Community Update: Python Software Foundation with Dawn Wages
Published October 23, 2025
This video features Dawn Wages at DjangoCon Europe 2023 in Edinburgh, Scotland.
Keynote: A New Adventure Is Born: How Open Source Dinos Unite
by Dawn Wages
https://pretalx.com/djangocon-europe-2023/talk/DJF7SF/
A coding brontosaurus teams up with other Open Source dinosaurs and unite as a community to advance their project. Let's dive into a fun approach to personas, developer experience and OSS.
A coding brontosaurus teams up with other Open Source dinosaurs and unite as a community to advance their project. From The Apprentice-Rex to the Ptero-Python-Party-o-Dactyl, this rag tag group of friends' unique approaches work together to make our journeys better and our communities sustainable.
This keynote, inspired by the playful characters of the classic children's film A Land Before Time (1988) brings a fun approach to how persona marketing applies to the participants in Open Source. We compare project ecosystems, talk friction reducing activities tending to your community, and how iterative product and developer experience design through empathy is essential to project adoption.
Dawn Wages presents personas and complex-systems thinking as practical tools for stewarding open-source communities. Seven dinosaur personas represent different motivations—learning by doing, watching, discussing theory, building relationships, pursuing mastery, or racing toward a working feature—so community leaders can design better onboarding, documentation, events, safety practices, and feedback loops. She argues that healthy communities need iterative code and community health checks, clear definitions of success, inclusive safety work, a fast path from beginner to productive contributor, and documentation. She also asks communities to examine how money, sponsorships, employment, trademarks, and unpaid labor shape open source, while recognizing that growth is not always the goal and that bias cannot be eliminated completely. The closing message is that personas can build empathy, comparisons depend on vantage point and contain bias, and communities accomplish more by working together.
Summarised automatically from the transcript.
Automatically transcribed, so expect mistakes in names and technical terms.
Speaker 1: Hi everyone, it's so exciting to be here. I see a lot of familiar faces, I see a lot of new faces. Um hopefully by this end of this presentation we both feel like we know each other a little bit better. And also this is a perfect size group for it to be. a little bit interactive so if you have hands if you're able to raise your hands on a couple of things I'm gonna do a little bit of call and response Also, there are interactions, there are graphics. My lovely fiance is a designer and she's like did this thing for me. So we have it It is present throughout the presentation as this thematic. There were a lot of d dinosaurs. I love the theme. I leaned into it heavy. So just Bear with me and just like enjoy it.
Speaker 1: I also am not the only dinosaur presentation this week, so you're getting dino heavy. Um So kind of a subtopic and to get a little bit more like kind of academic and on the meat of the presentation, we're going to be talking about personas and complex systems. both are different disciplines um and can have some scientific and academic terms. But we're gonna I mean we're open source. We're just gonna like move fast and do as best as we can with the resources we have. So we're going to try to put our uh product manager hats on as well and then dive into some of these scientific concepts that may not be uh that may be a little bit more hairy when we don't have the ability to uh collect all the data.
Speaker 1: So my name's Dawn Wages. I'm a Python community contributor. So my nine to five, I work at Microsoft as a Python community advocate. So if you have thoughts on Python and any of the Microsoft ecosystem I may not be able to help you immediately, but I can find someone who does. But if you specifically have thoughts on Django in VS Code and IntelliSense That's my bag. I would love to talk to you about that gathering feedback. That's my nine to five. And also thank you to my employer for sending me here. I am the treasurer for the Python Software Foundation. I've been on the board for just a year. Elections start June 1st. The closing of nominations for board seats
Speaker 1: is the 15th. Elections start June 20th and end on the 30th. So please uh participate. There are several levels that are able to vote. That's the contributing, maintaining, excuse me, contributing, managing, and sustaining members. and fellows. So please, please be involved with your the Python Software Foundation board elections. And you can also do that online today. I'm also on the core team for Wagtail CMS. I have been since 2020. Wagtail is a content management system built on top of Django and it is an opinionated CMS for the opinionated web framework. You'll see a lot of intersecting communities as I talk about them today,
Speaker 1: some concentric and eccentric circles, and a lot of facets of community talk. I'm also the creator and uh an advocate for the op um organization for ethical source. Uh me and my team of Lee Gill and John Soto, we created the first anti-racist ethical source license and we are down for the conversation of what does capital O capital S mean and can your ethical source license still be capital O capital S open source And I would love for everyone to keep in touch. So it's so exciting to be here. And so now that you know me, you can keep in touch with me on Macedon. online. You also have my website there and like my web as the social apps
Speaker 1: Keep changing, you'll find all of the latest there. So what this presentation is and is not, it is a fun way to view stewardship of your community. We talk about personas, complex systems, these gonna that it's gonna get actionable tools, uh not just philosophical concepts, but there's not going to be any type of tutorial on community. I don't even know what a tutorial on community actually would look like, but this is not that. Um and there's not a case study, although I will be name-dropping certain uh certain uh communities and comparing them amongst each other and we'll probably disagree and that's for a good thing. Here are the slides for my presentation
Speaker 1: and also a QR code. I'll give you a second to do that if you want to follow along. There are some random weird words that I created because you know I I like to be fun and punny and are excellent uh team of of captioners are gonna were able to help you out with that. And my 3D models by my excellent fiance who's starting an agency and she's absolutely brilliant. Um it's called Augmented Reality Design and Development Ard. And if you're familiar with Philadelphia, I'm from Philly. If you're familiar with Philadelphia uh slang art is like uh art So it's art, so when you're doing a confirmation. Anyway, it's a very cute pun.
