Latest / The Bootstrapped Tech Founder with Fexingo: Profitable Software Companies Without VC / Bootstrapped Founders Building on Open Source Foundations
Transcript
- Lucas: So, you're a bootstrapped founder. You've got an idea for a productivity tool, and you need to ship fast. You could write everything from scratch — authentication, file storage, real-time sync. Or you could stand on the shoulders of open source projects and just wire them together. Luna: Which is what most of us do, honestly. I don't think I've ever started a project without at least three open source dependencies. Lucas: Right. But here's the question: if you build your whole business on an open source foundation, what happens when that foundation changes? Or when your success makes you a target? Today I want to look at how bootstrapped founders can use open source strategically — not just as free code, but as a deliberate part of their business model. Luna: And I'm guessing there's a cautionary tale or two. Lucas: There is. Let's start with the best case. JetBrains — the company behind IntelliJ IDEA, PyCharm, ReSharper. They're famously bootstrapped. No VC money. Billion-dollar valuation. And their entire product line is built on top of an open source platform. Luna: Wait, IntelliJ is open source? I thought it was a paid product. Lucas: The community edition is open source, under the Apache 2.0 license. That's the foundation. JetBrains then builds proprietary features on top — the advanced debugging, profilers, database tools, framework support. They call it 'open core.' The core is free, the value-add is paid. Luna: So they get the distribution and community contributions for free, and they monetize the stuff that enterprise users will pay for. That's clever. Lucas: Exactly. And it works because the open source core is genuinely useful on its own. Developers download IntelliJ Community Edition, they like it, they hit the limits, and they upgrade. JetBrains doesn't have to spend a dime on customer acquisition for that first touch. Luna: But what stops a competitor from taking that open source core and building their own proprietary layer? Lucas: That's the risk. The Apache 2.0 license explicitly allows it. And that's exactly what happened with Android — Google took the open source Linux kernel and built Android on top, and now they dominate mobile. But in JetBrains' case, the moat isn't the code, it's the ecosystem. Their plugins, their integrations, their brand. A fork of IntelliJ without the JetBrains tooling would feel empty. Luna: Okay, so open core can work if you have a strong brand and a deep product. But what about the other direction — when the open source project you depend on changes its license? Lucas: That brings us to MongoDB. In 2018, MongoDB changed its license from the GNU AGPL to the Server Side Public License, or SSPL. The new license explicitly says: if you offer MongoDB as a service, you have to open source the entire stack around it. That's a direct shot at cloud providers like AWS, which were offering MongoDB as a managed service without contributing much back. Luna: Right, I remember that. But it also hurt startups that were building on MongoDB, because now they had to worry about whether their use case would trigger the SSPL. Lucas: Precisely. One day you're using a friendly open source database, the next day your legal team is telling you to migrate. And if you're a bootstrapped founder, you don't have a legal team. You just have a ticking clock and a codebase that suddenly has an uncertain future. Luna: So what's the lesson? Don't build on open source? That seems extreme. Lucas: No, the lesson is to know what you're building on. The license is only one dimension. The other is the project's governance and financial health. Is the project backed by a company that could change direction? Or is it a community project with a foundation behind it? Luna: Like the Apache Software Foundation, or the Linux Foundation. Those have neutral governance. Lucas: Right. If you're building on something like PostgreSQL, which is governed by a community and has no single corporate owner, the risk of a license change is near zero. If you're building on something like MongoDB, where the trademark and the copyright are held by a single company, you're borrowing their land. Luna: And that's fine as long as your interests align. But when they diverge, you're left holding the bag. Lucas: Exactly. So here's the framework I'd offer to a bootstrapped founder. First, classify your open source dependencies by risk. Low risk: projects with a foundation, permissive license, broad adoption. High risk: single-vendor projects with a copyleft license or a history of license changes. Luna: Second, you want to isolate those high-risk dependencies behind an abstraction layer. Lucas: Yes. Don't let MongoDB's query syntax leak into every corner of your codebase. Wrap it in a repository interface. That way, if you need to migrate, you swap one module, not a thousand files. Luna: And third — if you can, contribute back to the projects you depend on. Not just code, but money or advocacy. It keeps the project healthy and gives you a seat at the table. Lucas: Good point. There's actually a bootstrapped company called GitLab that started as an open source project. They later added a proprietary tier, but they always maintained the open source core. And they've been very careful about licensing — they use the MIT license for the core, which is about as permissive as it gets. Luna: So GitLab is an example of a company that did open core right, like JetBrains. But they also started with a lot of community goodwill. Lucas: They did. And they're now a public company. But the bootstrapped version of that story is something like Plausible Analytics — an open source alternative to Google Analytics. They're bootstrapped, they charge for hosting, and the code is open source under the AGPL license. Luna: And the AGPL is a strong