Tom MacWright is a prolific contributor in the geospatial open source community. He made geojson.io, Mapbox Studio, and was the lead developer on the OpenStreetMap editor. He's currently on the team at Val Town.

In 2021 he bootstrapped a solo business and created the Placemark mapping application.

He acquired customers and found steady growth but after spending two years on the project he decided it was financially unsustainable. He open sourced the code and shut down the business.

In this interview Tom speaks candidly about why geospatial is difficult, chasing technical rabbit holes, the mental impact of bootstrapping, and his struggles to grow a customer base.

If you're interested in geospatial or the good and bad of running a solo business I think you'll enjoy this conversation with Tom.

Geospatial Companies mentioned

Transcript

You can help correct transcripts on GitHub.

[00:00:00] Introduction

Jeremy: Today I'm talking to Tom MacWright. He worked at Mapbox as a, a very early employee. He's had a lot of experience in the geospatial community, the open source community. One of his most recent projects was a mapping project called Placemark he started and ran on his own.

So I wanted to talk to Tom about his experience going solo and, eventually having to, shut that down. Tom, thanks for agreeing to chat today.

Tom: Yeah, thanks for having me.

[00:00:32] Tools and Open Source at Mapbox

Jeremy: So maybe to give everyone some context on, what your background was before you started Placemark. Um, let's talk a little bit about your experience at, at Mapbox. What did you work on there and, and what would you say are like the big things you learned from that experience?

Tom: Yeah, so if you include the time that I was at Development Seed, which essentially turned into Mapbox, I kind of signed the paper to get fired from Development Seed and hired at Mapbox within the same 20 seconds. Uh, I was there for eight and a half years. so it was a lifetime in tech years. and the company really evolved from, uh, working for Human Rights Watch and Amnesty International and the World Bank and doing these small, little like micro websites to the point at which I left it.

It had. Raised a lot of money, had a lot of employees. I think it was 350 or so when I left. and yeah, just expanded into a lot of different, uh, try trying to own more and more of the mapping stack. but yeah, I was kind of really focused on the creative and tooling side of it. that's kind of where I see a lot of the, the fun and programming is making these tools where, uh, they can give people the same kind of fun like interaction loop that programming has where you, you know, you do a little bit of math and you see the result and you're able to just play with, uh, what you're working on, letting people have that in other domains.

so it was really cool to figure out how to get A map design tool where somebody changes the background color and it just automatically changes that in your browser. and it covered like data editing. It covered, um, map styling and we did, uh, three different versions of that tool over the years.

and then Mapbox is also a company that was, it came from, kind of people who are working on the Howard Dean campaign. And so it was pretty ideological and part of the ideology was being pretty hardcore about open source. we hired a lot of people who were working on open source projects before and basically just paid them to work on the open source projects, uh, for their whole time there.

And during my time there, I just tried to make as much of my work, uh, open as possible, which was, you know, at the time it was, it was pretty great. I think in the long term it's been, o open source has changed a lot. but during the time that we were there, we both kind of, helped things like leaflet and mapnik and openstreetmap, uh, but also made like some larger contributions to the open source world.

yeah, that, that's kind of like the, the internal company facing side. And also like what I try to create as like a more of a, uh, enduring work. I think the open source stuff will hopefully have more of a, a long term, uh, benefit.

[00:03:40] How open source has changed (value capture by large companies)

Jeremy: When I was working on a project that needed offline maps, um, we couldn't use Google Maps or any of the, the other publicly available, cloud APIs. So yeah, we actually used a, a tool, called Tile Mill that I, I hadn't known that you'd worked on, but recently found out you did.

So that actually let us pull in OpenStreetMap data and then use this style, uh, language called carto to, to basically let us choose what the colors would be and how the different, uh, the roads and the buildings would look. What's kind of interesting to me is that it being open source really let us, um, build something we otherwise wouldn't have been able to do.

But like, at the same time, we also didn't pay Mapbox any money. (laughs) So I'm, I'm kind of curious, like, if it's changed, like what the thinking was in terms of, you know, we pay for people to build all these things. We make it open source. but then people may just not ever pay us, you know, for all these things we did.

Tom: Yeah. Yeah. I think that the main thing that's changed since the era of tilemill is, the dominance of cloud platforms. Like back then, I think, uh, Mapbox was still using, we were using like a little bit of AWS but people were still just on like VPSs and, uh, configuring things in cPanel and sometimes even running their own servers.

And the, the danger of people using the product for free was such a small thing for us. especially when tile Mill was also funded by the Knight Foundation, so, you know, that at least paid half of my salary for, or, well, sorry, probably, yeah, maybe half of my salary for the first year that I was there and half of three other people's salaries.

but that, yeah, so like when we built Tile Mill, a few companies have really like built on those same tools. Uh, there's a company called Carto coincidentally, they had the same name as Carto CSS, and they built on a lot of the same stack they built on mapnik. Um, and it was, was...

I mean, I'm not gonna say that it was all like, you know, sunshine and roses, but it was never a thing that we talked about in terms of like this being a brutal competition between us and these other startups.

Mapbox eventually closed source some stuff. they made it a source available license. and eventually Mapbox Studio was a closed source product. Um, and that was actually a decision that I advocated for. And that's mostly just because at one point, Esri, Microsoft, Amazon, all had whitelisted versions of Mapbox code, which, uh, hurts a little bit on a personal level and also makes it pretty hard to think about.

working almost like it. You don't want to go to your scrappy open source company and do unpaid labor for Amazon. Uh, you know, Bezos can afford to pay for the labor himself. that's just kind of my personal, uh, that I'm obviously, I haven't worked there in a long time, so I'm not speaking for the company, but that's kind of how it felt like.

and it yeah, kind of changed the arithmetic of open source in this way that. It made it less fun and, more risky, um, for people I think.

[00:07:11] Don't worry about the small free users

Jeremy: Yeah. So it sounds like the thinking was if someone on a small team or an individual, they took the open source software and they used it for their own projects, that was fine. Like you expected that and didn't worry about it. It's more that when these really large organizations like a, a Microsoft comes in and, just like you said, white labels the software, and doesn't really contribute significantly back.

That's, that's when it, the, the thinking sort of shifted.

Tom: Yeah, like a lot of the people who can't pay full price in USD to use your product are great users and they're doing cool stuff. Like when I was working on Placemark and when I was like selling. The theme for my blog, I would get emails from like some kid in India and it's like, you know, you're selling this for a hundred dollars, which is a ton of money.

Tom MacWright on Shutting down Placemark