Episode 56
· 32:51
The the system was every engineer was basically an advocate, which is like a great way to start your company. Like, I honestly think most companies, instead of hiring a proper advocate, like, hire engineers who don't mind talking. Get someone who knows your product, like, very intimately, but also can talk about it. Like, that's that's extremely valuable. So think about that when you're you're thinking about your first couple of hires.
Jack Bridger:Hi, everyone. You're listening to Scaling Dev Tools, the show that investigates how dev tools go from 0 to 1. I'm joined today by Brian Douglas, who is also known as b doggy on the Internet, and is the founder of open sourced, the open source intelligence platform. Brian was previously DevX leader at Netlify before becoming director of developer advocacy at GitHub, all the way through to IPO, I think. Not IPO.
Jack Bridger:Sorry. Microsoft acquisition. Acquisition. Yeah. Yeah.
Jack Bridger:And was also on, Showmic's software SnackBytes episode. So thanks to Showmic for introducing us to Brian. Thanks so much for joining, Brian.
Brian Douglas:Yeah. My pleasure. Yeah. I'm super excited to to dig in and talk dev tools and, yeah, my journey through them.
Jack Bridger:Yeah. I mean, you've been there at, like, the early stages of some of the most iconic dev tools. And one of the things you spoke about on the ShowMix podcast was about when you first joined GitHub and you first joined Netlify, and you kind of went in with this plan of 30, 60, 90 of what you were gonna do in the 1st days. And I wonder if you could talk a bit more about what your plan was when you joined and if there's any changes that you would make now.
Brian Douglas:Yeah. Yep. You know, in hindsight, always always thinks you'd change, but also it you wonder if, like, if you change something else, would there would the journey be different, or would you even be at this place? Because, like, at Netlify, so I I joined as employee number 3. I was a long time a year I spent as a customer.
Brian Douglas:Well, honestly, I was a user because they didn't have a paid plan. But I was just a power user and had a blog and a podcast, because I yeah. Just wanted to host my myself on the Internet and talk about stuff that I was working on. And, so in that experience in the I guess, what it was an interview, but it was like a coffee chat really. They pitched me on joining the team based on my background in sales, but also my years as an engineer, to be, like, one of the first non I was, like, the first non head of, in engineering person.
Brian Douglas:So, like, they had a head of infrastructure, they had a CTO, and then there was me. And I came in this to do some front end work. Ironically, I interviewed in Go, but then got a JavaScript job. But I digress. 36 to 90, like, in that conversation, they're like, hey.
Brian Douglas:What would you how would you make impact if you if you join our team? And I was like, oh, let me just, give you a 30, 60, 90. And they kinda were taken aback because it's more of a a sales plan or a business development plan than the engineering. But my whole thing was that Netlify was doing something very different. They were taking what normally you would do on s 3, and they were providing a a better experience.
Brian Douglas:We'd have to zip your project and upload it to AWS console, which was like a behemoth that was hard to use back in the day. And within the 1st 30 days, my my pitch was like, first, you know, make the product good. Like, we're the the plan was to move to react at that point from Angular, and then add some other cool features and onboarding tools. And then within the 1st 60 days, my pitch to them was, like, go to every single boot camp and teach those boot camp students how to build their portfolio. Because it's low effort, you just build a template, and then you ship it on Netlify.
Brian Douglas:And the goal is within the first 36 30 seconds to a minute of you hearing the name Netlify, you should have a site up and running. Because, again, like, we're just, like, closing the gap on things like s 3 and GitHub Pages, and we were just making it a lot easier. So it was possible. So when I joined, I did just that. Like, I would go do a conference talk at a in a city.
Brian Douglas:So I I mapped out all NFL cities. So there's 32 NFL teams in the US. If there's a team in the NFL, it means it's a big enough market, which means there's probably a university close by, and there's probably a boot camp close by as well. So, the goal was, like, go to the city, meet boot camp founders, meet college professors and CS programs, and teach those students how to use Netlify. And then whenever I would do that, I'd leave with a curriculum of, like, hey.
Brian Douglas:If you wanna embed this, it's just a a standard URL, where you can take the markdown. It's also a GitHub repo, and you can embed this into your curriculum to teach all your students how to use build a portfolio. And to success, it worked. So we would grow. By the time I left, I think we're approaching, like, 4 or 500,000 users.
