Blake Ross and Dave Hyatt started Firefox as a side project while working at the Mozilla Foundation. They were working to revive the struggling Netscape browser, but became frustrated by the constraints imposed on them. So Ross and Hyatt decided to build a browser that they would actually want to use. Working in their spare time, they began developing a new browser that was fast, simple, and reliable. In 2002, they launched the initial version, called Phoenix, and in 2004 they released Firefox 1.0, which was an instant hit. Like a lot of things described in this book, Firefox was something new. It was an open source project run like a startup, both in the concern for the end user and in the attention paid to marketing. The results were impressive: Firefox has cut into the formerly overwhelming market share of Internet Explorer, and dominates among technical users. In 2005, Ross took a leave from Stanford University to start a startup with fellow Firefox developer Joe Hewitt. Livingston: Tell me about how Firefox got started. Ross: Firefox grew out of Mozilla, which itself has a very long history that I won't go into now. I personally started working on the Mozilla project in 2000. It was open source; anyone could work on it. I started working closely with the Netscape team, because they were basing their product on Mozilla. I was helping them fix bugs, and they invited me out for an internship one summer, so I went out to Netscape, which was a pretty cool first job. Livingston: You were only 14, right? Ross:Right. I worked out in California, and it was great the first summer. Then I started working from home, and when I came back the next summer, things 395 Blake Ross Creator, Firefox 29 CHAPTER had gotten much worse. Netscape kept sliding further and further in the market. At this point, they had something like 5 percent market share. This is post-AOL, post-browser war and all that. Things got a lot more desperate when AOL tanked and started to demand more revenue from the browser. They wanted a return on investment, and they'd bought Netscape for about $4 billion. So the browser started to turn into nothing more than a vehicle to drive people to Netscape.com. There were search buttons everywhere, advertisements everywhere. It was a mess. The culture didn't focus on users. It was painful to be working there. Firefox was more a response to our experience at Netscape than to the dominant browser, Internet Explorer. Explorer had basically been abandoned at that point; in 2001, Microsoft disbanded the IE team. So we started Firefox as a way to work on the browser that we knew we could make if we weren't being controlled by marketing, sales, and all these other influences inside Netscape. It started off with just three or four of us--the people who had always been fighting these battles within Netscape to make the right decisions for users. For example, we wanted to include pop-up blocking in Netscape 7. It would have been the first mainstream browser to include pop-up blocking. The Mozilla folks had all the code ready, but Netscape wouldn't include it because Netscape.com had a pop-up ad. Those kinds of decisions were painful, and it was frustrating to have our names on the product that was getting released. So we started a project called Phoenix, which was supposed to be an allusion to the mythical bird that is reborn from its own ashes. It was like the project was being reborn from the ashes of Netscape. Livingston: Who was involved? Ross: David Hyatt, Joe Hewitt--who is now my partner on a new startup, Parakey--and I were on the development side, with Brian Ryner and Asa Dotzler providing build and QA support. The project was like an afterthought for the first 6 months to a year, something we worked on at Denny's after work. I went back home to Miami, and we worked on it online for a while. Phoenix was basically a fork of the Mozilla code base that we controlled. We closed off access to the code, because we felt it was impossible to create anything consumer-oriented when you had a thousand Netscape people in search of revenue and a thousand open source geeks who shunned big business trying to reach consensus. We just wanted to close it off and do what we thought was the right thing. We went through a couple name changes, Mozilla offered us more support, and that's kind of how it all got started. Livingston: What were some of the other names? Ross:It started off as Phoenix, and we quickly encountered trademark issues. It was just the three of us, we weren't lawyers, and we were broke, so at that point we probably would have done anything someone asked of us. In this case, Phoenix Technologies complained because they had some kind of web browser, too. We renamed it Firebird, because it's the same imagery, but there was an 396 Founders at Work open source database already called Firebird. So we renamed it again. At that point, it was fairly popular--though not nearly as popular as it is now--so we wanted to keep the "Fire" part of the name. We just went through Fireanything names for a couple of months, and somebody came up with Firefox, which is actually the Chinese name for a red panda. Livingston: Were the Firefox developers all in different places? Ross:When we first started doing it, we were all at Netscape. Then Dave left to go to Apple to work on Safari, and we had some other folks like Ben Goodger from New Zealand, Pierre Chanial from France, and Jan Varga from Slovakia come on board. I went back to Miami, and