The Nautilus file manager is just one part of the Eazel plan for the Linux desktop. The company is taking a two-pronged approach to making Linux an attractive choice for desktop users. Eazel won't be depending on software sales, for instance, to provide revenue as it contributes to the Linux community.

"When it's software that needs to be added, we'll make free software," Adler says. "When it's services, we'll make services that involve Eazel, and that gives us an opportunity to make money."

Adler sees Eazel's future projects as involving a continual reassessment of where Linux lags in usability and desirability to the average desktop user.

"I see this as an iterative process," he says. "We're going to keep doing more work on the areas that prevent people from choosing Linux as their desktop."

Current services Eazel intends to offer include a software inventory system that provides an easy way to remain up-to-date on available packages, as well as a web-based storage service. According to Adler, other services are in the works but won't be part of the initial outlay provided by Eazel.

Early Choices

According to Adler, picking GNOME as the Linux desktop project was a relatively quick decision, despite the existence of another well-developed Linux desktop project: KDE.

"It was definitely dependent on initial conditions," he says, noting that "the main issue for us wasn't making the right choice or the wrong choice. The main point was getting moving building something."

The decision was made easier at the time by the uncertainty the Eazel team had over the now largely resolved questions regarding the license of KDE's foundation: Trolltech's Qt toolkit.

"The last thing we wanted to do was build something that turned out not to be free software," says Adler.

On the other hand, Eazel's programmers were extremely comfortable with their use of C++, and even with the licensing issues, Adler notes that there was some concern that the company might have fit better with KDE.

Eazel was only warily accepted by the core of the GNOME community at first. According to Adler, concerns existed within the GNOME community about Eazel's long-term commitment to the project, and how maintainable any code they produced would be should other GNOME programmers need to finish the work they'd started.

"There was a lot of pressure from the other folks in the GNOME project to do things in a way that matched the core GNOME pieces. This wasn't going to be just another module, this was going to be a part of the core GNOME...they convinced us to code in C, and they also convinced us to pick up the work someone else (Elliot Lee of Red Hat) had already started."

The issues were ultimately resolved with Eazel's decision to switch tracks in programming choices to conform more with the existing project "We really wanted to be a part of the community, not some sort of rebel faction."

Adler also observes that the early conflicts between Eazel's programmers and members of the GNOME community are periodically distorted in terms of their importance:

"People are always looking for conflict. That was a pretty short episode, very early in development," he says.

He characterizes current rumored conflicts between several key players in the GNOME community as being less centered around project-endangering differences, and more around day-to-day conflicts that arise in most software projects. His attitude toward his work with the GNOME community reflects his belief that the initial period of mutual uncertainty is over:

"It's a total joy to work on this project," Adler says. "We hope we're setting an example by doing the work out in the open."

Next: Coding in a Dynamic Environment: Working with GNOME, Working with Linux »