Before I move on to criticism let me say that I think quickly looks like a reasonable good approach for rapid development for interpreted languages. It will be interesting to see how it develops and there is pressure to build up a more feature rich IDE interface on top of its core functionality.
But on to the criticism....I think name spacing at the distribution level is probably short sighted as a policy model to follow. Now GNOME and KDE namespaces for templates make sense...as these upstream projects do define a framework on which to hang applications. What does it mean to be a fedora project or an ubuntu project or a debian project really?
Distributions do not define a technology framework. I understand that Canonical is pushing launchpad backed bzr and desktopcouch pretty hard because they think they are best of breed technologies...and I respect that. But does it make sense to brand that combination of technologies as Ubuntu specific? And then to encourage other distributors to brand a different stack of technologies as yet another template and have competing technology frameworks? I don't think it does. For new projects, a distribution specific namespace is an arbitrary definition at best. At worst its going to end up being a barrier to adoption across distribution boundaries. What independent project writer really wants to see their end-user application bound up in distribution specific naunces or dependant on distribution patches to upstream codebases? Amd what I would really hope that Canonical realizes is that pushing technologies like desktopcouch needs to be done in a brand agnostic fashion. The last thing Canonical needs is for a competitor to think of desktopcouch as Ubuntu-centric...and then introduce a competing framework for application writers to choose.