we continued to work together online. Joe and I still collaborate through IM on Parakey, even though we're about 20 minutes apart, because we're so used to that environment from Firefox. It's just so much faster to collaborate online than it is for him to drive down to me or me to drive up to him. Livingston: Were there any conflicts with Dave working at Apple? Ross: Yes. They were also making a simple end-user browser, and he was not really supposed to be working on a competitor to that. It wasn't on our end that we had a problem. Livingston: Did he leave Apple? Ross: No. He still works on Safari right now. He did Firefox and then went off to Apple. Livingston: So then it was just a few of you. Ross: The Firefox team is always changing. It's not fair to say there are just a few of us, because we're based on Mozilla, which obviously has dozens of developers, and there are a lot of developers working on Gecko, the core layout engine. The Firefox team itself--the people worrying about everything wrapped around the engine and working on the separate fork of the code base--was always about four or five different people for the first year. Now there are a lot more, obviously, because it's the main source tree. All those people that were working on Mozilla now work on Firefox. Livingston: What was the first turning point when you knew you were really onto something? Ross:I think it was when we put out our first milestone, which wasn't even... We put it on an FTP site and had an article on mozillaZine, which is a community news site. It was already getting as many downloads as a Mozilla milestone. On the one side, you had a lot of Mozilla people--the hardcore developer types--who didn't like what we were doing, because focusing on "mom and dad" is heretical in much of the open source world. Then there were a lot of people who were saying, "Finally, Mozilla is stepping away from its geek roots and doing something more mainstream." We got a lot of coverage early on from bloggers and PC World and stuff like that. It got out of control pretty quickly. Blake Ross 397 Livingston: Did you have problems getting users at the beginning? Ross: No, but the users we were getting weren't really the target audience; these were people that downloaded beta builds from Mozilla. So it was still a geek audience. We had to transform the culture at Mozilla because it was all based around open source ethos, which says programmers are kings, marketers are sleaze, and everyone else can read the manual. All the branding for Mozilla looked very Communist--the logo was a dinosaur and the banners ads were... I can't even describe it, but very odd, technical kind of imagery that didn't appeal to most people. We had to move a lot of that into a more mainstream world. Livingston: How did you do that? Ross: The first thing that happened was that Netscape split off Mozilla into an independent entity. Mozilla was once just the open source technology arm of Netscape--they made technology and Netscape distributed it. When Netscape said goodbye, Mozilla didn't really have any kind of major distributor anymore. As Firefox matured, Mozilla decided that they could try to distribute it directly to the user without having to go through a middleman like Netscape. At that point, the culture started to shift out of necessity; the organization had to cater to more users or potentially collapse. Livingston: As you were working on this, did you worry about competitive threats? Ross: No, Firefox was very different from traditional startups. Companies usually worry about competition for financial reasons, but when we did Firefox, money was just always sort of there. There were donations, seed money from AOL; we eventually got this Google deal, but it wasn't a source of fear for us, because we knew if it didn't make money... It wasn't even supposed to make money--it was a hobby, right, so we didn't really care. I was in school. It didn't have to succeed. It sounds bad, but the project was kind of just for us at the beginning--to make something that we knew we could make, but not inside Netscape. It was an outlet for those frustrations. We wanted people to use it, but we weren't going to kill ourselves if it failed. We defined success in terms of users, not competitors. In any case, the IE team had been disbanded, and Netscape had bowed out, so the market was wide open. We didn't crunch numbers or conduct market analysis; we relaxed and followed our gut. There's a lot more pressure now with Parakey. People expect another Firefox or something like that. Livingston: People must have high expectations for you, which is not a bad thing, I suppose. Ross: Not a bad thing, but you have to deliver. It's hard to under-promise and over-deliver when everyone's promising things for you. We're trying not to hype up what we're doing until we've got something people can use. People expect the world, so if you hype up what you are doing, you have to deliver, and it's not easy. 398 Founders at Work Livingston: Did any competitors ever do anything to anger you? Ross: Not directly. The only thing that bothers me is that Microsoft seems completely driven by competition. We tried to be driven entirely by users. There was a need, so we tried to cater to it. We didn't say, "We're going to try to crush Microsoft" just to crush Microsoft. That wasn't the intent, even though that's kind of the stated goal of some misguided open source projects. Whereas Microsoft, they win a browser war, so in 2001, they