Brian Douglas:I don't know the exact number because my whole thing is, like, I don't wanna pay attention to number and then over obsess over it. So I never really tracked numbers that way, but every now and then, like, we'd have an all hands and they would mention it. But, yeah, that was, like, that was my 30, 60, 90. It's, like, get attached to the actual funnel of where developers are coming in. And then 3, 4 years from then, like, everyone you see is in Netlify, which became the actually, it became very realistic because I like, some random people I just run into, it was like, hey.
Brian Douglas:I'm wondering how to code. Here's my portfolio. And it's like a Netlify dot app. And I'm like, oh, it worked.
Jack Bridger:Could you actually, like, tell us a bit about, like so you kind of you joined, and you were like, okay. Day 1 of actually getting boot camps to or going to these cities, like, what what was it actually like on the kind of, like, implementation of that strategy?
Brian Douglas:Yeah. So it it was, when I joined, I really was just engineering. I we were just trying to build the application that was working, and we shipped this feature called deploy previews, which is embeds into your GitHub PR, and it'll give you a link, to be able to preview your Netlify deploy. It was pretty novel. It wasn't we weren't the first people to do this.
Brian Douglas:Like, Heroku was actually the first company to do it. They got a Heroku pipelines, but Heroku was like, we don't wanna embed in the GitHub, like, come to come to Heroku. So they did it, like, kinda clunky, but Netlify is, like, distribution. 3 at that point, it was, like, 5 people at the company. So let's just embed in GitHub and do, like, the GitHub way.
Brian Douglas:And it worked out that the CTO previously worked at GitHub and had a very deep understanding of the GitHub API, which is how I learned how to use the GitHub API pretty well is what my and why do you open soft today? But, yeah, what I'm getting at is, like, the first 3 months to 4, like, that summer was just really just, like, building the ground game of, like, makes your product good. I make sure people can use it, make sure that it's discoverable. And then that that summer I joined, create react app came out. It's create react app is like the, CLI that the team created internally at meta, but it with PMA community project still exists where you give this, like, CLI tool to create a react app.
Brian Douglas:And, because that came out, you could create react app, a new React app, and then you could Netlify deploy all from the command line. So you could deploy a Netlify site in 30 seconds basically. So I wrote the blog post 30 seconds to deploy, React, and it actually did pretty well. It got on Hacker News. Hacker News, like, the create react that got a Hacker News, but then we, like, were fast followers of it because we were just, not planning to be it.
Brian Douglas:Like, it's I just thought it was cool because everyone was like, Rob, Rob, I create React app. So I built that sit well, I didn't build anything. I just wrote the blog post. Like, here's the guide of how to do it. It was, like, 500 words.
Brian Douglas:And, so, like, once we set up the structure of, like, the product work, then people could discover it. At that point, I've there's a lot of boot camps in San Francisco. So I would just talk to all the boot camps I knew because previously, I worked at a boot camp, building out their dashboard, previously, it's block that IO, which is now Thinkful. The knowing boot camp leaders from that experience, I would just go to boot camps in San Francisco first and test out. Hey.
Brian Douglas:Would you like this curriculum? Like, would you like basically this your URL, use it for your, like, LinkedIn and your resources? You don't have to rewrite anything, like, we'll maintain it, but link in any resource how to build your portfolio, and I'll be using that offline. Like, we'd sponsor, you know well, actually, we didn't sponsor anything. I would just show up and do, like, a lunch and learn, basically.
Brian Douglas:And that was it. And like folks would just love me forever because I came to talk to the students about how I transitioned my career into being full time engineering, but also while doing that convert that that talk, I would say, hey, here's Netlify, Here's how to use it. And it became like the the perfect situation to do that. So then my goal when I joined Netlify is, like, I just wanted to do a conference talk because I just wanted to talk about just something, like, get on stage one day, and, I did that. And then I would send that same CFP to other NFL cities.
Brian Douglas:And every time I got accepted, then I would then the plan was, like, 4 ideas, but 3 touch points. So at least either 2 boot boot camps if it was, like, Chicago had, like, 3 or 4 of them. So I spoke to, like, 3 of them there. I went to a conference. I also spoke at a meetup, and I would call up the meetup organizers and be like, hey.
Brian Douglas:I saw you on meetup.com. I'm gonna be in the city. I know your meetups on this day, but I'll be here on these days. Like, do you wanna like, we'll sponsor a pizza from Netlify, which is, like, what, $500 or whatever. And I'll come and speak about deploying React apps and building our portfolio, in 30 seconds.
