In this session from DevRelCon Earth 2020, Austin Parker talks about the origins of Deserted Island DevOps and a few lessons learned about what community and events can mean going forward.
Austin: So, hello, everybody. My name is Austin Parker. And today I’m going to talk a little bit about Animal Crossing, a little bit about DevOps, a little bit about running virtual events, and a couple other fun 10-cent academic terms I found while Googling.
So, if you don’t know who I am, I’m the guy that did the Animal Crossing conference, and this is now my personal brand that I must wear around my neck like a millstone for the rest of my DevRel career. I have to gesture this way to the slides. Cool.
So, this is my Animal Crossing character. I think, the resemblance is pretty good, I think. The glasses’ shape is wrong, but, you know, you go to a war with the Animal Crossing accessories that you have and not necessarily the ones that you want. So, let’s go back to March to start off with, and kind of talk about how we got here and what the heck Deserted Island DevOps was.
So, you know, throughout February and March, the world was obviously changing. And the world for DevRel professionals, people who make a living by getting on a plane, and giving a talk, and drinking bad conference coffee was rapidly falling apart in front of our eyes.
At Lightstep, we had actually been having some internal conversations about like, “Hey, it feels like things are going to get bad soon, should we try to do a virtual event?” And I was kind of on the like, “No, I don’t think we can pull it off. I think it would take a lot of work.” That said, I was still, you know, sitting around at home, going a little stir crazy and enjoying quarantine with everyone else.
And like everyone else in the world, I got a copy of Animal Crossing. And wow, that’s a fun way to kill time when you can’t go outside. So, I made this tweet, I put together a little booth in Animal Crossing. You can make your own little designs. So, I made a Lightstep logo. I found those fun…
You see the glasses that look like something that you would probably get at a booth, right? I made this, and I tweeted it out. And a couple people responded. And someone was like, “Oh, we should do a conference at Animal Crossing? Wouldn’t that be fun?” Yeah, it would be fun. It’s amazing what you can do with a website that doesn’t look very good. And I can say that because I made this website.
And a really good, like, gimmick, right? And I think the story of Deserted Island DevOps is… It’s really amazing what you can do if you kind of go for it. So, I put together this website, I actually launched this on April 1st. And one of my takeaways is if you have a really weird idea announce it on April 1st because then if it completely blows up, you can just be like, “Hey. April Fools’.”
But with nothing but this website, no speakers, very little idea of how we will actually do it, and not a lot of, you know, other prep work other than, like, saying, like, “Ah. This is probably possible.” You know, I threw this up, and then the first day I had 100 registrations. Now, that in and of itself was really the indicator idea that “Okay, this could be something that will be successful, right?”
And I think there was a couple of reasons for that. One is that, you know, we kind of started out with a few guidelines. One, this is going to be free, two, you know, this is going to be a community event, right? And that’s an interesting thing to say because when it started, there wasn’t a community.
But I knew there was a larger community outside of any specific thing, right? If you’ve been here since the keynote, you know, Sarah talked about this, there’s this idea almost of a community within a community. There’s a community of, you know, open-source, but then there’s this larger tech community that we’re all sort of a part of.
And so, I thought to myself, like, “Okay, that’s the community I want to talk to,” right? I want to talk to people, and give them something to do, give them something that sort of takes away the sting, I guess, of being stuck at home. So, put the website up, the next 30 days or so was a blur of find, you know, getting speakers, asking some people to speak, actually figuring out how to produce this in, you know, OBS, and Twitch, and Animal Crossing, and setting things up, and doing rehearsals.
And the result was this, right? This was what people saw when they were watching the stream. We streamed the entire event on twitch.tv, used the Animal Crossing camera view to sort of act as a virtual camera. The speakers would come in to Animal Crossing, they would join my game session, and then I would overlay their slides kind of on the top.
If you see where, like, the DevRelCon logo is on the screen right now, the slides would be sort of picture-in-picture in that side of the screen. So, you could still see someone’s avatar in the game world, and you could also see their slides. So, one important thing, you know, no one could actually see the speakers other than what you saw here, right?
As a viewer, you’re only really connecting to everything inside the game world, you’re seeing, like, people’s Animal Crossing characters and not necessarily their real faces, which was actually very interesting. And I’ll talk a little bit about, like, how that, I think, helped later. But let’s talk about the result, right?
We did this one-day Animal Crossing thing. How’d we do? Well, we had over 8,000 unique viewers on Twitch. We had over 11,000 viewers in total. From our YouTube VODs, we’re sitting at 4,800 views right now on all those videos, and more coming in, on a fairly regular basis as people kind of discover it and want to go watch the keynote or something, which is pretty amazing considering we started from nothing.
And I have a theory about why this happened, or why this got to be this kind of moment in time. And to talk about that, I’m going to introduce a academic term, which is called Pace layering. Now, if you’re not familiar with Pace layering, I will briefly explain what it means.
The idea is, and this was formalized by Brian Eno. Yes, that Brian Eno, and Stewart Brand, I believe back in the ’70s or ’80s. And the idea is that society, that everything moves at different rates. And at the top, you know, the top line here is, you have things like fashion, you have very ephemeral kinds of concepts that move fast, and explore, and take risks.
