 2: Could is this documented? Could we document this? It is one of the most helpful questions you can ask. Push for things to be written down and findable. But don't be surprised when they ask you to do it. Next, vote with your time by choosing healthy gardens. Life is too short to contribute to communities with toxic governance. Invest your volunteer time where it's appreciated and sustainable. And lastly, participate actively in garden planning by showing up to governance discussions, give feedback on proposed changes, and volunteer for roles when you can, because every gardener makes the garden stronger.

Speaker 1: Yeah. We want you to remember that governance should be a human-first endeavor. This means that you should build your governance with people in mind. So like letting a field remain fallow, you can build in forced rest periods to force people to take self-care sprints. And you should try to set time limits on projects because you're never going to be able to do everything, and they can help you tend to the plots you're able to. And lastly, you should nurture ways to allow people to express themselves through something like a life channel or a random channel. Just know that no perfect system exists. Each of our gardens is unique, and you know what? That's pretty cool and we're celebrating. So thank you for listening and wanting to tend to your governance garden. We hope you dug it.

Speaker 3: Round of applause to the speakers. Thank you so much. So we only have a minute Anybody with a question?

Speaker 4: Thank you so much. Uh I actually really enjoyed this presentation. So quick question, did you ever come across the situation when there would be too much of the governance and how would you detect it and what would you do with it?

Speaker 1: Too much governance? Yeah, I I think that's um we've seen that in the Django community. Uh yep. I I I The solution there is that you have to find ways to pair it back, I guess, and that takes a while. What Rachel was saying earlier of identify what you do and what your governance does. If you look at what you do and document that and then anything else, just remove that then. That's the simple answer, I suppose.

Speaker 5: Um I'm curious uh Yeah. How do you have any thoughts on like voting? Um I I think sometimes with the voting process uh it's uh uh maybe more of a popularity contest than uh who would be a great leader in the community. Do you have any thoughts around that or how how to kind of identify leaders versus popular people?

Speaker 1: So I I've thought about this. I think when you reframe that from popularity to trust, it gets a little bit easier, under more understandable. It's everybody's voting on who they think they trust the most. And In my opinion, it's more of the fact that we don't have clear roles of like what's actually expected of them. So then you don't know if you can actually run or like how to evaluate somebody. It's really hard for us to elect people on um merit for the exact position rather than just trust. And that that's my current thought.

Speaker 2: And I think like It's a popularity contest, so we it's based on like how much we know these people. And it's usually because they have community buy-in. They've been in the community doing things. So like he said, we have trust of them. New people, they might be great for the job, but how do we know um without having that experience first. So I mean it's a valid point, but it also has its benefits. Thank you.

Speaker 3: Okay. Thank you everyone. I think they'll take other questions in the hallway. Thank you. And we have Lightning Talks coming in five minutes. Thank you.