In this episode of Semaphore Uncut, I chat with Alex Ellis, creator of OpenFaaS, Inlets, and Arkade. He shares his experience of following his open-source passion while also making his work pay.
Key takeaways:
- When an open-source project has traction and community, but no business model
- The sponsored app business model
- The Freemium business model
- Entrepreneurs have to take blows and get up again
- Your own open source path is possible
Listen to our entire conversation above, and check out my favorite parts in the episode highlights!
You can also get Semaphore Uncut on Apple Podcasts, Spotify, Google Podcasts, Stitcher, and more.
Like this episode? Be sure to leave a ⭐️⭐️⭐️⭐️⭐️ review on the podcast player of your choice and share it with your friends.
Edited transcript
Darko (00:00): Hello, and welcome to Semaphore Uncut, a podcast for developers about building great products. Today I’m excited to welcome Alex Ellis. Alex, thank you so much for joining us. Can you give us an intro about you and your work?
Alex: If you asked me in 2017, I’d say I’d created an open-source project to make serverless functions portable, to run on your own infrastructure. It got to the point where people are telling me to quit my job, and focus on it full time. And I’m looking at a few offers, and I’m not quite sure what the best option is for me. I took a balance between compensation and retaining the ownership and the direction of this OpenFaaS project that’s been created.
By 2019, I am back out in the world again. I don’t have a good option to work on OpenFaaS full-time. There’s just no job available for that. The users of the project don’t pay the company, don’t pay the project, don’t pay me anything. So it’s kind of difficult.
So, over the last 18 months up to now, I’ve been helping companies, almost flat pack assemble communities and partnering with companies like Equinix Metal and Rancher to help them really engage and reach and excite developers with technical marketing, with engineering, with podcasts and webinar presence. On one level, I’m not working full-time on OpenFaaS, I am actually spending a considerable amount of time helping other companies. However, I still get to work on it and with a significant amount of my time. That’s kind of where we are today.
When an open-source project has traction and community, but no business model
Darko (03:12): What was the motivation for the OpenFaaS, and do you have some concrete plans where it should go and how it ended up?
Alex: One of the things that are a constant struggle, and I’m sure people are tired of hearing about this now, is that open source is not a business model. There is no way to do open source and get paid for it.
So, in 2020, we’ve done some custom development work within the OpenFaaS community, hired some of the community as consultants to come in, and sort of stretch what we can do in quite a small company but it’s not quite enough. So I’ve been doing business development and figuring out what kind of features wouldn’t compete with the community but would make sense for a company using OpenFaaS in production. One of those features is SSO. You go there, and you bring your own OKTA, you configure it with OpenFaaS, pay for a license, and then you can invite your team, and it just upgrades the authentication for you.
Going forward into 2021, there’s a bit of a juncture here. It’s either, we wind down OpenFaaS, we close down the project, and that’s it and just say: we tried it for four years, couldn’t get any money out of it, and users will have to figure out an alternative.
That’s something that over the last four years, almost every month, I’ve been coming back to. It has a really energizing community around it. There are many different open-source users and end-users at commercial companies. But I’m not a charity worker, and neither is the community; we need some way of getting this to the point where it can pay for itself.
Arkade: the sponsored app business model for open source
One of the things that I particularly enjoyed is reading Getting Real by Basecamp. The author talks about starting with a really small thing that can deliver value that you can sell and build upon. And I have seen a resurgence or even a kind of a subculture of micro-SaaS indie developers, and I don’t know if that’s what I want to be – if that could pay for OpenFaaS, the project, and community. However, I’m interested in it. And because I put this business head-on, it’s allowed me to go on and try everything.
I saw an opportunity, a pain point, and developed a solution for it. There was a lot of repetition in tutorials I was writing. I was doing technical marketing or engineering as marketing, I would call it, for these companies, having to repeat 10 commands to install cert-manager, three commands for OpenFaaS, five for Nginx, 20 for Istio.