Speaker 1: I'm very excited for her. So let's start with some assumptions. First iteration. This is never going to be a one-and-done activity when you're managing a community. We're going to iterate over KPIs. KPIs are key performance indicators. This takes macro objectives into your micro actions, like for example, your monthly or quarterly meetings, if you are interested in gaining new contributors. Maybe at the beginning of your meeting you should announce the number of new contributors you've had. We iterate over some of these concepts. It doesn't have to just mean metrics. My favorite type of check-ins is when I have small meetings with Organization for Ethical Source.
Speaker 1: We do PIS check-ins. So that's uh P I E S, that would be starting with your physical, that would be your intellectual, your emotional, and then your spiritual check-in. That's it's easier with smaller groups, but it really brings uh an uh uh the the ethos of your organization what's important to you first. So I I I'm going to continue hearkening back to the theme of iterative iterative approach. Bias will always be there. Um it is not something that we should ignore or also just accept. accept, but we should know that it is on the horizon and it is should be an ever-present goal to try to eradicate bias as best as we can, but we will never be done bias will always be there.
Speaker 1: And I'll talk more about vantage points and and how we address bias or how bias creeps up as well. And capitalism. So value of our time is now equated to the value of the yield produced in that time. We are all production engines in this system. Sometimes community can just be the praxis. Sometimes community can be the goal. And as I go through this presentation, I might not always mention that, but sometimes the goal is just community, it's not always about uh creating the best product out there. Although that is our goal. We want our widgets to do the thing really, really well. So I I want to also uh emphasize that capitalism um
Speaker 1: will give us the idea to practice practice our craft, uh but we also want to uh implement community praxis. And growth is not always the most important thing as well. So sometimes small groups are really important and they're nimble and agile and we don't always have to get bigger and sometimes we join bigger groups to do bigger things. So we will start with some dyno definitions, then we'll look at um dyno examples and classifications, then we'll look at a community map, and then we'll do some community map comparisons. This talk is a more grown-up
Speaker 1: version of my first ever talk that I've done. I was noodling over these concepts between like 2016 and 18, where I was learning about context uh complex systems and uh network clusters. Um and if you've watched the PyCon keynote by Kara Willing, she had amazing metaphor and talked really uh candidly about our community and different nodes and relationships in our community. And I I'd like to think this isn't like the same like cinematic universe of Carol Willing's talk and I really appreciate uh her lens on that. So if you haven't seen that yet when it releases, go check that out. And hopefully this is also a journey as well. So complex systems are systems whose behaviors are inherently difficult to model.
Speaker 1: It is in the name. Dependencies, competitions, relationships, and other types of interactions Complex systems also are is a discipline in science. So some examples are like climate. We have organisms, human brain are incredibly complicated. infrastructure contraint constraints, transportation, communication. We have complex struct uh software and then in this example we'll be talking about social groups. And there are also social groups that interact with complex software, but the the noun today, the it, is the other people. As I mentioned, and this is a science in and of itself, so if you're interested in learning more, there are several levels of evaluating this science. So the interdisciplinary approach, there are scales to complexity.