And then below that, you have things like commerce, infrastructure, governance, culture. And at the bottom, nature. Nature moves extremely slowly, right? You’re talking about eons, you’re talking about things like climate and ice cap, polar ice caps, and tectonic plate movements. And then as you move up, these things all move a little bit quicker.
So, culture, you know, culture changes slowly, I think, but less slowly than nature. Governance changes faster than culture, you know, in America, at least every 8 to 10 years, you sort of have these shifts in governance. Above that, you have infrastructure, right? Like, you can walk, you know, take a drive at night, and you’ll see infrastructure changing, right?
We used to have one type of a lamp in a light post, and now we have another one. People have those LED headlights in their cars now, rather than on, you know, the older style which is also related to commerce, right? Commerce also changes, you know, more slowly than fashion, but the basic ideas of commerce, the technologies that sort of feed into this, these are things that are evolving at a more rapid pace than the underlying infrastructure but less slowly than fashion and trends.
And the idea presented through Pace layering is really that this is a way that civilization as we know it becomes resilient to change and shock. In a durable society, each of these levels can operate at its own pace sustained by what’s below it, and then refresh with what’s above it.
Our climate changes, obviously, very slowly, you know, but if we didn’t have climate, if we didn’t have nature, then nothing else up that list will work. And you can see this basics, the same concept is pretty prevalent throughout psychology, throughout sociology, you know.
You can make a reference to, you know, Maslow’s hierarchy of needs onward in terms of how we can conceptualize this in other forms, there is some base that is, you know, must be sufficient in order to do higher-level things. One kind of take away from this is this idea that fast learns and slow remembers. And so, when we’re moving fast, these things are at the hour rim, the outer layer of our layers, that’s where interesting stuff happens.
And we have to remember that while, you know, everyone that’s listening to me, and me, myself, you know, has been impacted by COVID-19, has been impacted by the pandemic. It’s impacting us all differently. And I think that leads to this, you know, this thought experiment, like, “What is it in DevRel itself?”
Like, “What is the actual nature of what we do that moves fast or moves slow?” I think relating to that is that what I’ve learned, at least from doing Deserted Island DevOps, and doing this sort of virtual event, and watching how we’re all changing, is that I think we need to be prepared for something new and almost a new normal in terms of, “What does it mean to be a DevRel? What does it mean to have events?”
So, with that in mind, I want to move on to a couple takeaways about, you know, everything that happened and kind of discuss real lessons learned from doing Deserted Island DevOps. And finally, we’ll come back to Pace layering.
So, keep that in mind. One of the things we did that I think worked really well was the idea of interactive segments. And I think this is something that is generally a great idea for doing any kind of virtual event, but you have to be very careful with moderation, especially if you have a public, public event, right.
We streamed on Twitch. And since it was fairly popular, I mean, I think at one point, we were, like, number two or three in terms of concurrent viewers for, like, Animal Crossing as a category. But keep in mind, Twitch has a very large audience that is not exclusively comprised of developers. So, you will get the internet bleeding in. I erred a little on the side of being overly cautious and had the chat very restricted.
There are things you can do, like followers-only chat. You can restrict people doing hyperlinks. There’s a bunch of different bots that will help you moderate your chat, and you can add in things like, you know, a list of words that get muted. But we actually had a speaker that did, like, an interactive thing where it’s like, you go to the Slido and you can ask question, you know, you type things in, which did have a couple moments where the bad part of the internet decided to go in and put things that were not great.
So, just keep in mind that you want some sort of, either a delay or a two-stage system where things don’t go on to the main stream without having someone make sure that there’s nothing hateful, or rude, or whatever going on in there, right?
I’m not going to say I got the balance right. I think I probably could have biased more towards, like, people chatting more, but, you know, you live and you learn. The second big takeaway is I didn’t expect that this would become as big as it did. But I think one of the reasons it did become kind of a little bit of a hit was because we generally, I took a very hands-off approach to, I don’t know, I don’t want to say leading it, but, I guess, kind of leading it.
So, one great idea. And what you’re seeing a screenshot of here is a watch party, a virtual watch party. And the way these come about is about, like, a week before the conference, someone on Twitter was like, “Hey, is there a way that we can, you know…”
My cohost… Or my host, actually, Katie said, “How are we going to handle Q&A?” And I was like, “Well, people will put them in the Twitch chat.” And then it’s like, “Well, the Twitch chats can scroll very fast.” I’m like, “Oh, that’s a good point.” And we didn’t want to do a Slack because everyone’s on, like, 40 billion different Slacks, including this Slack, which you can use to asking any questions.
And there’s the, how you join it. But everyone’s on like a billion Slacks, right? And there’s actually, in terms of being easy to access, Slack has some difficulties, I think. Discord is another Slack-like alternative, you know, fancy IRC. But one nice thing about it is you can do what’s called an invite link.
