Hello! This is my last week working on rbspy. I’m planning to write more about the profiler and how it works soon, but I wanted to take a minute to talk (again) about how I ended up working on the project and how it’s funded. I want to talk about funding because it’s an important part of how a lot of open source software gets created and maintained!

Just about a year ago today (March 23, 2017) Segment announced their Open Fellowship – funding for 3 months to work on an open source project.

The blog post said:

The primary goal of the fellowship is to enable participants to fully dedicate themselves to a project for a few months. We’re hoping to give them a chance to speed the adoption of a new, fast-growing project. Or maybe help them build some long-awaited key features of a library that’s already widely used. Or perhaps even jump-start an entirely new idea altogether.

When I saw this, my thought process went:

1. That sounds amazing!! I want to do that!
2. But what would I work on? Oh, I have this Ruby profiler prototype I built last year, maybe that?? That would be fun and I think it might help a lot of people!
3. I’ll apply and see what happens!

So I filled in the form, thinking “who knows if they’ll accept me? I’ll just apply and see what happens!! If you’re not getting rejections you’re not being ambitious enough, julia!” and waited. They accepted my application, and the project happened!!

open source funding models

Before I talk about what made this fellowship + sabbatical work for me, I wanted to take a second to catalog a few open source software funding models that I’m aware of:

• a company employs people who work on the software (like Hashicorp or Google or Mozilla or Igalia)
• companies who use the software donate to a software foundation (like the Linux foundation or Ruby Together) that pay developers to work on the project
• Crowdfunding with Patreon (successfully used by the vue.js project)
• Google Summer of Code / Outreachy pay a ~$5000 stipend to students / folks from underrepresented groups to work on an open source project for 3 months. Participants get a mentor.
• Segment Open Fellowship / Stripe Open Source Retreat (pay someone to do 3 months of work on an open source project they choose)

There are certainly more that I’m leaving out – see A handy guide to financial support for open source and getting paid for open source work for more!

One difference is that some models involve full-time employment and some models are about sponsoring sprints of activity on a project. In this case, the “sprints of activity” model worked really well for me.

flexible starting time: very helpful

Originally, the fellowship was supposed to run from June 15 to September 15, 2017. Obviously I didn’t do it from June to September, because it’s March now! So here’s what happened. Segment offered me the fellowship (yay!). I realized I couldn’t take time off work by June 15, and told them so. They extremely kindly said I could delay the fellowship until a later time that worked for me. I suggested January -> March 2018 (6 months later) and they agreed to it!

Then I asked my manager if I could take 3 months off work from January-March (unpaid) to do the fellowship. Since there was about 6 months notice (lots of time), we got permission to do it. Amazing!

I think it was amazing that Segment was so flexible about the dates of the fellowship – their flexibility (and Stripe’s willingness to let me take 3 months off work) was what made it possible for me to do this. Otherwise I probably won’t have done this project.

Giving my team a lot of notice was great – I was responsible for a pretty big project at the time, so putting the sabbatical 6 months into the future gave me lots of time to make a lot of progress on the project and help get other people on the team up to speed so they could take over while I was away.

doing the fellowship remotely

The other thing that Segment did that I thought was amazing was – they said it was totally fine if fellowship participants did their projects remotely! This was important to me (I wouldn’t have applied otherwise) because, realistically, I wasn’t going to move to San Francisco for 3 months to do this project.

I blogged about my progress on and off as I went, which I think helped them feel like they knew what was going on even though I wasn’t in their office. They didn’t ask that I do that, though – they were just supportive of me working in whatever way I chose to work.

Working from home is how I work usually so this felt pretty normal.

sabbaticals are cool

I really like my job, and I didn’t want to quit just to be able to dedicate some time to a project that I was interested in. So I think it’s awesome that I was able to take time off work to do the project. I’ve been at Stripe for 4 years, so Stripe’s sabbatical policy said that I was allowed to take a 3-month unpaid break and come back after.

Working on a smallish (~4000 LOC) open source project has been a nice thing to do for a relatively short/focused amount of time. I don’t think rbspy needs more full-time dedicated attention right now (it works! time to step back, see what people think of it, and fix bugs / add features as they’re needed), so I’m happy to go back to a job & team that I like.

Another benefit of doing this was that now I have actual code that I’ve written out in the open on GitHub! I don’t really believe in “github is your resume” (lots of great programmers don’t do any open source work! that’s fine!) but it does feel good to have.

build a prototype to build confidence

Taking time off work to build a programming project felt a little risky. What if my ideas didn’t work? I think the biggest thing that helped me be confident in my ability to do the project was – I had a prototype!! In May 2016 (blog post: “How to spy on a Ruby program”), I’d written a small prototype of what eventually became rbspy.

That prototype wasn’t really something people could use – it was pretty fragile and, while a few people had used it and contributed to it, it needed a lot of work. But it did kind of work, and so I felt pretty confident that I could make it into something more robust if I just spent time on it. (spoiler: I was right! it’s way more robust now! people are able to use it!)

I don’t do programming projects after work

Some people program after work. I think this is cool, but I don’t! I don’t really feel like I have the energy to do it. For whatever reason I do have the energy to write hundreds of blog posts about programming, but I guess that comes out of a different energy reserve for me :).

I’ve done exactly 3 side programming projects in the last 4 years, none of them particularly time consuming (turn off retweets, computers are fast, the rbspy prototype).

It’s been useful for me to accept that I don’t really want to program after work or on the weekends. But of course there are projects I’m excited about building, so this sabbatical was a great opportunity to do programming work that I wouldn’t otherwise do.

apply to things, even if you’re not sure how they’ll work out

One lesson I learned from this was – apply to things! I did a couple of things “wrong” when applying to this fellowship:

• The application asked for a resume and I didn’t have an up-to-date one, so I just filled in the “resume” section with a link to my blog (since this blog lists a lot of the work that I’ve done)
• More importantly – when I applied, I didn’t know whether I’d be able to actually accept the fellowship if they offered it to me. It was short notice (I applied in April for a fellowship starting in June), and I didn’t have a plan yet for how I’d take time off work.

But I applied anyway, and then when they offered me the fellowship and the dates didn’t work for me I told them that I couldn’t do it (I actually said “sorry, I actually can’t do those dates, please offer the fellowship to somebody else!”), and they offered to change things around to make it work! So I asked for it to be delayed by 6 months (which at the time felt like a big thing to ask for), and they agreed!

It actually took me quite a while to work out the details – they told me I was accepted in June, and I only got permission from work to take time off finalized about a month later. But that was okay! They were very patient and it worked out :)

So I think it was a great idea to apply with no clear plan of how I’d handle it if they accepted me, because we figured it out and it was fine!

maybe take a sabbatical!

It looks like Segment might be doing the same fellowship program again this year. If you’re interested in getting email when they open applications, the open fellowship homepage has a mailing list.

For me taking this sabbatical has been great – it’s really fun to have an focused goal that’s outside of my usual comfort zone. “Ship a cross-platform binary”, “do a medium-sized systems programming project that people actually use”, “build a profiler”, and “lead an open source project” are all things I was interested in but had never done before, and the end product is something that people are using and that I’m happy with!