If there's a recipe for sure disaster for a commentator or reviewer, the first ingredient is having an opinion about Mozilla. For over two years, the Linux community has been engaged in a love/hate relationship with the project, revealing the many fault lines and stress points running throughout the Linux world. In the absence of news about the project, a steady trickle of commentary flows, usually eliciting more energy and passion than the author likely put into the instigating piece. Mozilla has become a lightning rod, and the opinions surrounding it are heated. Even Suck, a passionate exercise in detachment, has worked up enough concern to advocate that the project be disbanded and rendered for whatever good bits have come of it: the virtual glue factory for a horse too late out of the gate and too plagued with ailments to contend.

Fortunately, for those who keep one eye out for Mozilla-related headlines, there's news now: the Mozilla project has made Milestone Release 17, and Netscape (who opened the source for the project and has provided much of the core team) has released its second preview of Netscape 6.

Despite the near-constant background buzz over Mozilla, milestone and preview releases serve as energizers for the Mozilla-observing community (be they admirers or cheerful participants in an alleged deathwatch). The incremental improvements from the last milestone rally the faithful. The ongoing list of bugs large and small fuel the pessimists to greater heights of eloquence regarding the project's impending doom.

We aren't here, however, to dwell overlong on the community's assorted reactions to the project. In fact, we've chosen to concern ourselves with the commercial manifestation of the Mozilla effort: Netscape 6.

The Netscape "product" is the commercial face of the Mozilla project. Where nightly builds of the bulk of the code going into Netscape are available in the form of tarballs at mozilla.org, Netscape has preferred to make only two releases so far, making a small amount of to-do about each.

Reaction to the first Netscape 6 preview release were mixed. Speaking to Microsoft users who downloaded it without any particular interest in the underlying meaning of its open source development reminded me that the average Linux user probably has a lot more patience for code "under development," than those who prefer to use words like "unfinished" or "missing stuff" to describe the same thing.

Reviewers were a little more mixed in their perceptions. Some took the high road, noting that as a preview release, Netscape 6 was something to behold. Others called it too little, too late, and joined the Mozilla obituary-writing community.

Now the second preview is out, more complete, more usable in general, and supposedly (alongside Milestone 17), the last release before the business of optimizing and bugcrushing are undertaken in earnest. In other words: this is what you're going to get, it just doesn't work as well as it will.

Before going any further, we'll issue a few warnings:

Netscape 6 will not present itself as usable to people in need of a stable browser. It has bugs, it misbehaves, and it disappears from the screen over mysterious vexations known only to itself. If casual users can spend an hour of worry- (and crash-) free browsing, we're happy for them.

We also don't believe it's somehow a mark of an imperfect Open Source/Free Software advocate to point out that the product isn't ready for prime time. Netscape 4.73 is more stable on our test machines where Linux is concerned, and when we ran the Windows version on a dual-boot machine, I found much the same thing. The Gecko rendering engine is very good, and it does a nice job. The wrapper around that engine, though, is flawed. It's no poor reflection on Open Source software that this is so, and if you've been around the community for long, you know this makes its own sense. It is, however, a mark of how far a product has to go if Netscape 4.7 is consistently better behaved.

Next: Getting and Installing Netscape 6 »