Speaker 1: There's also adaptability versus efficiency. And the paper that I mostly reference when I talk about complex systems in this talk, but then also other things that I write is the introduction to complex systems science and its applications. By Alexander F. Siegenfried and Jenner Barel Jan. So my favorite part of this domain is that it's interdisciplinary and how we're organizing ourselves. So my undergraduate degree is in business. And my favorite class was strategy. And we had this one class called where we talked about Porter's Five Forces. So we had uh competitors and systems and and um uh it talked about the symbiotic relationship so keep in mind some of those systems
Speaker 1: uh in this example In personas, what do people think, feel, inside? Also a little bit more about me is my background is in e-commerce. So we had big budgets to sell widgets, to track down people, to talk about how they bought widgets And as we got more specific with our communities, it's very common to find marginalized communities of age or race or gender. Inviting those and those people into the conversations, making sure that these are data-backed assumptions that you're making.
Speaker 1: And then the goal is to just paint a fuller picture of a person. But we're not using people this time because this is a lesson scientific approach. We're using cartoons. So this is all based off of the uh archetypes of uh dinosaurs in my Childhood favorite movie, The Land Before Time, a 1988 classic. Uh can I can I get some hands on who knows The Land Before Time Oh yes, yes, yes, this is so good. Okay, great. I was really worried it was gonna get lost. Okay. You don't have to know a land before time to to get this presentation, um, but it's just gonna be a way that we talk about people's motivations and strategies I come up with some fun and silly names. Please laugh. I enjoy that. At my heart I am a comedian. Um not a good one.
Speaker 1: So please enjoy this with me So here are the dinos. We have, let me introduce you to my seven friends, the OSS dino personas, that serve as a mental model for us. So we have the Apprentice Rex. The talk about it if us, the bang your head against the wall of Saurus, the watching the listen of saur. The tarot python party dactyl. I think there's gonna be some of those here. Yeah, we've been we've been partying it up this week, I guess. Uh we also have the race to the finish raptor. And the aquiredon. I think that's all seven of them. So a lot of words. I'm gonna be comparing them.
Speaker 1: These are their motivations The Apprentice Rex, we have a species known for their uncanny ability to find the experts and ask the questions. And these are people who love to dive into the work and treat development like learning a trade. To the right of that, we have the bang your head against the wall of source, similar to the Apprentice Rex, where they love to jump into the code. But they start with building and thrives on a demo. It's distinctively more solo Um and prefers quick starts uh into jumping in the code, running into bugs and problems, and just hammering their way through it. On the next line, we have the watch and listen a source. This open source dino enjoys case studies, videos. Uh let 's just sit back and watch tutorials
Speaker 1: specifically um watching it work while it's happening and learning their way. So we may have some here at the conference who love and who love to learn this way Beside our friend we have the watch and listenosaur, we have the tarot python party Dactole. They're interested in community building, friendships. not just the music and social gatherings, but really perhaps learning through osmosis, making friends that last along the way. On the far right of the second row, we have where theory meets practice with the Talk About It office. They love to chat in the hallway. They like to dive into the theoretical and the conceptual. And this one this one I think is the most likely to me to be the polyglot.
Speaker 1: They're comparing uh concepts in one language to another. Have um how does garbage collection work in one language versus how it works in Python, for example On the third line, we have the Waste to the Finish Raptor where we are zooming to our proof of concept, then to our MVP. until our alpha. They are really focused on the features. It is maybe less about the community and is mostly about does the hammer knock the nail in. And finally, we have the inquiron. So this may be a researcher-first approach. They love to uh uh become an expert and really focus on mastery of their skill, perhaps.
Speaker 1: And then how do they interact with the community? What is their motivation? So, apprentice programs, uh, pair programming, sprints open source contributions for the Apprentice Rex. We can see core code and tracebacks, maybe demo projects and third-party packages for the bang your head against the wall of source. The watch and listener Sora might like live streams, for examples, or podcasts, or videos. The Python, the tarot Python Party Dactyl will appreciate the code of conduct, for example. I've been to a really great uh conference where we had like a knitting circle. It was kinetic the uh something hour or whatever. Very nice, lovely ways to communicate with your with your t
Speaker 1: uh with your peers without necessarily doing it right in front of code I think I really resonate with them, although I don't identify as a tarot python party dactyl, because I think I learned through osmosis. I've gained a lot of friends through the community and that was one of the quickest and best ways for me to learn when I first got started. And I consider myself one of the people who is a community taught engineer or a self-taught engineer through the community. And that was really helpful. The talk about if us uh enjoys maybe Twitter spaces, for example, or Twitch streams or webinars. They like meetups and office hours, they liked Django Under the Hood, for example, or deep dive tracks. The race to finish Raptor will love
Speaker 1: Django's to-do list app and the other demos that just kind of start with you know first-party third-party packages. blog posts with demos and poil prate cl code. And the Inquire Don, as I mentioned, they're interested in mastery, so maybe mastery courses and They're addicted to collecting reference books, as I know a lot of us can relate. Here are the personas again. As I mentioned, I'd love to get to know you all. So uh what about the Apprentice Rex? Who in here thinks they might be an apprentice rex? That was me. I don't know. That was okay.
