00:00:16.170
This is my first time giving a talk to an audience like this. I guess it’s my first time at a conference, so I wasn't sure if I wanted no one to show up or for the room to be packed. This seems like a good happy medium, and I’m feeling pretty good.
00:00:23.680
If we could lock the doors and just prevent anybody else from coming in, that would be good. Today, I'm going to talk a little bit about building open-source centric companies and why you want to.
00:00:29.410
Traditional slideshow theory tells me that I should first talk about who I am, but I want to ensure we have all the right people in the room. I want to set the right expectations, so I’ll start with the tagline, which answers why you would want to do this.
00:00:43.000
I'm giving this talk, so clearly I have a lot to say and I think it's the right idea. But I'll start with a little clip and let somebody a bit smarter than I explain it. People say you have to have a lot of passion for what you're doing, and it's totally true.
00:00:55.180
The reason is that it’s so hard that if you don’t have passion, any rational person would give up. It’s really hard, and you have to do it over a sustained period of time. If you don't love it, if you're not having fun doing it, you're going to give up. That’s what happens to most people. If you look at the ones that end up being successful, it’s often the ones that loved what they did and could persevere through tough times.
00:01:14.500
So, that’s my big reason: if you want to do something successfully, you have to be passionate about it. Thankfully, open source is something a lot of people can get passionate about.
00:01:20.030
If you're here, my assumption is that you’re the right people, and you’re included in that group. I’m going to focus a lot on sustainability. A business’ main goal is to stay alive over a long sustained period of time, which again means you need to be doing something that you’re excited about.
00:01:31.760
More importantly, you need to be working on something that will carry you through the hard times. You’ll keep hearing this as a central theme in my talk, so keep it in the back of your mind.
00:01:47.150
Back to the beginning: who am I? Hi, I’m Leah, and I’m the co-founder of a company called Tilde. We build open-source software. We just released our first product, a Rails profiler called Skylight. Hopefully, you stopped by the booth and met some of our team.
00:01:56.860
In my opinion, there are several companies I’ve been involved with and some projects I’ve worked on listed here. Over the years, I’ve had the chance to contribute to multiple initiatives, and I believe you need a diverse group of people to run an open-source project successfully.
00:02:07.479
This means not just coders. I’ll elaborate more on this when I focus on jQuery in a minute. People often find it shocking to hear that I am on the core team of projects like Ember, but there are many different skill sets needed to make an open-source project successful. Not shockingly, writing code is not the only skill required.
00:02:31.300
People sometimes think that if they write it, they will come, but there's more to it. There’s also the question of what to do when they arrive, and how to prevent your servers from falling down, which is more related to infrastructure than actual coding.
00:02:50.780
After this, I tend to focus on areas of projects that are underserved. For me, this means events logistics and team infrastructure. I think about how to build a core team that will help this project advance and stay alive. Evangelism is important for how I want to run companies and certainly for open-source projects.
00:03:06.030
The companies I've spent the last couple of years with have focused on open-source projects. They came together around a specific project, usually one that was important to them and central to their commercial success. I spent about five or six years on the board/core team of the jQuery Foundation, which was one of my more formative experiences.
00:03:16.839
This taught me about having core teams that weren’t only made up of coders. This idea originally came from John Resig before it really crystallized as something he was doing. I watched others take it from there and bring it to more projects, making it a reality at jQuery.
00:03:37.150
I mostly ran events and even managed accounting for a while, which is something you usually don’t need to worry about. But at jQuery, we had a guy named Gary who worked on evangelism and later ran an army of evangelists. We also had people working on the website and infrastructure.
00:03:49.450
I probably don’t need to convince you of jQuery's success. When I started getting involved, Prototype was king and we were essentially laughed out of the room. We all know how that turned out.
00:04:08.420
Currently, Ember is my main focus and the primary project that Tilde is built around. It benefits from lessons learned at jQuery and other projects over the years. Hopefully, in three years, I can joke about Prototype and jQuery, but it will be about Ember and something else, and we’ll all agree it has a vibrant ecosystem.
00:04:26.560
It has companies doing training, consulting, screencasts, multiple conferences, and it’s a really long healthy list. So hopefully it’s headed in the right direction.
00:04:49.890
Before I move on, I want to discuss what I consider the elephant in the room: the benefits. I talked personally about why open source matters, but if you’re here, you’re likely on board with that.
00:05:04.450
You have other stakeholders to keep happy: VCs, partners, co-founders, and all these people that usually aren’t coders. This could be where some hesitation arises. It’s a tough sell to say, 'Do this free thing and success will follow.'
00:05:23.410
I want to talk about some low-hanging fruit: baseline arguments to convince them that building an open-source centric company and supporting your project is the right move.
