Pick Your Distro: Cutting-Edge, Bleeding-Edge, Blunting-Edge

You might remember Bryan Lunduke from his extremely pointed "Linux Sucks -- Let's Fix It" presentation of a few months back. Now he's aiming fighting words at Fedora over F11, and from that I've gleaned a few larger questions about what is the real role of any given Linux distribution.

Some of the feedback in the above-linked article takes Lunduke to task for expecting too much from Fedora. It is, after all, the precursor to RHEL, so some instability is part of the picture. But even with a distribution like Fedora, which is meant to be bleeding-edge and the alpha-state proving ground for RHEL, when does bleeding-edge / alpha-state shy over into flat-out broken? How well should something like Fedora work, and for whom?

A question like this only seems to have an answer in the light of other distributions, all of which are also evolving (or are moving targets, you choose the wording). If you want to avoid patent encumbrances, you pick Fedora; if you want something for commonplace productivity and broad hardware support, you pick Ubuntu; if you want something really stable, you pick Debian. And consequently, with Fedora, the main emphasis has been on keeping things patent-less.

But the problem with Fedora using "free and open" as its big draw is that, when you get down to it, almost every other distribution has something of the same approach anyway. Or close enough to it that most people just walking in the door aren't going to notice any differences that matter to them. It's not as distinguishing as it might seem. (Speaking for myself, I like Fedora. The lack of patent-encumbered pieces has been a bit annoying, but if nothing else it serves as an object lesson in just how much of what we do has a patent encumbrance of one kind or another.)

Other people do things differently, to be sure. Ubuntu segregates patent-encumbered software into its own repository, while such things can only be added to Fedora through the use of a totally separately-maintained, third-party repository. That might sound like a good deal of difference to those in the know, but trust me -- to the vast majority of either potential users or newcomers it's wholly the same. And it's the participation of precisely such people that helps a distribution become more than just ... another distribution.

The problem, as I see it, is that you shouldn't have to make a list of "unencumbered, stable, up-to-date" and then pick only two. The very thing that makes a distro a distro -- its focus, its specialization -- can also be its weakness. It's high time to stop pretending it's actually a strength in disguise, or that it simply means you have a very specialized audience for your work. That approach has dated itself. Doubly so when you consider the possibility that come late this year people with no particular platform commitment will have another fresh new reason to go back to Microsoft's easy embrace.

InformationWeek Analytics has published an independent analysis of the current state of open source adoption. Download the report here (registration required).

Follow me and the rest of InformationWeek on Twitter.