Building a Better Browser at Mozilla

Let’s dispose first of the rather mundane David and Goliath story that has been the focus of most recent news coverage of the Mozilla Foundation. “We’re not out to hurt Microsoft,” Brendan Eich says, “so much as to help the Web.”

Eich’s words might sound like mere braggadocio if not for the startling success of the first consumer product released by the foundation, a nonprofit descendant of Netscape Communications Corp. that he serves as chief software architect.

Mozilla’s free Firefox 1.0 Web browser, unveiled Nov. 9 to ecstatic reviews, has been downloaded since then by more than 6 million users; counting earlier test versions, it may already be running on more than 30 million computers.

It may also be responsible for the first recorded decline in market share experienced by Microsoft Corp.’s Explorer browser in at least five years.

Firefox’s advantages over Explorer make its rapid acceptance unsurprising. Among other virtues, it’s faster, more resistant to viruses and spyware and full of useful features that Microsoft, complacent in its near-monopoly, has never provided for Explorer. (Firefox is available at www.mozilla.org.)

Even more interesting, Firefox is the product of an informal group of fewer than 20 programmers, many of them volunteers from around the world working for free, assisted by thousands of technology aficionados who have contributed ideas, identified bugs and tested interim versions on their computers over the years. Their emotional investment in the project resembles that of Apple Macintosh fans: Programmers have already developed Firefox versions in 24 languages, including Slovenian, Chinese and Asturian.

Firefox could be the most successful general-purpose program ever created by the “open source” process, in which the programming code of a project or system is publicly available for enhancement or extension by programmers at large.

For all that the term “open source” may conjure an image of thousands of programmers hacking away in isolation, Firefox -- which Mozilla will soon supplement with an e-mail program, Thunderbird -- also shows that a successful open-source project can’t be merely a public free-for-all.

To be successful, it must be carefully managed, although not as firmly as a corporate effort aimed at creating a proprietary product. “There have to be one or two minds in charge,” says Eich.

The trick is to strike a balance between authority and indulgence. Says Walt Scacchi, a research scientist at UC Irvine’s Institute for Software Research, who has been studying open-source projects: “It’s a question of how much guidance and coordination is required to move things forward without being corrupted by authority.”

Indeed, Eich has spent years mediating among participants who work on Mozilla projects for personal reasons and with varied philosophies.

“It’s a very heterodox community,” he told me in the Mozilla headquarters tucked away in a corner of a Mountain View, Calif., office park. (The office is dominated by a model suspension bridge, fashioned entirely out of empty soda cans and paper clips, that once adorned the Netscape offices.)

“Some are hobbyists who only want to scratch their own itch. Some are into the challenge of doing complex software. Some are motivated by getting their code out to millions of users.”

In a development gratifying to Mozilla, many are engineers who have been assigned full time by their employers to develop applications exploiting its code.

Although Eich says that animosity to Microsoft ranks fairly low as an inspiration, Mozilla owes its birth to the giant company’s ruthlessness. In 1998, Netscape’s business of selling its pioneering browser and related software for profit was destroyed by Microsoft’s decision to give away Explorer for free.

Forced to follow suit and thus unable to support a large engineering staff, Netscape decided to publicly distribute its source code -- the basic programming blueprint. It hoped that volunteers would use it to develop their own innovations, preventing Microsoft from securing an absolute browser monopoly, while allowing Netscape to develop its own specialized (and hopefully profitable) applications. Supervising the effort was a team of Netscape programmers named, after the Netscape browser’s code name, the Mozilla group.

The strategy failed to save Netscape, which was absorbed in 1999 by America Online. AOL last year donated its remains, including old Netscape software code, to the newly formed Mozilla Foundation and donated $2 million in seed capital.

The foundation was largely the idea of Mitchell Baker, a former Netscape lawyer who now serves as its president. Baker says she always understood that a large-scale project aimed at producing products that would have broad utility and require constant innovation required an organizational structure that would give its products an identifiable brand and allow it to employ a crucial core of personnel. The payroll now comes to about 15, she says, most of whom are engineers supervising the open-source work.

The Firefox launch plainly has forced Baker to consider how to deal with the attention and responsibility -- and possibly the revenue -- that may come from market success, without sacrificing Mozilla’s open-source ecology. It may be tempting to bask in Firefox’s rapid acceptance -- “People tell us that it’s such a better Web experience that if they try it, they really like it” -- but a new set of challenges lies ahead.

“I’m not sure of all the ways the stresses will manifest themselves,” she told me this week. “We’re going to be looking at the pressures of success.”

Golden State appears every Monday and Thursday. You

can reach Michael Hiltzik at golden.state@latimes.com and read his previous columns at latimes.com/hiltzik.