Brian Douglas:So that was the that was the game, like, just doing that. It it took about probably 6 months before I started traveling to do that, and I enjoyed it. I enjoyed, like, getting on a plane and being super cheap, staying on Airbnbs and stuff like that and really having no budget. But conferences would pay for your travel if you got accepted. So that was sort of how we got to got to kinda see this thing grow.
Jack Bridger:Yeah. That's amazing. It's like such a yeah. Kind of there's a lot of hustle there. If you were I mean, I guess, like, DevTools listening, they might be thinking like, oh, cool.
Jack Bridger:Let's just go start hitting up all the boot camps and tell them to to not tell them, but, you know, can can I, you know, to show you how to use our tool? But I imagine that may not work as well now because it's like you it was like because it was a new strategy.
Brian Douglas:Yeah. And I'd say boot camps are still, like, they hit their pinnacle point. Well, now it's like there's less boot camps. Well, there's there's a lot of boot camps, but I don't think it's as vibrant of a space. Like, definitely go talk to boot camp students and get them to use your stuff.
Brian Douglas:So that way they become experts in 4 years, and then they're they embedded into whatever a future company is. But, like, you could take that same model and go to live streamers on Twitch or YouTubers on YouTube and say, hey, you've got a community or you have a Discord, but, like, hey, we built a Discord integration built on Replit and, it uses this one API, which is our API, but, like, we're to show people how to build Discord bots to do x y z. And you go shop that around, which is like it's business development. It's like sales 101. It's like go figure out whatever the repeatable way to knock on doors.
Brian Douglas:And it's like the it's the the law of averages. Like, the more doors you knock on, the more chances that people will say yes, and then more opportunity you have when that content lives forever within the community. And that's it. The one thing I I learned at so GitHub, my, my 30, 60, 90 for GitHub was actually and I have very similar was my which we're trying to empower marketplace integrations, and build on top of the API GraphQL. I only did that for, like, probably the 1st 6 months and then we went directly to get up actions because get up action shipped and that was like the better solution.
Brian Douglas:But it was this it was the same deal, but the difference was I was going to communities and that GitHub explicitly does not have the GitHub community. Like, there's no GitHub Discord or there's a GitHub Discord, but it's not maintained by GitHub. There's, like, the GitHub Reddit, but it's not maintained by GitHub. But there's, like, all these other avenues that are not GitHub, like JavaScript and TypeScript and Rust and React. So explicitly, it was just go become best friends with all the maintainers of all these projects and all these communities and make sure they knew who to contact when they had a question about GitHub.
Brian Douglas:And that was the goal. It's like we go talk to these folks, but it was the same thing as most dev tool startups, like build up the Rolodex. So that way whenever you need a speaker at a conference or we need to, like, do a one off, you know, interaction for whatever feature we're sort of testing in whatever community is, like, build up that that muscle of, hey. We're shipping a feature. Could you give us, like, 15 minutes to, like, a quick demo?
Brian Douglas:And, also, could you implement this so we can talk about you at GitHub Universe? So, like, at the point where GitHub Actions was shipping in 2019 well, sorry. GitHub Actions shipped in 2018. But 2019 is when GitHub Action shipped CI, so we could do automate build tooling, DevOps, like, proper DevOps inside of GitHub actions. My the the goal was to get all the top projects in open source using GitHub actions.
Brian Douglas:And the question was, like, what are the top projects? And And it was funny. It was, like, not funny. It was, like, ironic that most people didn't know what the top projects would get out of it. Why?
Brian Douglas:And the answer was never stars internally, and it's also never externally. Stars is not the metric, for growth, because, like, jQuery has 50,000 stars, but jQuery is a maintenance bone. So, like, do you need to reach out to jQuery now? So we've been building the algorithm internally, for identifying what the top 100 projects were. And then we had to figure out, okay, how do we make impact really quickly about something that's brand new and people don't have a lot of context on?
Brian Douglas:So we chose JavaScript as the as a language because JavaScript top language was easy easy to see that in the data that we had. Everyone can see that data. So we found the top 25 JavaScripts for the top 100 list. It's really top 500, but the top 100, we'll call it that. And for the DevRel team, for me, because at the point we had 2 developer advocates on the team at GitHub up until, I think 2020.