And an invite link is just a link that you click, opens in your browser, and then will join you to a Discord server. And Discord uses a federated identity system where you can have a single identity for multiple Discord servers, which you can also kind of do, which kind of works with Slack, but their model is very different, and so you’re having to sign up for every Slack individually.
Whereas in Discord you have one account and then you can join that one account to multiple different things. So, we went in, created a Discord, put that link out there and had, like, 1,000 people, 800 to 1,000 people join it, which was “Whoo.” But in that, people would start saying like, “Oh my gosh. Hey, I’m going to have a multiplayer session.
Does anyone want to come over and, like, do a watch party?” where they’re all kind of sitting in another Animal Crossing Island, watching the stream on their computer, and then, like, interacting with each other in the game. And my take was like, “Okay. Yeah, sure. Here’s a channel. Go.” I would say, not everything needs to be super tightly controlled, right?
Especially in a virtual event where the cost is so low of sort of trying new things. You know, maybe it might not have worked at all. It might’ve been a, you know, it might’ve… And it did probably, it certainly was one of the things that like, “Oh, this could be a minefield,” right? Because all you’re saying is, “Well, you’re participating, so you need to stick to the code of conduct.”
Thankfully, everyone was on their best behavior, and there were no problems. But there’s a balance there. I think given sort of the state of the world, like, it’s better to be a little bit more, you know, to have a bit more, like, bias towards experimentation, bias towards doing something new and different rather than maybe kind of doing things the way that feels safe.
Because if you do nurture that community, it can surprise you, and you can get really interesting stuff like this, which is kind of my biggest takeaway. Like, ultimately, you can’t be afraid of trying something new. What I’ve seen a lot of in sort of in the virtual event space is, yes, we are trying new things but the new things we are trying are mostly trying to replicate sort of our existing model of how a tech event should go and then importing that into a browser.
And I don’t think that can work. This fundamentally, these are different. I can’t see you watching this. And as someone that likes to stand up here, and gesticulate, and, like, make jokes, and kind of look for an audience reaction, like, I can’t get that from you. The two things that I think that we did that were very helpful to sort of ameliorate this was one, all of the speakers were presenting live, but they were all in a Zoom together, and everyone had their camera turned on.
So, when you’re talking, you’re not just talking to your camera, you’re not just talking to, you know, my video, my picture-in-picture here. I can see everyone else I’m talking to in the little Zoom, you know, boxes. And what our speakers said after this was like, “That was such a great experience,” because it felt like you had an audience.
You could actually see someone’s face. You could see the reaction. People were present in a way that you’re never really sure if they are when you’re doing a talk like this. Like, I think everyone’s present. I think everyone’s listening to me. I can get really close to the camera, and maybe make you uncomfortable with how close I am to the camera, but it’s not the same, right? And it will never be the same.
It can’t be the same. Like, we can’t map the experience of going into a room and talking to 5, or 50, or 500, or 5,000 people into me standing behind my desk, talking to you through the internet. We have to operate at a different sense of scale. So, I think that leads into, let’s see…
I told you I was bringing it back. I was bringing it back to Pace layering. So, what changes? And what doesn’t change? What are the things in DevRel, in this career that make it different, right?
What are the things that are going to move fast? What are the things that are going to move slow? Well, I spent some time thinking about this, and I would still like to improve on it. But I think at the top level, at the highest end of this, the thing that changes the fastest is actually our audience. People will fall into and out of events. They will grow in their career.
They will learn things. They will forget things. But ultimately, you know, we are not personalities, we are educators. And when you educate someone, they leave, right? Like, if you teach some… If you’re a teacher, then at some point, your student will surpass you or at least get what they need, and they’ll move on.
So, our audience is changing. And that change in audience does mean that events will…the structure and concept of an event will need to change as well. It will need to absorb shocks and blows. One thing that we don’t talk about a lot is the switch to virtual has been probably amazing in terms of accessibility for events to a wide range of people that either didn’t have the financial ability or didn’t have, you know, the location availability or whatever to actually go to these conferences, right?
Like, if you are in Africa, or you are in India, or you are in a lot of places in the world, like, your access opportunity is going to be significantly different than mine. And while I can go and say, “Hey, let me get on a plane and fly to Europe or fly to California to do a talk, or to attend a conference,” there’s many, many more people that can’t do that.
So, our audience is becoming more broad, thanks to COVID. But that also means that our events need to change to be inclusive of that new and adjusted audience. What underpins all of this is the idea of being an educator, I believe. As a DevRel, I feel like my purpose is to educate and give back to people. And I think those people, that forms that underlying community that I’ve talked about this entire time.
And that is the ultimate base that we need to build on. So, with that in said, thank you for listening to me. If you’d like to read more, I wrote everything up on my blog, aparker.io. You can also read about Pace layering, or you can watch the talks on my YouTube channel, youtube.com/c/oncallmemaybe. Thank you so much, and enjoy the rest of DevRelCon Earth.
Do you need to have a fully formed dev rel metrics framework right from the off? Or can you take a more gradual approach?
Understanding the journey a developer takes through our products is essential if we want them to be successful.