Speaker 1: What about the talk about it if us? Yeah? Okay, we're getting more. Uh what about your bang your head against the wallasaur? Okay, this is a large group. I got that. Okay. What about the tarot python party doctor? Okay. Okay. Yes. Okay, we got some energetic ones too. The watch and the listenosaurus. Okay. And the race to finish Raptor. Okay, there we go. That's maybe like a good five percent. Um and the acquire a dawn, loved mastery. Okay I love that. So as you raise your hand, you might have seen your friends raise your hand as you evolve
Speaker 1: evaluate some of these things later. you'll be able to find groups that are like you but also team up in in complementary ways and you can figure out other ways of a of approaching it. But really this is just a mental model on how other people approach things differently than you and as you're tending to the community evaluate how other people's incentives may not be the same And who am I? So I said we're gonna get to know each other. I am definitely the apprentice rex. I started this journey um through kind of an apprenticeship for it, for example. It was It was hard for me to get the first Django job. Um and so I did my first apprenticeship and then I consulted for um two and a half years before I got my first, you know, full-time engineering um
Speaker 1: senior engineer Django job. Um and so it was a little bit of an untrod non-traditional path, but maybe it's more common um than I think and maybe I'm putting the signal out there to meet you and we can talk about that I also like to be a talk about it office and maybe this whole presentation just like outs me very quickly as one of those. But I love concepts, I love being able to to wax poetic about some of these things and really, you know, uh bring it home that way. So it's I find that fun. So complex systems. So here's this is not at all the full list, obvious obviously, or else it would not be a complex system.
Speaker 1: But here are ten of the of the key themes that I find in creating a healthy community. And the ones in a little bit darker green are the ones that I'm going to talk about today. So funding sources, code and community health checks. uh code of conduct and safety tools, the zero to hero experience, so really kind of turning around very quickly and uh giving that feedback loop and some of those endorphins quickly and documentation Funding sources. So this gets complicated and can be an entire presentation in and of itself. I'm the treasurer for the Python Software Foundation and I've also been in sponsor roles on a number of uh Django conferences, uh
Speaker 1: Europe. Um DjangoCon Africa is coming up in November. I hope to see people there. And if you are part of organizations that might be interested in sponsoring DjangoCon Africa in November, please feel free to reach out to me or the team sponsors at DjangoCon. africa. We also have DjangoCon US, which I'm a co-sponsor chair for, and that's coming up in October. Money and Sponsorships has been working harder for that relationship since 2020. I'm so excited that we're coming back in full force. um in doing so in in really safe ways. But a lot of times those are contingent upon pre-existing relationships. So the people who are most connected have an easier time funding their conference. And it's not fair and I don't have an answer to that solution, but being able to put that um that on people's radar and recognize
Speaker 1: that uh you people with the coffers with the funds should be reaching out and looking for opportunities to invest back into their community in ways that don't always uh float to the top of their timeline So, thoughts, some thoughts on how to understand follow the money in open source. For example, who owns the trademark? That costs money. Figure out who does that. Who finances the conference? That costs money. Where do people who write the code work? Is there anyone maintaining the code who is employed as perhaps a developer in residence or a fellow? How about a company or companies uh that employ a particular person, maybe full-time or part-time, to work on that particular project.