Brian Douglas:Yeah. 2020 was when we actually hired. We absorbed some folks from Microsoft, but, basically, 2 people and the goal was, like, how do you get GitHub actions leveraged in these popular open source projects? Yeah. So we had the 25 and then all I did was, message, DM people and message people on Twitter.
Brian Douglas:Hey, get a back to his shipping CI. Do you wanna test it? Let me know if we can jump on a call. And, by the by Friday so I was like I mean, it might have been, like, Friday to Friday, but by Friday, we had 20 of the 25 using GitHub actions in some part of the the organization.
Jack Bridger:Yeah. That's amazing. And this it sounds like it's like a lot of things that GitHub is so already people are using it, like, a core part of it at least. And they know about GitHub, but they're not necessarily, like, using all the things that you want. And and so you're kind of building that Yeah.
Jack Bridger:Relationship and making sure they they know all the stuff that you can be useful. Yeah. I GitHub had, did
Brian Douglas:a great job in sort of the bottom up strategy for adoption. So the 1st 4 years, GitHub took no no VC money at all. Like, they were bootstrapping it and we're building out the product for the $5.5 a month for private repos for the longest time. Yeah. Up until 2012, I think, was our first first round of funding or first institutional.
Brian Douglas:Might have been a little later than that, to be honest. But so they had such good adoption and everyone just knew GitHub because they just they were just at a a good time in the market to convince people to use Git, and use GitHub. So the the system was every engineer was basically an advocate. So I was the first proper developer advocate hired in 2018. Prior to that, like, if an engineer loved doing open source and they were hired, then you'd ask that engineer, hey, can you talk to the rails community about this thing we're working on?
Brian Douglas:It was always like word-of-mouth, which is like a great way to start your company. Like, I honestly think most companies, instead of hiring a proper advocate, like, hire engineers that don't mind talking, like, hire engineers that don't mind writing a blog post because that's that's the hardest thing. It's like, get someone who knows your product, like, very intimately, but also can talk about it. And that's what I did at Netlify. And that's why my career sort of shifted into this this space.
Brian Douglas:But it's the same thing, like, when you hire your first engineer as well, like, hire engineers that could be product managers. It's because, like, adding another product manager when you have a burn rate and you're taking on capital or you're bootstrapping, like, that's an extra amount to feed basically. But if you hire an engineer that really intimately knows your product and could be product manager and can, like, triage issues and can ask questions to users, like, that's that's extremely valuable. So, think about that. Like, I said, your listeners think about that when you're you're thinking about your first couple of hires.
Brian Douglas:But because that was way GitHub was built, there was tons of people and, like, no rolodex, like, no idea of who to talk to. And as GitHub becomes mature, it gets acquired by Microsoft, people leave. So it ended up being I was there before Microsoft, but I was able to meet all the people who knew all the knew all the people and and had, like, built a network because that was my job was, like, become friends for everyone in GitHub, outside of GitHub. And so and in some ways, I was kind of the last man standing. Like, there was a lot of people that were still there for years, but, like, I'm not gonna reach out to Mike McQuaid and be like, hey, Mike.
Brian Douglas:Can you intro me? Did the like, he would do that, but, like, Mike doesn't wanna do DevRel or, you know, yeah. Anyway, I could name drop a bunch of other people, like, not everyone wants to be an advocate or do DevRel and send emails and plan conferences and organize meetups to boot camps and and some lunch and learns. But I was the weird one that didn't mind doing that and shipping code. So, I'm sorry.
Brian Douglas:I'm, like, trailing off. Basically, the yeah. I already forgot the original thought that I was actually going down.
Jack Bridger:I think it was just the whole, like, 30, 60, 90. Right? Was it without I think that was the main we just Yeah. We were
Brian Douglas:on the same thread. Yeah.
Jack Bridger:Yeah. I think we were just on that thread. Yeah.
Brian Douglas:But, yeah, I mean, it at the end of the day, it's like it's the the thing that I I always take it back on because I do a lot of I do a podcast as well, and I do a lot of angel investing. And my whole thing is, like, you have to for getting adoption for developers doing that bottom up strategy, which I recommend for a lot of folks, you kinda have to talk about the vision and, like, where this thing is going and, like, what are the possibilities. So, like, even when before we shipped CI for GitHub actions, I remember speaking at Jamstack NYC and talking about how you could build CI on GitHub actions. And I was at actually in Barcelona at full stack fest, and I was talking about how you could build CI with GitHub actions. And then then we shipped CI with GitHub actions.