So Arkade started in the same vein as OpenFaaS, this is not going to be sustainable and, okay, then it will be too big to work on. I floated the idea of sponsored apps: a company like a Semaphore, a Sysdig, a Venafi that has an agent or a commercial product that’s not open source, would pay us an amount to create an app for them to write a blog post promoting it, and then to support it for 12 months. And I signed the first customer last year. You just have to try different things and see where there’s an opportunity.
Inlets: The Freemium business model for open source
I created another project called Inlets. And this one was because of that company I was working at in 2018, we needed to get webhooks. You may turn to something like Ngrok. Well, in enterprise companies Ngrok is always banned. You can’t use it.
I thought, what can I do? And I thought, well if it can go into the HTTP proxy, it should just work. So that’s what we did, built a small prototype with a web socket. I was able to get my Raspberry PI taking photos of the window on the internet, through a portal, all through wifi. And goodness knows how many routers and hops, and it didn’t open any ports.
That project, Inlets, people were really keen on it. It’s got 7,000 GitHub stars now. I got to the point where I wanted some more features, and I knew it was going to take me two, three weeks to build it. And I thought I’m not giving away three weeks of labor for nothing. And so I just didn’t, I made it closed-source, and then called it Inlets Pro. It had TCP capabilities. You can use it as a VPN. You don’t even have to expose the public end of it.
And I launched it, and pretty much most of the sales have actually been developers so far, but it’s still early. What’s kept the community there, even with the closed source parts is that, whether you use the open-source tunnel or the pro one, there’s an integration into Kubernetes that doesn’t care. It now supports maybe 10 different public clouds and IS platforms. And I just wrote the first two to show people how it was done.
Entrepreneurs have to take blows and get up again
I remember chatting to Jamie Dobson from Container Solutions when COVID hit, and consulting contracts got canceled with no notice. He said if you want to be an entrepreneur, if you want to get into the arena, you’ve got to get some blows, and you’ve got to be used to taking them. And it’s just how many times are you going to get up from that. And the people that succeed are the ones that get up one more time. I found that quite inspirational. Even people who outwardly may be succeeding and have some revenue coming in probably don’t feel much different to me.
And so, yeah, it’s getting into that mindset of, I don’t just want to be a conventionally-minded process company man. I want to create stuff, build a community, help companies. I want to bring other people into this and do interesting things.
And I think also you need to stay in your own swim lane. The body and the mind have a natural state that it gets dragged off to. It’s like a reflex or an instinct, fear of missing out, wondering if other people are doing things better if they’re making more money. And it’s an exercise to keep bringing yourself back to the center.
Your own open source path is possible
Darko (26:41): It was a super interesting conversation, we went deep on many subjects. Some people in their careers are reaching a certain threshold, and they want to figure out what’s next for them. So I would say that we can maybe encourage people to start their own thinking. It is doable. You want to beat the struggles.
Alex: The lifestyle is doable. There’s not some point when you become a principal level engineer that you can suddenly go off on your own. I think it’s more the leaning that if you feel frustrated with the technologies that you’re working with, if you feel like under a shroud or a shadow of past innovation and the way we always used to do things or that you don’t have a voice, just have a think about whether you’ve got the support network and whether you’ve got the sort of energy to go in and do something on your own.
It could be that you just become a contractor, and you just go between companies doing interesting stuff. Or you might like the path that I’ve taken, which is, I know that there’s an opportunity for so many companies now in the Cloud Native Landscape that need to connect to developers and struggle, not because they don’t have enough money, but perhaps they’re not doing enough of the right things.
And I know that was a good move because that business has sold. We’ve got a pretty reasonable client list after 18 months. And so I think if you’ve got some skills, if there’s something that you can do and that you enjoy, and that makes sense to you, you can package that up, and you can sell it, and you can do okay.
Darko (29:16): Yeah. Let’s wrap up with that. That’s a nice flavor of the whole conversation. Thank you so much. Whether it will be OpenFaaS or something else, I’m sure there’ll be a success at the end of the road. Thank you. Good luck.