bow out. Which is completely irresponsible, because this is the most used software application in the world, and they just stopped developing it. Now they are back in the game, because they have competition, so that pulls them back in. I will say that Internet Explorer 7 is shaping up to be a good browser; I just wish it came a few years earlier. We also see them trying to emulate a lot of the more genuine community spirit that we've built up. People like Mozilla because we're open source; we try to be transparent and honest with the community. We're a free product. We work with people. And we're starting to see that kind of thing emanate from Microsoft. They have a team blog now and they are trying to be very buddybuddy, but it feels like a PR pitch, as if they looked at our situation and now they are trying to bring that sense of goodwill over there. If it were genuine, it would be great, but it feels like a sales pitch. It's getting better though. I respect companies like Opera, which also produces a browser. They aren't doing that well right now, but at least they're in it for the right reasons. They've been around for a decade now, and they are passionate about the Web. Microsoft just kind of comes and goes with the money and the competition, and that doesn't seem like the right motivation to make a good product. Livingston: Looking back, what did people misunderstand about Firefox? Ross: Many die-hard open source fans misunderstood our goal. Usually, in an open source project, if you're not a developer, it's kind of like, "What are you doing here?" A lot of people misunderstood the real audience we were going after. It's hard to explain exactly what that means, but you can imagine, here's this Mozilla project, it's very open, everyone gets a say. If you are a developer, you get to vote on whether or not a feature gets implemented. Then we come along and say, "We're making a product for mom and dad. You still have a voice here, but some of the features that you think we should add may not be the ones that they want to use. So you have to take our word for it that, even though 500 of you want something right now, you may actually be in the minority of a much larger group that we're pursuing that's going to be silent during this phase of development." It's hard to convince 500 flesh-and-blood developers that their pet feature may not be desirable to 500 million imaginary users, especially when you have no hard evidence to back it up. In some ways, I'm glad it's just the two of us again on Parakey. We can work very fast and there are no politics. Blake Ross 399 Livingston: So in open source projects, you have to listen to the opinions of other developers. Ross: Sure, they're the ones building the product. We just have to be wary of our inner geek voices and make sure we're considering the needs of the world at large. I don't think Mozilla did that, and the project stagnated at a few million users. Livingston: Do you think Firefox has reached the mainstream because it is better? Ross: There are a million different reasons. Many people think it's easier. Others were just sort of weaned onto it when their children put it on their computers. Of course, we've also done plenty of legwork to reach the mainstream. It's all word-of-mouth marketing. We have a site called Spread Firefox that Asa Dotzler and I started in 2004 when we launched Firefox. It's basically a way to leverage the talents of people who are not coders. We said, "Instead of just being developer-only, like most open source projects, how do we leverage college students and Toastmasters and people who knit--just every kind of talent you have and every organization you're a part of. How do we match you up with other people in your region and give you tools to spread Firefox?" That was a huge success. We've had over 250,000 people sign up. We also did an ad in the New York Times. Ten thousand people donated between $10 and $30 each to buy two full-page facing ads in the New York Times when Firefox launched. Of course, that's a couple hundred thousand dollars, but we didn't have a marketing budget. That was all communityfunded, which is pretty unusual for any software project, let alone an open source project. Livingston: So Firefox spread because the browser is better and through word of mouth? Ross: Yes. We don't have people shaping a message or working the press. It's all been grassroots, word of mouth, done through Spread Firefox. It's been interesting because we've seen about a dozen companies adopt the same model since then. There's GoTrillian.com, SpreadOpenOffice.org--there are all these different copycat sites. Livingston: Was there ever a point when you were really worried? Ross: Not really. But I'm making it seem like startups are so stress-free, and of course that's just not true. It's just really freeing not to be... We weren't trying to strike it rich with Firefox. It's open source and it's free. We weren't trying to take over the world; we had kind of modest goals, and it was OK if it failed. We were a lot freer to make risky decisions. If you can afford to do things that way, it's just so much better. You're not thinking about venture capitalists or marketing or sales. Just product and users, all day every day. 400 Founders at Work Livingston: You were pretty young when you worked on Firefox. Was there anything you found you were better at than you thought? Ross: I thought marketing was something that required a degree and formal experience. It