00:05:37.120
The first is expertise. If you contribute to a project and it’s used to build your product, you have the best in-house expertise. Depending on how successful your open-source project gets, this typically means the kind of expertise you’d otherwise pay a lot to consultants and other outside experts.
00:05:53.710
The second is influence and access. If you contribute to the project and use it, it would be nice to have some influence over its future. There’s a lot of value in having a seat at the table during conversations about feature prioritization, breaking changes, or API support.
00:06:11.300
You all likely know the experience of using something that isn't widely supported. Thankfully, most open-source projects strive for meritocracy. If you’re smart and contributing, you’ll have opportunities to be included in discussions.
00:06:27.669
This is generally the easiest argument to sell to external stakeholders. The third aspect is recruiting. Being involved with open-source projects gives you access to top talent. Those working on your technology may want to join your team.
00:06:43.810
It makes hiring faster, transforming the idea of hiring someone you'd never dream of to being in a position where you’re like, ‘Hey, that super famous influencer is someone I chat with online regularly.’
00:06:56.640
Lastly, related to recruiting, building an approachable brand around open source is very valuable. In communities like Rails and JavaScript, developers appreciate those supporting and ensuring the success of open-source projects.
00:07:09.340
When developers feel gratitude towards your contributions, they are more likely to engage with your product and services. One example of a company successful in this area is AT&T. They hired my good friend, Aaron Patterson, even though he humorously critiques them at RailsConf.
00:07:28.100
Nonetheless, it has been a good strategy that has paid off. If you observe the roster of AT&T's interactive development team, you'll find impressive Open Source luminaries who have come through.
00:07:44.780
I mention AT&T particularly because they stand in stark contrast to Tilde. Tilde is a small company, and the primary thing I’ll discuss is how a company like AT&T can support open source in ways that may not significantly affect their bottom line but creates a substantial difference in attracting talent.
00:08:02.829
We’ve talked about why we want to go this route and how to get people on board. You’re still sitting here, which is great! Now, let’s dive into how you can make it happen.
00:08:24.710
I know that the first thing a lot of people think of when considering this model is funding. Interestingly, when I first wrote this talk, I didn’t discuss funding at all because I find it to be the least interesting part.
00:08:39.350
However, I understand it's a top concern for many, so I think there are some low-hanging fruit ways to approach funding. But first, I want to point out a thread—the importance of building a symbiotic relationship between your open-source project and your funding.
00:08:55.390
The ultimate goal is to create a relationship where the open-source project supports product development, and product revenue supports the development of the open-source project.
00:09:11.540
A great example of this is Basecamp. They’ve used the development of Basecamp to elevate Rails and vice versa, which has helped them achieve an impressive level of success.
00:09:32.880
One cautionary note is that while they got about 90% of this right, understanding that their model of symbiosis kept both aspects moving forward is essential. Let’s talk about the things that typically don’t work.
00:09:49.460
The first is venture capital. An open-source strategy is inherently a long game. It needs time and faith, but most VC arrangements focus on providing a quick return on investment. They need momentum and evidence of ROI to justify funding.
00:10:08.920
Typically, it’s difficult to pull off a strategy that relies on faith and trust for success over several years. This results in frustration and disappointment, making it a poor fit for many.
00:10:22.820
The second remains donations, such as simply placing a PayPal donate link on your website. While these approaches aren’t harmful and should absolutely be part of your strategy, achieving sufficient revenue solely through donations is challenging.
00:10:38.190
Even with jQuery, which was immensely popular, we saw only thousands of dollars through donation buttons on the website for a significant portion of time, which was insufficient to support ongoing efforts.
00:10:54.520
We also spent time thinking about optimizing donation button placements for higher visibility, and our donation drives often required much effort for minimal returns.
00:11:13.519
Selling support is another idea that comes to mind when discussing open-source projects. However, making this model work on a small scale is exceedingly difficult unless you're at a level like Red Hat with vast resources.
00:11:38.520
The first hurdle is that your price tag needs to be high enough to cover your costs, leading to high client expectations for value, which often proves challenging for small companies. If it’s costly on their end, they will seek high value, leading to mismatched expectations.
00:11:57.560
Secondly, running a support infrastructure demands significant administrative overhead—scheduling, sales management, ticket handling, and team coordination—that cannot be scaled effectively without lots of additional resources.
00:12:12.780
Furthermore, you’ll add emotional complexity with urgent customer needs, distracting you from your key goals. This creates unnecessary stress when the priority should instead be on delivering your product and its deadlines.
00:12:26.230
Finally, selling support contradicts the goals of your open-source project. If your revenue relies on offering paid assistance, you're incentivized to make it harder for users to succeed independently and gate specific features providing clients with support.