Brian Douglas:But it was like, I was doing showing the hard way to do it to show, like, the pain, show, like, what's possible because, like, end of day, like, CI existed on GitLab. CLI existed in Circle and and Jenkins. But the challenge is, like, no one built CI in a way that was, like, truly open and that you could share across ecosystems. I guess, get that. I've never used get help lab c l CLI but or sorry, CI.
Brian Douglas:But, I think they didn't have an open, version. But building the ubiquitous, ubiquity ubiquity. You anyway, basically, building the standard for okay. Now you can do CI, like, we've basically, frankly, I don't work at GitHub anymore, but, frankly, what GitHub did is they commoditize CI and CD to the point where, like, if you're a CICD company today in 2023, like, you're like, you're a you're more of a bespoke or a niche product and you have to build this like the thing on top of it. And, and I think it's the same thing like no one's building orchestration software, unless you're building on top of Kubernetes because Kubernetes is a standard.
Brian Douglas:So, like, why would you go build that again when Kubernetes is good enough? And, like, I think we have to get a little more time away to figure out what the next thing is, when it comes to orchestration but and maybe that's coming maybe someone already has something out there. But, yeah. And then the day it's like going back to like my original mention and I think I what I was talking about is like being the sort of advocate, the first advocate of GitHub is like figure how to scale yourself in a way that you don't have to be the person always doing the conversations and talking.
Jack Bridger:Yeah. I think you've spoken a lot about creating advocates or kind of identifying Yeah. Advocates in the community. Yeah.
Brian Douglas:The number number one job for developer advocate is to create other advocates for your product. So, like, if you're all if you're the only one that can walk in the room and talk about that product because you're the only one that cares, then you you're not doing a good job. Like, it's gotta be at a point where you could take a step back and watch the machine start. And that's what I was doing with the whole boot camp thing. It's like, everyone got the moment of, like, oh, wow.
Brian Douglas:I thought I had to, like, sign up for s 3 and, like, configure the AWS CLI and, like, do all this, like, random stuff. And now you get this, like, connect your GitHub account to Netlify and you deploy, like, in 30 seconds and you're good. And then, like, once you once you figure out that base layer of, like, what people what you just the pain point you solve, now everyone gets to go solve harder problems. So now it's like the next thing was like, oh, how do you do headless CMSs? How do you connect my WordPress to a Netlify site?
Brian Douglas:Or how do I do, like, background jobs? Or how do I like, all that those problems are the next things that Netlify ended up solving later on. But, like, we were the move past that pretty quickly. And I think, like, even when I'm working on with open sauce, like, the challenge that I've been trying to solve with the intelligence around open sources, a lot of people like this, this is 2 completely separate things that people are working on, which is I'm a developer. I wanna do open source, so I'm gonna, like, go do Hacktoberfest or I'm gonna go connect my profile to this thing that's gonna recommend any places to go contribute to.
Brian Douglas:But then there's also, I'm a maintainer. I can't sustain this project because I need funding. I need people to help contribute with me. Like, how do I even find people to contribute alongside of me? And, like, very few times people actually cross and say, oh, maybe I should go talk to maintainers and help them out, or maybe I should go talk to contributors and help them out, but no one's like, oh, let me just create an ecosystem where everyone can just hang out and talk.
Brian Douglas:And, I see countless amount of, like, projects that do either or and don't have, like, the sort of middle ground to for discovery. And I think the challenge is most people think GitHub's gonna ship this. And GitHub has shipped a lot of these pieces, but there's no, like, story to, like, really make this thing work.
Jack Bridger:Yeah. Like, I guess the only thing is, like, the trending sort of stuff that you can see popular projects. But on the other side of it, I don't know.
Brian Douglas:Yeah. And the trending is the same problem for stars right now because the trending it's yeah. If you ship a a 1,000 commits so, like, if you don't squash your commits, it's a lot easier to be trending. And if you do a 100 automated PRs, it's a lot easier to be trending. But how do you how do you distinguish like, okay, who's this automating their life away to, like, make sure PRs are released all the time, and who's like either squash and commits?