turns out that marketing is just making the product good enough that people spread it on their own, and giving them ways to do that. It's a lot easier and more natural than I thought it would be. Now I can't stand meeting with professional marketers who try to "craft" the "message" and all that junk. Livingston: What surprised you most? Ross: How easy it was to get Firefox to take off, at least in light of the death knell people had been sounding for years. We'd been hearing forever that nobody downloads a client anymore, and browsers are dead, and Mozilla can't make it. It's never going to go anywhere; the market has been monopolized. We just ignored all that and did it anyway, and it worked. It's a bit harder to take analysts and other "industry insiders" seriously now, because Firefox proved them wrong. There are a lot of people in the industry who aren't actually the ones writing the code or contributing to the project, but they want to feel like they are relevant somehow, so they make sweeping predictions that draw attention. I think you have to be in the project and be the one moving it forward to truly understand whether you have a shot at success. One analyst has already announced he's "skeptical" about Parakey and he barely even knows what it is, let alone tried it out. Smells like Firefox all over again. Those kinds of comments are so motivating. I love the challenge. We talked to plenty of people at the very beginning of Firefox. It was obvious that people were not happy with their browser, and it was very clear that, if we could do something better, we might be able to get them to use it. Livingston: Do you remember people's reactions when you gave an early demo of it? Ross: People loved the simplicity and went crazy over tabbed browsing. What's weird is that I didn't really talk to anyone I knew personally throughout the course of Firefox development. My parents and my friends--most of them didn't really know I was working on Firefox until it came out and there was the Business 2.0 article. That's when everyone was like, "Wait, you work on Firefox?" They knew I "did something" with computers, but... Livingston: Your parents didn't know? Ross: Kind of. I think they knew I worked on Mozilla. They knew I worked at Netscape, so they knew I worked in browsers, but they didn't really know my involvement in Firefox until they read about it in a magazine. Which is kind of how I prefer it, because it's much easier to spend a couple months on something, fail silently, and just go back to school, than it is to tell everyone that everyone is going to use our product. It's easier if people aren't bugging you until you have something to put in their hands, and then they can tell you if it's good or not. Blake Ross 401 Livingston: So the stakes were lower. Did you ever want to quit? Ross: Well, I did in a way. I went back to school for 6 months, and I wasn't working on the project much during that time. It wasn't that I walked away-- we knew there were people working on it--but it was leisurely because we knew that Microsoft wasn't coming back any time soon. Livingston: Now you are in a "real" startup. How did you get started? Ross:In some ways, the media and the venture capital industry made it happen. From our earliest days at Netscape, Joe and I were always shooting the breeze about how terrible software was and what we would change if we could. After some Firefox press hit, we started getting emails from investors saying, "We want to meet." And we'd think, "Meet about what? It's an open source hobby project." Then we realized, "They want to meet about funding us, so we should probably get some kind of company together." We figured this was the perfect opportunity to act after years of talk. Livingston: You wanted to take advantage of the rising tide? Ross: Right. We already had ideas around software, and we said, "This is the time to do it if we are going to do anything. People are going to listen to us right now, so we might as well go for it." Livingston: Do you have a name? Ross:We're calling it Parakey for now, but who knows if it will stick. Firefox was our fourth name. Livingston: Can you tell me about any of the challenges you've faced? Ross: One thing is just time. Whenever I'm doing something now, I feel like I should be doing something else instead. If I got married tomorrow, I'd probably be worrying about a code issue during the ceremony and deliver my vows in Python. It's a nonstop state of stress. The first couple months we did the startup and all these venture capitalists were emailing us, we felt like we had to meet with all of them. We thought, "Oh my God, we have to say yes; we can't say no to these people." Now we realize that time is our most valuable resource, and every minute we spend in one of these meetings just sitting there is time wasted. Things are getting better. We're starting to push people away to give us space to work, but in some respects it would be so much easier if the Firefox thing hadn't happened. We should be setting our own timeline, but people are already waiting for what we're going to do next, so it's hard to relax under these kinds of circumstances. It's a lot of pressure. Livingston: Who are your mentors? Is this Joe's first startup, too? Ross: Yes. That's kind of the problem. We don't have that one person who has done this a thousand times who can advise us. We have a good lawyer. We're looking for a mentor who doesn't have ulterior motives and who is aligned with our interests. 