Computerworld

Firefox fills the IE void

In January 2004, 94.8 percent of Web surfers used Microsoft Internet Explorer 5.0 or higher, according to the Web analytics research company OneStat.com. Not me, though. For many months I've been using a Mozilla-based browser that can't seem to settle on a name: Phoenix, Firebird, now Firefox. Identity crisis notwithstanding, it rocks.

Trust me on this -- I'm no knee-jerk open source bigot. During Mozilla's long nuclear winter, I stuck with IE because I wasn't willing to live with compromises. Then the tables turned. Suddenly, IE was the compromise I could not live with. Bugs didn't get fixed. Standards support didn't improve. New features didn't appear. And the last vestige of cross-platform ambition evaporated when IE for the Mac was killed last year. The message is clear: Internet Explorer is dead in the water.

Ironically, although Microsoft cited competition with Apple Computer's Safari as the reason for killing IE for the Mac, I've abandoned Safari on OS X for the same reason I've abandoned IE on Windows. Firefox does more, it's moving faster, and -- here's the kicker -- it runs identically on Windows, OS X, and Linux.

On each of these platforms, I enjoy a state-of-the-art end-user experience. Tabbed browsing, search plug-ins, and pop-up blocking are the headline features. But there are wonderful small touches, too. My favorite is "Find as You Type," a built-in incremental search that finds text on the current page as soon as you start typing it.

I've also come to rely on a bevy of features for power users and developers. The LiveHTTPHeaders extension, which opens a window onto the HTTP protocol, is an invaluable aid to integration chores. The DOM Inspector, now included in Firefox, reveals the internal HTML, CSS, and JavaScript structure of a Web page. Using Firefox's XML capabilities, I've built browser-based applications that fetch, transform, and search XML payloads.

During Mozilla's tortuous years of incubation, the project's goals were contentiously debated. Some wanted to focus on producing a fast, reliable, standards-compliant browser. Others sought to create a cross-platform toolkit for rich Internet applications. The first goal has been achieved, and the second is within reach -- a fact that Microsoft (and others) would rather you didn't notice.

The future of "great Windows applications," we're told, lies with Longhorn's next-generation presentation subsystem, Avalon, which will reboot software development sometime in the latter half of this decade. Of course, even Microsoft can't wait until then. Consider InfoPath. It's a great Windows application and a rich Internet client that had to ship in 2003. Its foundation is none other than Internet Explorer -- or rather, the suite of components and Internet standards on which Internet Explorer depends. Could InfoPath have been built on a Mozilla foundation instead? You bet. And the result wouldn't just be a great Windows application. It would be a great application, period.

As a development platform, Mozilla's reach still exceeds its grasp. Its XUL (XML User-interface Language) technology, for example, isn't as polished as the Macromedia Flex markup language or Avalon's XAML (Extensible Application Markup Language).

IE is arguably good enough for the 95 percent who continue to use it for basic browsing. But browser-based rich Internet software isn't nearly as good as it could be. If you build it, they will come.