Speaker 1: What about perhaps a company paying for a specific feature? So understanding some of the nuances and dynamic of the way that your open source code has been funded and whether or not it's funded and whether or not people who are creating that tool for you are making a living living wage Figure that out. Are you donating to them on some type of semi-regular basis? Or people in your community on type of semi semi-regular basis? There are so many ways that these relationships shake out and not with aligned incentives, and it really behooves us to think about who and what jives with our sense of justice and equity and compensation and open source. And I think there's definitely a way it is complicated, but there is a way that our open
Speaker 1: our source code can be open and free while people are compensated for their labor. Code and community health checks. So I put these in the same um bucket, although they are very different concepts, for example, in their application So all check-ins should be happening iteratively, as I mentioned. But this is really an opportunity for you to create a rubric for success. And to frame it in a different way, it's also a check on common gotchas. A healthy project has new contributors, so as I mentioned, you could be checking in quarterly or monthly meetings to see what the number of your new contributors are month over month. A healthy Django project should be able to, for example, manage its files.
Speaker 1: So did you know that you can run python. manage. py check? And Django uses the system check framework to check for common Django problem uh gotchas. Uh I think in this approach it's very simple one and the same. You are using a systematic way to evaluate what a community health is looking like. Do yourself a favor of building those guardrails for your community by defining what good or successful looks like. And then checking in Code of Conduct and Safety Tools. Never underestimate uh the power of building around the people and the most marginalized people in your community.
Speaker 1: It ends up making the entire community feel more safe. Um and it is hard and uncomfortable and weird, fluffy conversations that is not like code at all, and it really relies on some human and interpersonal reactions, but that practice is really important and you will learn a lot by practicing it and they will gain a lot by you practicing. Theoret to Hero experience. So these are one of the things that I really love from um being in Django Girls. That's actually my first way that I wrote any Django at all was in Django Girls. And there is a setup of the tutorial that lets you have endorphins
Speaker 1: at a really nice pace And so as you're building your project, understanding what it takes to get to that first bit of serentonin may keep people coming back. Documentation. So where intuition falls short, documentation should get us the rest of the way, which is all of the way. Our favorite communities have a little bit of all ten of these. Um but uh attributes as I mentioned that complex systems get very complex and these are broad strokes and simplifications of what healthy communities can look like. So our most favorite communities um may be able to relate to some of these. For example, Django.
Speaker 1: So the ones that are highlighted are not at all, uh or the ones that are not highlighted doesn't mean that it doesn't exist. I've been in the Django community for a while and I work in um work in it and I see some in all of it. Um but this is really where I feel that Django s uh stands out. Django, the batteries included, Python Web Framework. And I've resisted the urge to talk about Django in each of the previous slides because I knew we were going to get here. But we have the I wanted to highlight the the code and community health checks and the ease of use and abstraction, code of conduct and safety tools. is really part of their their success story as well as the documentation. It's been representing present represented, but uh key to some in the next few slides are my opinion's
Speaker 1: advantage On uh some of these other frameworks and languages. Python. So uh in the spirit of a challenge, I did not highlight all of them, uh, but uh For the most of these areas, Python's readability, I mean some even call it a delight to write more than I've ever heard in other languages. Again, I am biased But DevTools and WebTools have had an excellent few years in Python in particular that's been coming up in recent PyCons and DjangoCon. So it's been a really exciting story to see um a lot of support in uh for Django in particular. Uh
Speaker 1: for example, um VS Code is very um dedicated to having a a ta first class Django IntelliSense. We also have PieCharm, which is fantastic and there's you know obviously like no love lost. I love Pie Charm as well. So just having more to the ecosystem and we have choices um to be able to help us write Django faster and better. Also there's been really uh really cool um news uh about Python security and infrastructure Something that's always been a priority for uh the PyPA team, um Python um packaging authority team And I'm really excited for events to come back. So for those running Python events, there's a really great Python event Discord called Python Conference Organizers.
Speaker 1: uh I'm happy to connect you uh to them if you are interested in joining that chat and it's uh I think twice a month um and they just get on Discord and they just like talk about trends Wagto CMS is one of the leading um content management systems built on top of Django. Um they 're really focused on the zero to hero experience. Again, I'm biased, I'm picking uh frameworks and tools that I have access to and I've seen I'm really proud of. This is a popular framework of a different size that provides a very specific solution to Pythonesis Um, they 're really extending some of the opinions of Django to an opinionated CMS. Um, and you know, we can see if we agree on their opinions.
Speaker 1: the ease of use in abstraction and how to turn around how the turnaround time for a quick wow are among them other notable things. JavaScript. So a high-level, dynamically typed, just in time compiled interpretive language. Uh that's 98 % of client-side applications have in their stack. I would I would call that a very successful language. Um You know, it's I would call that pretty successful. And so why do we think it stands out? Some of these things we would not be able to say uh like five, six, maybe ten years ago about uh JavaScript with their maybe documentation or the zero to hero.
