Mozilla's New CEO, John Lilly Steps Up To The Task

InformationWeek: Is there any reason for Mozilla to go public? Just over a week ago, Mozilla announced that COO John Lilly had become the organization's new CEO. Lilly is a veteran of Apple, Sun Microsystems and Trilogy Software. Recently, he served as CEO, CTO and VP of products for Reactivity, which was acquired by Cisco in 2007. On Friday, he spoke with InformationWeek about the future of Firefox and of Mozilla.

InformationWeek: What's next on the agenda for Mozilla?

John Lilly: Right at the moment we're trying to get Firefox 3 out the door. That'll happen the first half of this year. I've been using it for a while obviously, but since beta 2 I've been encouraging everyone I know to shift over because I think it's quite a high-quality release, especially for a beta.

Getting MailCo launched, our Thunderbird sister company to Mozilla Corporation, getting that up and off the ground is a pretty high priority for us in addition to Firefox. And as I said in my blog, I'd like us to help people understand who we are a little bit better. My in-laws of course are very proud of what I do and what my wife does, and they've been sending around articles about me and the CEO job. And many of their friends, the characteristic response is, "I've used Firefox for a long time but I had no idea about the mission and the non-profit orientation of Mozilla." So one of my hopes is to help people understand what we're trying to do in terms of keeping the Web open and participatory.

So, get Firefox out, get Thunderbird out, and start to be able to talk a little bit better about what our mission is and how're getting things done, because it's quite unconventional.

InformationWeek: Are there any particular technical issues you're wrestling with that threaten that agenda? To judge by the recent retreat on music DRM, openness seems to have the upper hand at the moment.

John Lilly: For me, it is clear that openness is coming on all fronts. I don't think I'm quite ready to say that DRM is disappearing. I think in music it is. In movies, it's sort of the opposite. And in spite of my appointment at an open source project, I bought an Amazon Kindle and I really like it. But it's decidedly not open. So there are still closed projects and closed-source projects that are significant and meaningful. I think that things like dataportability.org and some of the open API things that are happening are pretty key. We have one, a new initiative in labs called Weave, that's more about making server-side services more accessible and open to anyone.

InformationWeek: What's standing in the way of further openness? What remains to be done?

John Lilly: We're an 18% browser, so I'd say that the hard part of the fight's ahead. People like to write about us as if we've done something, and I think we have. I think we're made a significant impact on the world. But there's huge contention around standards. Getting JavaScript 2 into the world, through the standards groups and implemented, not only in Firefox and Opera but in Internet Explorer, is a big fight still. I think helping Web developers not require proprietary frameworks like Flash or Silverlight, that's a big thing. I think there's lot of innovation to do on the Web; I think there're lots of reasons that there will be closed factions and open factions.

InformationWeek: Is there any reason for Mozilla to go public? InformationWeek: What about openness as it relates to security? Has your thinking changed at all over what needs to be done to keep users safe online?

John Lilly: I don't think it has changed. I think we've tried to get people to focus on the real issued and the real vulnerabilities we see. So we've never been too shy about telling people there's a security breach or a security patch. We try to be as responsible in our disclosure as possible. And actually, that's an area that concerns me not so much for us as for everybody else. The more that we can help put pressure on other vendors to disclose and be honest and open about security breaches when they happen, that's a really good thing. Obviously we put anti-phishing into Firefox 2 and Google's helping us with the anti-malware stuff in Firefox 3. That for us is an evolution, not really a revolution, in thinking. We really care a lot about days of vulnerability and days of risk, so you'll notice that our patch turnaround times are really fast. It doesn't take us very long to get virtually all of our user-base updated.

InformationWeek: With regard to MailCo, Firefox has a pretty good revenue stream with the ad fees from the Firefox search box. Can you offer any details about how Thunderbird will make money?

