At the beginning of a new development cycle, Ubuntu developers from around the world gather to help shape and scope the next release of Ubuntu. The summit is open to the public, but it is not a conference, exhibition or other audience-oriented event. Rather, it is an opportunity for Ubuntu developers -- who usually collaborate online -- to work together in person on specific tasks.
Small groups of developers will participate in short Forum and Workshop (formerly called "BoF"/Birds-of-a-Feather) sessions. This allows a single project to be discussed and documented in a written specification. These specifications will be used for planning the new release of Ubuntu, as described in FeatureSpecifications and TimeBasedReleases.
What was striking this time, compared to four or five years ago, the last time I attended a UDS, was the greater number of converging threads -- particularly in terms of hardware. OEM's and in particular, ARM hackers.
"Ubuntu developers" doesn't really describe the range of people there. A Linux distribution touches on a wide variety of vendors, upstream projects, designers, community enthusiasts, desktop environments, advocates of particular hardware (netbooks, tablets), or users (accessibility, i18n), user communities (education). It is loose, like an unconference, but not really improvisational, because almost everyone, including me, is showing up with a specific, probably complicated problem that usually involves complex coordination with other hardcore geeks in attendance. But the structure and tone of each public session varies wildy.
Nonetheless, both times I've attended I've felt like it would be valuable for people who have expansive ambitions for "open" processes in education to attend one of these things. Just getting a sense of the scope of what goes into a Linux distribution is both humbling and inspiring. It is vastly, vastly, vastly more complex and successful than the equivalent in any other field. Not that more than a tiny fraction of that happens at UDS, but it opens your eyes.
Also, in the case of Ubuntu, a stunning example of large scale venture philanthropy.
My big takeaway about Ubuntu in general is that the customization necessary to run Ubuntu (or anything else) on each piece of ARM hardware should give Canonical a good source of income. That is, ARM devices are much less generic and standardized than PC's. So if you are putting out a netbook that you'd like to ship with a Linux distro on it, you can't just grab a standard ISO like you do for PC hardware, and it will make sense to pay Canonical to do the necessary customization.