copyleft, right? That forces anyone who modifies and uses it over a network to release their changes. Lucas: Yes. But Plausible uses it strategically. They want to prevent competitors from offering a hosted version that's just a reskin of their product. The license is part of their moat. Luna: So the license choice isn't just legal — it's a business decision. Lucas: Exactly. And for a bootstrapped founder, you have to think like that from day one. You don't have a war chest to pivot if your foundation shifts. So you pick your dependencies like you pick your business partners. Luna: Let's talk about a concrete example. Say I'm building a project management tool. I need a database, a search engine, a file storage backend. What does the low-risk stack look like? Lucas: Database: PostgreSQL. Permissive license, independent foundation, decades of stability. Search: you could use Elasticsearch, but that's now owned by Elastic, a public company that changed its license in 2021. I'd look at Meilisearch — it's MIT licensed, and while it's younger, the license is friendlier. Luna: What about file storage? You could use S3 directly, but that's proprietary. Is there an open source alternative? Lucas: MinIO is open source under the AGPL. But again, that license might scare you. Or you could use the Apache 2.0 licensed version of the S3 API from something like SeaweedFS. The key is: you want a standard API so you can switch providers without a rewrite. Luna: So it's about API standards as much as licenses. Lucas: Exactly. And that's the fourth piece of the framework: standard interfaces. Use SQL, use S3-compatible APIs, use OpenMetrics for monitoring. That way, you're not locked into any one implementation. Luna: I'm thinking about the cost side, too. Open source is free to use, but the operational cost can be real. Running your own PostgreSQL cluster isn't trivial. Lucas: That's where you have to be honest about your capabilities. If you're a solo founder and you're not a DevOps expert, paying for a managed service like RDS or Supabase might be worth it. The open source project itself is free, but your time isn't. Luna: But then you're dependent on that managed provider, which is its own kind of lock-in. Lucas: To some extent, but less so if they're using the standard open source project under the hood. Moving from RDS to Supabase is a config change, not a code rewrite, because both are just PostgreSQL. Luna: Okay, that's a fair point. So the takeaway is: open source is a powerful lever for bootstrapped founders, but you need to treat it as a strategic decision, not just free stuff. Lucas: Right. And on that note — and I know this might feel like a tangent, but it's actually related — this kind of strategic thinking is exactly what we try to do with this show. No ads, no sponsors, just us digging into the decisions that actually matter for builders and operators. Luna: And if these conversations are useful for what you're building or running, honestly, the best way to keep it going is the smallest possible gesture. If today's episode was worth a coffee to you, that link is buy me a coffee dot com slash fexingo. Lucas: Seriously, that's it. A few bucks, no pressure, and it helps us keep the show ad-free and focused on the stuff that matters. Like how to build a business without giving up control. Luna: Yeah, absolutely. Anyway, back to the framework. You mentioned a fourth piece — what was it? Lucas: Standard interfaces. And the fifth piece is: have a migration plan. Even if you never use it, knowing how you'd move off a dependency changes how you design your system. It keeps you honest. Luna: So you're essentially prescribing a kind of 'design for replaceability.' Lucas: Exactly. And that's a discipline that pays for itself. Every dependency you can replace is a dependency you control. Luna: I want to zoom out for a second. This whole conversation assumes you're building on open source. But what about the bootstrapped founder who wants to release their own project as open source? Is that ever a good idea? Lucas: It can be, if you have a clear monetization strategy. Tools like Sentry, GitLab, and Mattermost all have open source cores and paid tiers. But the trap is thinking open source will automatically drive adoption. It won't, unless you also invest in community and documentation. Luna: And you need to be ready for people to use your code without paying. That's the deal. Lucas: Right. So if your business model relies on every user paying, open source is probably not for you. But if you can offer a better hosted experience or premium features that the open source core doesn't have, it can be a great distribution channel. Luna: It's a trade-off. But for a bootstrapped founder, it's one you can afford to make carefully. Lucas: Let me give you a quick example from my own past. I once built a small analytics library and open sourced it under MIT. I never made a dime from it, but it led to consulting gigs and speaking invitations that more than paid for the time I spent on it. Luna: So open source as a marketing expense. That's a valid ROI calculation. Lucas: Exactly. The key is to go in with eyes open. Know why you're using open source — whether it's to reduce cost, gain adoption, or build community. And match your license and your governance to that goal. Luna: Alright, I think we've given listeners a solid playbook. Let me try to summarize: choose low-risk dependencies, isolate them behind abstractions, use standard interfaces, have a migration plan, and contribute back when you can. Lucas: And if you're releasing your own open source project, have a business model that doesn't depend on everyone paying. That's the bootstrapper's way. Luna: I feel like I'm going to look at my next project's dependency list with a lot more scrutiny. Lucas: That's the idea. A little paranoia pays dividends. Thanks for listening, and we'll be back next week with another angle on building without venture capital.