John Lilly: We're spinning it out so that we can have a core founding team, to work on the mail project. So it's going to be their issue more than it's my issue particularly. But of course we're helping them. We don't know quite where the revenue will come from. We're trying not to focus on it overmuch. The first order of business is to get the organization up and running, and to get some renewed vigor and innovation in mail. I think once we do that, we'll be able to look at revenue. It's important to remember that before 2004, before we launched Firefox 1, we didn't know where the revenue for Firefox was going to come from either. Necessity is the mother of invention, they always say. And I think that putting several million dollars, as we are, into the mail company and letting them really try to be the most useful communications tool for users, and then focusing on the revenue as the opportunities come up, I think that is what we're going to do. It's worth saying that there are a number of open-source projects like this. Another one that I'm involved in is the Participatory Culture Foundation that does the Miro player. And they're experimenting with advertising in the video stream. So we'll see. We don't quite know what the Thunderbird version of our search box is going to be.

InformationWeek: How does the move toward mobile change things?

John Lilly: We're already deployed on the Nokia N810 tablet, a small factor tablet computer. We're working with other manufacturers to get the browser embedded. We've really waited for a pretty long time to get involved in mobile because the mobile industry was so heavily constrained structurally by the carriers and the manufacturers. The rise of Linux we think really opens up the market. We've got some versions of Firefox running in the wild...and we expect to see more this year. And I think it will really ramp up the year after this.

InformationWeek: What vendors do you see as allies in mobile? Among the various mobile OS camp -- Windows Mobile, Blackberry, iPhone, Android, Palm and Symbian -- Firefox seems to have few friends outside of handset makers committed to Linux.

John Lilly: They [Symbian] don't seem to be welcoming... but I think that we'll work on getting Firefox running on Symbian. I've been a Linux skeptic for a long time. But I think that Linux is clearly going to take off on mobile. Google thinks so too. It's the basis for the Android stuff.

InformationWeek: Is there any reason for Mozilla to go public? InformationWeek: Is there any reason for Mozilla to go public?

John Lilly: We're not going to go public. There're a couple of reasons to go public. One is to get money to give to people who worked on the thing. The other is to get operating capital so you can do more stuff. As to the first one, people are here because the Web is important, not for millions of dollars of financial windfall. So an IPO is not attractive for that particular reason. ...The other reason, working capital, we have some working capital in the bank now. I think our recent financial statements show us having about $50 million in the bank. ...We don't think we need to go to the capital markets right now for more money right. And I don't anticipate that will change.

InformationWeek: Is Mozilla doing anything to close the gap between local application performance and cloud application performance? The browser remains a poor choice for gaming due to lack of 3D acceleration support.

John Lilly: For casual games, Flash has been a fine thing to use. For things like virtual worlds, it's not always been powerful enough. I think with things like SVG and Canvas, that'll bring new -- we have some demos of some 3D work in the browser. That'll start to show up. Obviously we're working on desktop integration and offline storage, and I think we'll close the gap in terms of user experience that way. Performance... that's a good question. I think we're really just starting to explore 3D in the browser. That'll come with time.

InformationWeek: Is there anything else that's on your mind?

John Lilly: I think the Weave stuff is important. In labs we have a project around taking things that are in your browser and in the cloud and connecting them up. And we expect more of that to happen through the year. We just put up our test servers and put up a test extension. And I think there will be lots and lots of experimentation there as we figure out how to build a platform for applications.

InformationWeek: Can you explain a bit more about what Weave does?

John Lilly: People use synchronization services for bookmarks or [Web] history or searches or things like that now. They use Google or Del.icio.us or Foxmarks or something. They're basically storing their information on these third-parties. At Mozilla, we think we can provide that kind of cloud computing storage for people. And we think we can start with synchronization of browser state and history and that kind of stuff. And then over time, we can allow other service providers to connect up to our Weave servers so we can get more and more value-added stuff in the browser. So we think we can do this stuff, store encrypted data and give users control over who gets to access what, and how...that kind of thing.

InformationWeek: Do you foresee Mozilla running its own full-blown storage service or will it rely on third-party storage services like Mozy through APIs?

John Lilly: I think both things are in the realm of possibility. We talked about both. We'll have our own data store to start with. I think that hooking in third-parties, both from a services point of view and from a storage point of view is going to make sense before very long.