Speaker 1: I started JavaScript when jQuery was out so that was like, you know, let's drop it in and see if it works And I think one of the one of the ways that it really is successful in the community, not only is there a large ecosystem of like offshoots, for example, there's like the whole ECMA script um system and the organization of that um of that governing body, but they have consistent funding sources, um jobs remain consistent and the pay uh for each of those jobs kind of remain in uh a large range and remain pretty high over the years. So people keep flocking back. View. So View created in twenty
Speaker 1: thirteen is uh in TypeScript. extends HTML attributes and focuses the programming experience around the view layer. So people flock to View because of the documentation. There's also this view playground, the way that it's been able to use for small widgets, and you can kind of ramp up your experience and you can start small. are really practical approaches on why Vue has been really helpful um and the community has been really, really successful. The Zero to Hero um and the documentation and ease of extraction are also part of its success story. And we can't look at Vue without talking about React , a more mature and a much larger community where everything is a component. with his own life cycle and where is it in its life cycle?
Speaker 1: Um it's pretty standout, um well funded, um and there's consistency in React Careers and Pay and the ecosystem as well And we have it as a kind of a a standout in the progressive web app framework because of um because of its consistency. And Rust. Rust and Python interoperability has had a really awesome story in the past several years. Um, it's well-documented programming language. They don't say it's very easy to learn, but it is well documented. Um it uh has uh provides solving very modern problems in an interesting community So the community feedback and able to kind of ponder on some of these problems has drawn a lot of people
Speaker 1: as well. As I mentioned in the previous slides, there were no by no means a sense of measurement of the community, but these are all kind of hearkening back to having certain vantage points. So if you disagree on any of my um positionings of the community, that's great. And it means that you have a different advantage and so you can explore um some of the other opportunities or you can introduce um a friend to something new. And so we have how you can be a part of that community. So did I get those Python, tarot Python Partyodactyls? Uh well if you raised your hand maybe you want to do events but like think about community and code health checks Uh the way that you interact may be completely different from the process oriented um friends who want to do the health checks
Speaker 1: and you could be able to give an experiential view on exactly what may be going on and what a healthy community that feels in party with each other needs. I would also encourage you to partner with the watch and listen a source. We challenge you to team up and you get a really great picture of engagement and what content is being put out consistently. You'll get the IRL and the online activities. What about the race to finish a raptor? Like they're kind of like the the pragmatic, maybe the flyby uh Django Knot users. They really just need it for a tool and they don't really care about much else
Speaker 1: Um, I would say that maybe they're not particularly loyal, but that's okay. Like not no no harm in that. But they're notoriously hard to capture in engagement metrics and hard to get back If you have someone who really feels like there are ways to finish a raptor and you're sticking around in the community, maybe you have insights and ways that we can capture some of those metrics in the very beginning, at the start of their journey, and then figure out ways to keep them coming back. And then bang your head against the wall of saurus. Maybe you're really good at ease of use and abstraction and you can talk about that experience with your with your uh collaborators
Speaker 1: Uh you can determine on uh a lot of levels on what should be exposed and uh what are common common problems and reducing friction. And all of these are really in in congress with access and ownership. So as you're entering a new community , um being able to access elements of the community and then immediately feeling like you are able to own parts of that community um so you can engage with the system and and change it for the better We don't have to go it alone. And in the to quote an African proverb, if you want to go fast, go it alone. And if you want to go far, go together. And so in summary, complex systems are
Speaker 1: complex. Personas are a tool for empathy. Comparing complex systems is a contingent upon vantage and is riddled with bias. And we can go farther together.
Speaker 2: All right. Thank you very much for that. That was amazing. So we have enough time for questions. So if you have any questions, feel free to walk up to the mic. So just raise your hand. Hi
Speaker 3: Don, thank you. Wonderful model presented. I wonder if you have any tips on creating personas to match your community. Uh it seems quite difficult.
Speaker 1: Yeah. So I not of all of us want to use uh uh the lamp before time uh metric. So in previous roles um for e-commerce we had a lot of data. Um One of the ways the the tips that I've used for a number of open source communities has been um through job brows I mostly kind of. So I'll have one for a tech lead uh senior, junior engineer, apprentice. A lot of this will come from data and research. And even if it's uh anecdotal data or qualitative data, it is still data nonetheless. But being intentional with your approach and in trying to ask the questions of these different segments and what they need is important.
