Maintaining open-source projects that millions use

This story was originally published on Letters by Evernomic, a weekly publication for founders and investors where we write about what we’re learning from building and scaling companies across media and software. Subscribe here for the full versions every week.

Vladimir Kharlampidi is a solo developer in Miami. He built Swiper, an open-source touch slider library that runs on websites like TikTok, Samsung, Coca-Cola, and more. He also founded Framework7, an HTML-based mobile app framework, and commercial products like t0ggles and PaneFlow. Together they've been downloaded millions of times.

We talked about what it's actually like to maintain open source software that half the internet uses, why developers are the most brutal and most loyal audience at the same time, and how he eventually figured out the money part.

Let's start at the beginning. How did Swiper happen?

It was the early days of touch smartphones going mainstream. Everyone was building for mobile but nothing felt native. Everything was laggy and I wanted to challenge myself to build something that felt smooth, touch-first from day one. I put it out there and it just took off because everyone was dealing with the same problem.

There's a line people like to draw between open source as charity and open source as business. You've said it's neither. What is it then?

It's this weird in-between that nobody really talks about. You can build something used by millions of websites, be a core piece of the internet's infrastructure, and still have no obvious way to make money from it. That's the reality for most open source maintainers.

That's worth sitting with for a second. We tend to assume that if something is widely used, someone is making money from it. Open source breaks that assumption completely. The value it creates is enormous but it flows outward, to every company and developer using it, not back to the person who built it.

So when did the money question actually become real for you?

When I left my regular job and started working on my open source projects full time. That's when it hit me that I needed to figure out the commercial side. Before that, it was something I could afford to not think about.

And the route you took was interesting. You didn't try to monetize Swiper itself. You built separate commercial products.

Right. I introduced new commercial products that weren't directly tied to my existing open source work. That made a big difference. The community was fine with it. If I had gone the other direction and started putting paid licenses on Swiper or Framework7, that would probably be a very different story.

This is something a lot of open source maintainers wrestle with. The moment you put a price tag on something people have been using for free, it feels like a betrayal, even if it's completely rational. Vladimir sidestepped that entirely by keeping the open source work open and building new things to charge for. It's a cleaner separation, though it also means you're essentially running two tracks at once.

Swiper does also generate some revenue through sponsorships and donations. Brands pay to have their logo and a backlink on the Swiper site and GitHub repo. It's often an SEO play but it also generates good traffic for these brands. There are also donations from supporters but it's a far cry from what a product with millions of users would typically bring in.

It's worth noting that there are other ways to approach this too. A lot of companies today use open source as a deliberate business strategy. You release the core software for free so anyone can use it, build a large community of users around it, and then charge for a hosted or managed version so people don't have to deal with running it themselves. Wordpress, Ghost, Supabase and many others do this. The software is free and open source, but most people pay for a hassle-free version. The open source part isn't charity in those cases. It's the top of the funnel. It gets people in the door and the paid product is waiting right behind it.

Vladimir's path was different. Swiper was just a library he built and gave away. There was no hosted version to sell, no premium tier, no natural upsell. The commercial side came later and it came in the form of entirely different products.

What's the worst advice you received that you're glad you ignored?

When I started doing open source, a lot of people told me it was a waste of time. That it wouldn't go anywhere, that nobody would care, that I should focus on real work that actually pays. And honestly, on paper they weren't wrong. You're giving away your work for free with no guarantee anyone will even notice.

But I ignored all of that and just kept building. Swiper became one of the most used front-end libraries on the web. Framework7 gave thousands of developers a way to build mobile apps. That "useless work" ended up shaping how a big chunk of the internet handles touch interactions. It gave me a reputation, a community, and the foundation for everything I'm building now.

Part of this advice does sound responsible. Focus on what pays. Be practical. And for most things, it's probably right. But open source is one of those rare areas where giving your work away for free can compound in ways that no paycheck would have. Not always, and not for everyone, but the returns can be reputational, educational, and open other doors you don't even consider.

You've said developers are the hardest audience in the world. What makes them so difficult?

They'll tear apart your code on GitHub, they expect everything for free, and they'll switch to an alternative the second something shinier comes along. But if you earn their trust, if your thing actually works and you keep showing up, they become the most loyal users you'll ever have.

What does that loyalty actually look like? Are people contributing back, or is it more passive?

It's mostly that they keep using it and telling others. Swiper has been around for over a decade and it's still growing. That doesn't happen by itself.

You mentioned you thought every product you built after Swiper would take off the same way. That didn't happen.

No. After that kind of success, I thought I had the formula. Just build something good and people will come. I launched a few things over the years, fully confident, and some of them completely flopped. What I got wrong was thinking a great product is enough on its own. Timing matters, distribution matters, understanding exactly who you're building for matters. With Swiper the timing was perfect: smartphones were exploding and developers desperately needed what I built. With other things the product was solid but the market wasn't there, or I didn't know how to reach the right people.

Do you regret any of those?

Not at all. Every failed product taught me something I'm using now. Not in a cliche way but I'm way more intentional about who I'm building for, how I position things, how I think about growth. I'd rather launch ten things and have three succeed than sit around perfecting one idea that never ships.

Maintaining a project for that long, does it ever stop being fun?

For me it's always been mostly fun. I've gained a lot of experience and knowledge along the way, and new technologies keep emerging so it stays interesting. And I have loyal users and contributors who are supportive. Knowing your work is appreciated and used by that many people is a great feeling.

You work alone on all of this. What does that actually feel like?

From the outside it looks cool. Multiple products and millions of users, but the reality is you're one person making every single decision. Product, code, design, marketing, support, taxes, legal. There's no team to bounce ideas off, no one to cover for you when you're burned out. Some days you ship something amazing and nobody notices. Other days everything breaks at once and there's only you to fix it. It can be lonely and exhausting too.

But you wouldn't change it.

No. I get to build exactly what I want, move as fast as I want, and answer to nobody. That freedom is worth all the hard parts.

If none of this existed, what would you be doing instead?

Before I got into development I was completely addicted to making music. I could easily see a timeline where I ended up as a music producer. That same obsession with crafting something and tweaking it until it feels just right, it's the same energy, just a different output. Or a video game developer. Games combine code, design, storytelling, sound. It's THE creative product.

I could probably just get a regular job as a developer somewhere. But I know myself. I'd last about three months before going crazy. I need full ownership, I need to be able to follow my own ideas. I think no matter what, I'd end up making something of my own. The medium might change but the urge to build doesn't go away.

What's changed your workflow the most recently?

AI agents. Once I started seriously using them, not just chatting with them but actually building with them, my output multiplied. Things that used to take me a week I can do in a day. It's like having a small team without actually having a team. The realization was that it's not about replacing what I do but complementing me against things that may slow me down.

I get the same feeling I had when touch smartphones first appeared and everything had to be rethought. We're still so early with this.

That's it. You can find Vladimir @nolimits4web on X or GitHub.

Subscribe to Letters by Evernomic for more articles like this every week.