Brian Douglas:Like, what's the what's the story behind these contributions and why these folks top 10? And I think there needs to be a little bit more of a shake up around, like, okay, what is actually trending and what what is what is this like on the list all the time? So, like, sort everything from a 100, top 100 star projects. Like, I think number 3 is, like, 99 ICU, which is, like, a a massive markdown for free education and books. Like, it's why is that the top 10 repo on GitHub?
Brian Douglas:Like, it's it's a lot of data and a lot of books which is great, but it's just like a giant Dropbox folder for someone that drop a bunch of PDFs. But, like, that's and then there's, like, within the top 100, like, there's tons of, like, JavaScript interview questions and here's, this interview question and it's, which is great, like, they belong on GitHub, but, like, when we, like, that's not a place that I'm, like, I'm not looking for that maintainer because that main hitters, like, actively not participating. Like, people are just uploading and they have other people doing stuff. So and that's what we had to do with the GitHub. It's like we had to, like, remove those projects because they're not gonna embed actions and CI and a bunch of other stuff in their projects.
Brian Douglas:So, like, I guess, we basically have to think rethink on how we're thinking about the ecosystem and, like, what success looks like.
Jack Bridger:Yeah. I I mean, I completely agree because, just on, like, on the podcast and some other stuff I've done, like, I've sometimes tried to look, like, what beyond the noise on Twitter and stuff, like, which projects are actually, like, getting used a lot. And then sometimes you you find projects that have loads of gallop stars, but there's not many, like, issues created or, like, things are, like, kind of outdated or, you know, this is it's, like, a lot more complex than
Brian Douglas:Yeah. There's a a lot more nuance, yeah, when trying to find out, is this thing still maintained? Like, just ask that question and, like, how do you answer that? Like and then, like, who's maintaining this? Like, a lot of times that isn't a question that you wanna ask, but, like, it's not very clear until you, like, click a bunch of tabs or open a bunch of PRs and see who's approved, who's purged.
Brian Douglas:I'll look at a commit history like that becomes more challenging. Like, we're just trying to solve that. I answer that question sooner, with open sauce. And, like, another like, when you talk about noise, especially with dev tools because, I see this all the time when I I talk to somebody and they're like, hey. Help me out with my DevRel strategy.
Brian Douglas:And I honestly, I my DMs are open. I'm like, I I love talking to early stage companies or just dev tools, open source projects about, like, how do you how do you get started in reaching out to community. And, consistently, everyone's they they bring up their Slack channel, and like, oh, yeah. We have 2,000 or 7,000 people in Slack. And I was like, okay.
Brian Douglas:Cool. Yeah. What does that mean? Like, yeah. All these people love our product.
Brian Douglas:And I was like, are you sure? Like because, like, it's just 7,000 people in a Slack channel. It's like that it is a lot of noise. And that what that tells me is, like, that's a lot of time. And I can probably go to well, I benefit because I'm here in San Francisco, but I can go to meet up and, like, ask somebody, hey.
Brian Douglas:Have you heard about this thing? But, god, no. They go, okay. Well, they have 7,000 people in their Slack. It was like, oh, yeah.
Brian Douglas:But I'm not in there. So, like, what does that mean to me? But then that's also 7,000 people that are probably not paying for the thing and they're getting free support and help, and, like, feature requests and all this other stuff that how do you how do you actually sift through all that noise? In reality, like, it'd be so much easier just to funnel those folks into the comment thread of a blog post, or funnel that to your Twitter instead and like that become the community because at least that signal it's SEO driven or that signal is like discoverable and you could use that that actual signal of, okay, 7,000 people on Slack, but how many active commenters? And like a lot of times what this is is like when you sign up for their their thing, they send an email, it's like join our Slack or you automatically get invited to the Slack, which is cool because, like, I get what I so every time I'll get, like, 8,000 page views, unique page views because of something that went out.
Brian Douglas:But we don't get 8,000 people sign up for open sauce. So, like, how do you actually see what engagement looks like? And, which is why I put I push everyone to I do have a Discord. So, like, I I'm also part of the problem, but but it's, like, knowing what to do with that energy. Like, how do you convert that energy into, oh, 7,000 people in my Slack, every time we open a stack over for a question, it's answered within 30 seconds.
Brian Douglas:Or every time we we ship a blog post or a tweet, we get x, like, 10% click throughs or we get a 100,000 impressions because of, like, our army of Slack users. Like, how are you leveraging that to get the word out about your product? It's usually the question I ask, like, cool, 7,000? Like, what are you doing with that?