Speaker 1: So research is the short answer.
Speaker 2: Okay, um any other questions?
Speaker 4: Thanks, Dawn. Um can you give me an example of actually applying those ideas, those personas, maybe in this community or another one where you've seen it actually work and and what kind of results you've seen?
Speaker 1: Yeah, yes. So um for example, so um one of my at at Microsoft we have um A onboarding experience, we want to really focus on the, I guess we would consider the zero to hero experience of VS Code itself with Python. So we create opportunities for people to ask with like maybe zero experience and we try to meet them where they are. We're experimenting with different um uh contact methods so we have discourse, we're starting live streams more often, um, and then really doing that iterative approach on like what kind of questions are we getting. Um and then are are people doing the feedback loop on are people satisfied at the end? So the live streams have actually gone really, really well.
Speaker 1: I am not a live stream person I am learning. So some of you may see me on live streams and know that I am learning. But it's been going really well and people like the opportunity to have have those questions answered and uh Uh it also one of the things as well is because we don't necessarily say um outright that it needs to be um people new to our our pro um our product uh we have moderators of different skill levels in there to be able to be agile as well. And that's been pretty successful.
Speaker 5: Uh hello. Thank you for this very interesting talk. Um In the uh Django community today, these last weeks, there's been a sort of discussion about uh leadership and diversity in leadership. And I wonder if you think that you need the kind of diversity of different personas in the leadership as well.
Speaker 1: Absolutely. I think um it's specifically so you you can't have a new role for every for every persona necessarily and you can't just create one because you're a different personality than others. It doesn't always work that way. We gotta get things done. But I think the naturally a lot of responsibilities will fall in line with people's uh personal skill set. So really evaluating what persona you are and how you differ and then feeling like you can uh impart value in a particular role is one of the ways that happens Um, but I think it's a a service to ourselves to be able to offer what we can after we've done some of that internal work as well.
Speaker 6: Hey Dawn. Hello. Um you were talking about how growth doesn't have to be the only goal, which I really like. Uh but I do think that growth is easy to measure. And so what are some of the other goals you think something can strive for and how best to measure it?
Speaker 1: Yes. So uh some of the qualitative work. So it's really, really hard to describe um the shape of and like the value of a particular interaction. For example, is uh two people you met Who are now gonna walk away as um advocates and just like uh heroes of your product, is that more important than just letting a hundred or two hundred or two thousand people know that you exist Um and that's really conceptual and it's really part of what like future goals are and what you're actually positioning, so it's really personal to personal to each project. Um but I think I'd love surveys, people hate surveys, maybe give them a reward for following a survey doing a survey. But that is one of the ways that people can do it.
Speaker 1: But just kind of checking in and then really trusting that feedback that you give. And then we iterate. Yeah.
Speaker 2: Okay, uh more questions? Excellent. Thank you very much, Don.
Personas provide a mental model for understanding that community members have different motivations and preferred ways of learning and participating. They can help organizers offer complementary paths into the community and reduce friction for different kinds of contributors.
Discussed at 18:54The talk highlights iterative code and community health checks, funding transparency, a code of conduct and safety tools, a fast zero-to-hero experience, and good documentation. Communities should define what success looks like and check their progress regularly.
Discussed at 20:29Research is the short answer: segment the community by roles or experience levels and ask what each group needs. The data can be quantitative, anecdotal, or qualitative, as long as the process is intentional.
Discussed at 36:59At Microsoft, Dawn’s team applies the model to Python support in VS Code by meeting users at different experience levels through Discourse, live streams, and other contact methods. They iterate on the questions and feedback they receive, and use moderators with different skill levels to support users.
Discussed at 38:15Yes, although a separate leadership role is not needed for every persona. Different responsibilities naturally fit different people’s skills, so leaders should understand their own persona and contribute where they can provide the most value.
Discussed at 40:07It can measure qualitative outcomes such as the value of individual interactions, feedback, and whether people become advocates for the project. Surveys and regular check-ins can capture those outcomes, after which the community should iterate based on the feedback.
Discussed at 41:11Note: We understand that names change, people change, and bodies change. We respect each individual's journey and privacy. If you have any concerns about a video or need us to remove content, please don't hesitate to contact us. We will handle your request with care and promptly address any issues.
Published June 13, 2025
Published June 13, 2025
Published June 13, 2025
Published June 13, 2025
Published June 13, 2025
Published June 13, 2025