Mozilla Interview: Brendan Eich and Mitchell Baker

Mozilla Interview: Brendan Eich and Mitchell Baker

by David Sims
O'Reilly Network

March 14, 2000

Remember the dark days of the Mozilla project about a year ago? America Online had acquired Netscape, and many people were predicting that AOL, with its very minimal interest in the Open Source community, would let the project wither and die. Soon after, one of the project's leaders, Jamie Zawinski, left the team with a damning indictment online, expressing his severe dissatisfaction with an invasion of suits and a project that seemed to be going nowhere.

But Mozilla's star is rising again. Milestone 14 will probably be the last alpha before a beta release suitable for the masses to begin playing with. But there are other reasons for renewed interest. One is XUL, the Extensible User-Interface Language that builds user interfaces in standards-compliant web languages. Another is that the project is something of a poster child for the Open Source movement. More people have a vested interest in its success than those who are actively debugging the browser.

We recently talked with Mitchell Baker, the self-described "manager, problem arbitrator, and speaker to suits," and Brendan Eich, JavaScript creator and the main architect of Mozilla, about the browser, its place in the open source movement, and XUL. We started by asking when they hoped to have a beta out for end users.

Mitchell Baker: The release that comes out of Mozilla is aimed at developers, not to be a mass-market, consumer-based product. We do want to get something that is usable at the user level, but our users are not consumers. We hope to reach that point at the end of March.

David Sims: At the point of that release, what I have expected to happen is that people will take your release and create other customized and optimized browsers in different shells. Is that what you expect?

Brendan Eich: Certainly some people are going do that, some will just take the Gecko layout engine and JavaScript engine and put them in their own UI shell -- perhaps implement it with some kind of native widget set.

Other people will want the whole thing, including our Extensible User-Interface Language [XUL] and the mail/news component and HTML editor, mail composer. A lot of people want more than the browsing engine, but there's a good number of people who already said they like the browsing engine and that's what they'll be using.

We don't know exactly who they are, or what their schedules are, but they're all waiting for something stronger than an alpha as a baseline.

Sims: You mentioned XUL, the Extensible User-Interface Language. You see that popping up in other places, not directly related to Mozilla. Did you anticipate that this would become a development environment that people would apply to other projects?

Eich: Not really, we did it out of necessity really. We didn't really anticipate it, but it's the wave of the future, I believe. It reduces the cost of implementing user interface. And it has all the right limbs that HTML and the Web had against native widgets of custom C++ or Java code for implementing user interfaces.

Sims: In the time between when Netscape 4.0 was released and now, I've noticed a lot of people turning to IE 5, particularly people who have a strong interest in XML, because of its abilities there. Do you expect the browsers that come out of Mozilla to be strong XML tools?

Eich: Sure. We already have XML fully supported -- more than IE. Vidur Apparao is on the W3 DOM working group. He and others in the Gecko group have integrated James Clark's Expat XML parser, and hooked it up so that it has a DOM. And you can style it with CSS. And obviously we're using the XML language XUL for our user interface. We're not only interested in supporting people who want to use the standards such as they are, we're pushing the envelope and using it ourselves. We're aggressively implementing the relevant standards.

Sims: People in the Open Source community talk about Mozilla as an example project. Probably it's the level of visibility around it. At times there seems to be a lot riding on your success. I wonder if you're aware of that and what you think of the fact that you're sort of being held up as a model of open source working within, or at least very close to corporate environments.

Baker: Yes, we're aware of that. We've been talking with the Open Source community from the day Netscape decided to do something about this, back before Mozilla was formed. So these views of Mozilla as the model have been known to us. And one of the goals of Mozilla is to try to integrate the traditional Open Source community and methods and development styles, and the strength of that world with corporate, commercial involvement. There are other projects that do that as well. Apache, of course. But there aren't many and there aren't many that have existed for very long. So we're well aware of that.

We also try to talk with the Open Source community as much as possible, and get their feedback. And maintain communication, both in the code itself and organizationally. So we try to be good participants in that world as well.

Eich: Neither the proponents of Open Source like Eric Raymond or Bruce Perens, nor Mozilla staff feel like we're make-or-break for Open Source. Linux, obviously, has broken big in the last year, and Apache. So, Mozilla is a success and will be a bigger success once we finalize our code. But I don't think Open Source will live or die by us.

Sims: In the midst of this, there was Netscape's integration with America Online. I'm wondering if that affected your process at all?

Eich: Surprisingly, not at all.

Sims: Really? Are you far enough out?

Baker: We're far enough out, but more important our charter is different, and our constituency is different, and the way we need to work is in the open source world. And that, I think, is well understood. It's not so much that we're on the West Coast and forgotten about, I think, as Mozilla needs to function in a certain way. That's the way we have functioned, and we continue to do that.

Sims: Does AOL have a visible interest in you? Does that play itself out in any way? Is there a different level of interest than there was from Netscape?

Eich: Not that we can tell.

Sims: What do you guys think about Opera?

Eich: I've been kind of heads down working on Mozilla. We haven't really looked at Opera lately. Early on, it was the inspiration for the Gecko folks to do a small layout engine.

All I know about Opera, to tell you the truth, is what I read on Slashdot and a few other places. But I imagine we'll feel competitive pressure in some sense. I don't think Opera has quite opened their source yet. But people will look at their features and ours, and hold us to account if we don't behave as well.