00:12:40.710
Ultimately, selling support can create a dichotomy in your approach that can lead to failure. Now that I've vented extensively about what doesn’t work, let’s discuss the areas that do.
00:12:53.820
One viable source is consulting. Many people in our community bootstrap through consulting; it comes with both risks and rewards. Revenue from consulting can be quite attractive when established correctly.
00:13:09.839
However, it pretty much requires that you step away from your product for significant stretches of time. You might take on a 3- or 6-month project and need full focus on that, affecting the time available for your own projects.
00:13:25.370
That said, you want to ensure your consulting work is for those using your open-source project. Consulting for clients not involved with your project can widen your revenue pool, but also discounts the insights you can gain from the work.
00:13:43.710
The second key avenue is training. Ideally, you want to shift revenue from consulting to training, offering bursts of revenue without taking extended time from your product. You can squeeze in short training sessions that provide income while maintaining focus on your projects.
00:14:00.380
I personally prefer organization for smaller groups where you can charge a premium for carefully tailored content, allowing participants to receive individualized attention from you or other experts. It’s preferable to a bigger, more general group where engagement may vary.
00:14:18.140
Additionally, events can help scratch several itches. While they do have funding implications, building community through events is crucial to sustaining your project over the long term.
00:14:37.539
Investing time to build a committed community will lead to eventual returns: when a community is invested in your technology, your entire company becomes more resilient as those individuals will help push your success.
00:14:58.520
The whole thing becomes mutually beneficial. Community can lead to consulting and training demand, thus driving revenue. Conferences, while challenging, can facilitate income as well.
00:15:14.410
If conservatively managed with sensible sponsorship and contracts, you should be able to break even or even make money.
00:15:30.210
Sources of income under events include sponsorships, admission fees, and related workshops. I aim to mix all three to optimize revenue.
00:15:47.040
A recurring theme is sustainability. Unlike venture capital, which can lead to quick success, community building through sustainable means is crucial.
00:16:02.310
We've discussed why you want to pursue this path, the benefits, and multiple funding avenues. The next part outlines why this can be hard.
00:16:16.590
Firstly, be aware of emotional ups and downs. As your project grows, people will increasingly depend on your software and expect more from you. Therefore, careful management of those expectations is crucial.
00:16:34.950
Secondly, remain open to allowing others to contribute. When you achieve initial success, the desire may shift to grab all offshoots of income. However, to facilitate growth, building a community around your project is vital.
00:16:51.480
Thirdly, tracking your successes can help you remember the valuable moments when times get tough.
00:17:07.700
Fourth, meet people. It’s easy to become frustrated when negative comments arise online. Seek genuine interactions and positive feedback. You'll find a wealth of encouragement at events.
00:17:22.030
Fifth, remain above the fray. Avoid engaging in arguments with toxic online interactions. Focus on constructive debates on technical merits instead.
00:17:37.540
Finally, take time away. Perspectives can be gained by stepping back from urgent pressures, whether it's checking out another project or taking a vacation.
00:17:54.760
The next challenge is funding. Figure out how to be frugal without being cheap. Most expenses won't directly translate into leads or sales, so it’s critical to keep the team aligned.
00:18:12.780
Three key areas for sound investments are sponsorships, swag/branding, and user groups. Start small with sponsorships and make meaningful connections, instead of overspending on large events.
00:18:30.640
For swag and branding, focus on effective items that create goodwill rather than just pushing your logo. User groups are also excellent opportunities to create positive connections with the community.
00:18:48.580
Always remember to avoid being the evil corporate overlord. Ensure that you remain altruistic and valuable to your community, so they want to support you.
00:19:02.730
Additionally, ensure that your project will not falter while you focus on product development; build a community that can take over lead roles.
00:19:20.380
Next: it’s essential to let go of some control. Allow the emergence of leaders from within the community who can assist in managing. This creates a broader perspective and encourages diverse input.
00:19:35.940
Try the benevolent dictator method—build trust and empower individuals who are genuinely invested in the project.
00:19:50.290
Lastly, it’s easy to make the project a tool that serves only your’s or your company’s needs, but ensure it stays general-purpose and accessible to others.
00:20:06.150
Create an environment where both your open-source project and product thrive symbiotically.
00:20:23.700
Let’s wrap up with some key points: Number one—happiness and sustainability should be your goals.
00:20:39.150
Number two—maintain in-house open-source expertise and access to elite talent.
00:20:55.460
Number three—ensure funding comes from diverse sources to protect against tunnel vision.
00:21:11.470
Number four—do not let your open-source project goals conflict with your product goals.
00:21:27.120
Lastly, focus on sustainability. Building an open-source-centric company is challenging, and while getting to point B quickly sounds enticing, the real aim is to get to point Z while remaining alive and happy.
00:21:43.360
Thank you.
00:33:18.119
You.