Jack Bridger:Yeah. So it's like going beyond that kind of, like it's almost like a vanity one. It's like if you just go
Brian Douglas:It's like stars. Like, having 7,000 stars in a repo is good, but, like, what what are you doing with this? Like, how do you if you're a product that makes money, like, how do you cash that in? Yeah. There's there's a project, a popular project today, and they're onboarding for they had a guide.
Brian Douglas:And when he went through the onboarding, part of that onboarding guide is, like, you authenticate with GitHub, and they had a webhook that would also start a repo after you onboarded. So, like, they were actually, they were early in this, sorry. I don't wanna I called it out directly, but they were early in a movement. Like, they're a new way to interact with databases. And because everyone wanted to learn how to do that thing, they had, like, a how to x y z, tutorial, and they went to that tutorial.
Brian Douglas:You authenticate to, like, go build the thing you need to build, and then you would start a repo at the same time.
Jack Bridger:Oh, so they they told you, like, we're gonna teach you how to force people to start a repo.
Brian Douglas:No. No. They they would teach you how to use the the the technology.
Jack Bridger:Oh, okay.
Brian Douglas:Yeah. But when you would use the technology, you would create a repo, but you'd also star their original repo. Yeah.
Jack Bridger:Yeah. That sounds, like a
Brian Douglas:yeah. Yeah. It's it's it's stars could be gained. So, like, I guess, if you have over a 1000 stars, you're relevant. So, like like, the it's a top 1% of all GitHub has over 1,000 stars.
Brian Douglas:So, like, at that point, once you get a 1,000, it doesn't even matter. Like, unless your top 3 projects on GitHub, yeah, it doesn't matter because, like, most most stars end up petering out around 50,000 today. I imagine in a couple years, it'll be a 100,000 because of inflation. Inflation being, like, more people joining GitHub, more people learning how to code. Printing stuff.
Brian Douglas:Yeah. Yeah. Stars is just like it's not the thing to focus on.
Jack Bridger:Yeah. Yeah. Yeah. That's really good to hear. Do you have, like, things that do you have, like, a kind of a ranking of projects that I I've played around.
Jack Bridger:I've seen, like, you do, don't you? You have, like you can see, like, the hottest kind of
Brian Douglas:Yeah. So we have hot open sauce pizza, which is it's not ranked by stars. It's ranked by most, like, most recently created. So, like, usually usually, if you get, like, a 100 stars in a day, there's, like, a good signal that something's happening, and that's, like, a better metric. It's, like, instead of having a 100,000 stars, like, how many stars did you get today?
Jack Bridger:And if
Brian Douglas:you have a 100,000 stars and you got 0 stars for the past 3 months, then it's a good signal no one cares. But if you're constantly getting, like, a 100 stars each day, like, there's a good signal. Okay. There's something happening here or there's something if they're doing a good job at marketing or they paid a lot of money to get folks to start a repo. But, like, that's a better signal, and that's, like, the stuff that we're we're actually looking at.
Brian Douglas:We're not exposing this today, but we're trying to figure out, rebuild what we what I was using internally at GitHub. And we we just recently closed around the funding, which gives us time to, bring on another staff member to help engineer with us and help we're gonna start discovering some of this stuff pretty soon.
Jack Bridger:Yeah. Amazing. That's very exciting. I think that's what we've got time for, Brian. But is there anything if there was one thing that you would want, like, a DevTool founder listening someone working at dev tools to take away from this conversation?
Brian Douglas:Yeah. I would say, number 1, build a good product. And then number 2, please talk about it. Like, that's I think, easily number 2 is where people fall off.
Jack Bridger:Amazing. And where can people learn more about you and about, open sourced?
Brian Douglas:We have a a newsletter, sauce.ghost.io. It's a scroll to the bottom of opensauce.pizza, and you can subscribe. We'd love more people on our newsletter talking about our product. And, you find me b dougie, b dougie yo on Twitter, b dougie on GitHub. And, if you at mention me in the issue, I tend to respond to you as well.
Brian Douglas:So, like, use GitHub to to reach out. Let me let me look at some stuff.
Jack Bridger:Amazing. Thanks for joining Brian. My pleasure. And thanks everyone for listening.
Listen to Scaling DevTools using one of many popular podcasting apps or directories.
