editmoin 1.8 was just released, including support for moin 1.6, submitted by both David I. Lehn and Daniele Favara, and URL aliases, as suggested by Diogo Matsubara.
One interesting thing I should mention is that Daniele has sent me a reference to a Bazaar branch containing the change. It’s the first time I receive a bzr (as it’s also known) branch when the upstream project is using a different RCS (Subversion).
I actually intend to port this and other projects to Bazaar, but I’m waiting for the tags support. Hopefully it’s not too far away now.
The All Hands conference from Canonical is over. It was one amazing week in San Francisco, US. Seeing everyone was really great. A lot of new people, many new ideas, old friends… It’s amazing how we get pumped in these conferences.. 
One of the known issues in I’ve been trying to address in Smart for a while is the freezing effect that happens when a very complex upgrading situation (such as a full distribution upgrade) results in a combinatory explosion due to the number of choices to be analysed. Unfortunately, I never had time to really put in practice a reasonable solution for the problem. At this point, the beauty of open source software starts to shine.
A few weeks ago, Eran Tromer got close to the project and started researching and discussing about the issue. Not only that, but he produced actual patches that change the algorithm to prune the search space and find resonable solutions in acceptable timings. These patches were applied into the development version, and included in release 0.50rc1.
The preliminary results are quite impressive. David Farning has tested Smart 0.50rc1 with Fedora in several situations, and reported:
fc4->fc5, fc4->fc6, fc4->devel, fc5->fc6, all calculate updates and upgrades in a few minutes on a vm with 512M, using standard repos + Freshrpms and Dries. Much quicker than with 0.42.
This is really awesome. Thanks Eran!