Wired on Friday/Danny O'Brien: The Mozilla Project was an escape capsule, shed by geeks in 1998 trapped in the wreckage of the dotcom boom. Now it looks like that capsule may be coming back to earth - straight into Microsoft's back yard.

In 1998, Netscape was on its last legs. It had lost what pundits called "the browser war". Its free Web browser was being soundly trounced in the market by Internet Explorer.

The internal development team working on it was slowly being starved of resources by a company desperate to find other sources of revenue.

Then Netscape, heavily lobbied internally and externally by eager developers, released the browser's source code - the program's blueprint - under an open source licence.

It was a brave, innovative and, in a strange way, selfless move. It was clear that Netscape no longer expected to make much money out of its browser code: but it wasn't willing to let it rot on a back shelf. So it gave it to the general public. Anyone could work on it, improve it and fix it.

Programmers, some employed by Netscape but many contributing for free in their spare time, set about building a free competitor to Microsoft's new giant.

Work was slow - we reported on the first glimmers of a genuine competitor in this column two years ago. Even then the hackers had been hammering away for four years.

In the meantime, Explorer had cemented its lead with an almost complete monopoly of the browser world. Only a small Norwegian company, Opera, provided any competition to the browser that Microsoft bundled with Windows. Development on Explorer by Microsoft was largely stopped. The company has not released a new version of Explorer since 2001.

Since then, however, the pace has picked up. In late 2002, after accusations that many cooks had made Mozilla overweight and unwieldy, a renegade group splintered off the main effort and created their own slimmed-down version, now known as Mozilla Firefox.

Firefox quickly became seen as a genuine alternative to Explorer on Windows and the browser to beat on MacOS and Linux. When Apple decided to launch its own competitor to Windows, it was Mozilla developers that it turned to. Apple's browser, Safari, didn't depend on the Mozilla code directly but it certainly relied on the years of experience of the Mozilla coders.

Despite positive reviews and a considerable evangelism campaign by eager volunteers, Mozilla never made much of a dent in Explorer's market share. Search engine Google's "Zeitgeist" page shows how many of each browser visit its site every month. The figures must have made depressing reading for even the fiercest Mozilla advocate. The rebel browser makes a tiny fraction of visitors to Google, compared to Explorer's various versions.

But things appear to be changing. Explorer is showing its age, mainly in security flaws. One serious security breach in Explorer emerged last month and allowed rogue software to collect banking details and passwords from Windows' users. In response, the US government was reported by many media sources as recommending that surfers should switch to one of the alternatives.

Whether the reports are true (the organisation quoted by most reports, the Computer Emergency Response Team, has no record of such a recommendation), the media coverage had an effect.

For the first time, monitors of browser usage are reporting a drop in Explorer usage and a rise in competitors.

And now it gets interesting. In June, the three main competing browsers - Opera, Mozilla and Safari - formed two alliances.

The first was a simple agreement to commit to a shared plug-in standard. This means that modern extensions to the browser to play movies, animation and other multimedia can be installed in all three browsers without modification.

The second has a more far-reaching effect. The companies have formed a group to develop new Web standards - especially standards that will make Web pages behave more like normal computer applications.

This is far more worrying to Microsoft. It is the old Netscape dream renewed. Normally, application writers design their programmes to work with Microsoft's operating systems. Netscape's backers believed that, if you could get HTML pages to work like normal computer applications, applications writers wouldn't need to be tied to one operating system. They could write for the Web - and have their programmes run on any machine, Windows, Mac or Linux.

That's why Microsoft was so intent on smashing Netscape, and securing control of the browser market - and so loathe to update their browser once they had done so. If the Web became the platform that applications ran on, Windows was at risk of becoming irrelevant. It will be interesting to see what happens now that that possibility is raising its head again. We may be in for the second browser war.