Fedora Core 3 and the community

The "about" page adds this:

In this context, it is interesting to consider the Fedora Core 3 plan, which was posted on July 2. The plan calls for all kinds of interesting things, including:

• GCC 3.4 as the standard compiler.
• GNOME 2.8 - which is not yet released.
• KDE 3.3.
• Evolution 2.0
• Another attempt at SELinux, with a less ambitious, less intrusive set of policies.
• Indic language support

And a lot more. It looks like a bunch of good stuff.

One should note, though, that the scheduled date for the first test release is July 12 - ten days after the announcement. Before the plan announcement, there was very little public discussion of what FC3 was going to contain. At this point, there is not a whole lot of time to "include external developers in the process of making technical decisions." Instead, it looks much like, once again, the core decisions have emerged in final form from a smoke-filled room at Red Hat headquarters.

Let there be no mistake: Fedora Core is an unmitigated good thing. Red Hat is giving the world a high-quality distribution with (mostly) highly current software and a certain degree of visibility into the development process. One should not complain about such a gift; we are certainly richer as a result of it.

But Fedora clearly is not meeting its stated goals of being a community project, and, apparently, it is not even making much progress in that direction. Red Hat would do well to clarify its plans for Fedora at this point. If Fedora is to be a community project, interested developers need to see some progress in that direction. Opening up the promised CVS server would be a good start. Another promise that would be good to keep is this one:

The FC3 plan was clearly not developed in this way. The formation of the promised technical committee, which is supposed to include outside members, would also be a good step.

If, instead, Red Hat plans to keep Fedora in its current form (essentially, a development and testing platform for technologies eventually slated for the enterprise products), it should say so. Red Hat would be entitled to take this position, and, certainly, large numbers of users are content to run a Fedora distribution which is developed in this way. Who can complain? It is a free, high-quality distribution with good security support. But outside developers who would like to participate in its creation have a right to know whether (and when) that will be possible.