402 Founders at Work Livingston: Which also must be a problem since you are so well known. Some people must think, "This is my ticket!" Ross: I can't tell you how many times I get an email from someone who just wants to have dinner. So we'll have dinner, and we'll chat about politics, the weather, whatever. Then we'll have dinner again, and slowly it comes out that they want something. Eventually you find out that they want to come work for you, or they want to... sometimes they don't even know what they want, but they know they want something. It's hard to see what people's intentions are at the beginning. We're also overly paranoid because the first thing we did when we started the company was talk to a bunch of entrepreneurs who told us, "Don't tell anyone what you are doing. VCs are sharks." Meanwhile, you hear from the VCs, "You're too paranoid." So it's hard to find the right balance and be human, because you don't know who's genuine and who's not. Livingston: It must be frustrating not to be able to share your idea. Ross:Incredibly. If you ever want to stop a conversation dead in its tracks, just use my magic words: "stealth mode." I've also found "programmer" to work well in many situations. But we'll have our day. Livingston: Are there any lessons that you learned in the Firefox days that you are applying to this new startup? Ross: One is to make sure you are always in communication with the people who are eventually going to use your product. It's very easy to just lock yourself in a room and code all day, and you forget what the real problems are that people are having. So you have to keep talking to people and keep refining what you are doing. I also learned how you build up the right kind of buzz about your product in an honest way. With Firefox, we catered to the bloggers first, even though they weren't our primary target audience. Once you get the prominent bloggers to pick up the scent, you attract the intermediate press, the PC Worlds and the CNETs. You still don't have any moms or dads yet, you don't have any nontechies, but once the mainstream press sees PC magazine talking about it, then they start to cover the story, and they actually make it kind of a self-fulfilling prophecy. They write that "everyone is talking about Firefox" when, of course, mainstream users haven't even heard of it yet. But they are going to, now that the New York Times wrote about it. Livingston: What are your biggest challenges starting a startup? Ross: One is, in general, not knowing what's "normal." Investors hand us "normal" term sheets, consultants ask for "normal" fees. I'm 21--I haven't seen enough of the extremes to know what's normal. Our approach has been to make decisions slowly and methodically, do our research, and figure out who's on the level and who's selling us lines before signing anything. The other problem is just finding the time to finish the project and still see my family, my friends, my girlfriend. It's very hard as two people. It's a very big project. Blake Ross 403 Livingston: Is your time horizon several years? Ross: No. It's short term for launch. Livingston: Because there will be a race? Ross: We don't know of anyone doing specifically what we are doing, but you can just feel in the air that everyone's moving toward this kind of model. Who knows, someone could announce it tomorrow. Livingston: Are you able to say who you are most nervous about as a competitor? Ross: I'd say Google or Microsoft. It's a big enough project that I'm not sure a startup would be trying to do it, except us because we're nuts, but it's possible. Of the known companies, it would be Google or Microsoft. Livingston: So right now you are operating on a small amount of seed funding? Is that to pay your rent, etc.? Ross: We're going to take more before we launch, but we're trying to take as little as possible. We don't want $12 million. I don't know what we'd do with that. We don't even have an office. We're just working out of our apartments. Livingston: Do you plan to get one? Ross: Eventually. I need to see how many engineers I can fit in my bathroom and closet first. Livingston: Are you nervous that this idea is too big for two people? Ross: Yes. But we're also nervous about finding someone else, so it's hard. Just finding and interviewing candidates is stressful, because it's not like there's a team back home coding. If Joe and I are at a meeting, no one is pushing the product forward, and that's scary. There's a question of, "Is it better for us to spend all of our time iterating very quickly, or potentially ruin that dynamic by bringing on someone that we don't know well?" In short, I'm nervous about everything. If you're doing a startup and you're relaxed, you should be very worried. Livingston: So far, what has surprised you most about starting your own startup? Ross: One thing I didn't know was how tightly connected everyone is in the Valley. We'll meet someone, and then we'll meet someone who I would never expect to even know that person, and they'll say, "I heard you met Tony last week." It's such a small industry, and so much business is done through the network circuit, which is kind of upsetting, because I'd rather the good companies get the good deals and the bad ones don't get deals at all. Instead, it's more like, "Who do you know?" I can definitely see where the Google guys came from when they refused to play by these rules. They didn't know anyone, and they didn't schmooze their way in.