As you may have already gathered from Kim's blog (yes, we're both from the same province and home town, and no, the tidal bore is not a pig), I've recently been appointed to the Eclipse Architecture Council. It's an honor to be in the company of such great technical leaders, especially my Cloudsmith colleagues, two of whom (Thomas and Ed) are on the council as well.
I've actually been spending a lot of time looking at architecture (or lack thereof) of late, primarily within the Modeling project. As I've mentioned before, vision is one of the key contributors to a successful project, and a guiding architecture is an important part of such a vision. With the number of Modeling projects growing at an alarming rate (60+ and counting), it's going to be increasingly important to "bring order to the chaos", or risk the loss of potential consumers and contributors due to frustration, confusion, or both. Initiatives like Amalgam and the recently proposed Sphinx project certainly help, but there's a lot more that could be done.
Speaking of Sphinx, Stephan Eberle (one the proposed project leads) and I presented a talk at EclipseCon entitled "The Twenty Modeling Things", the slides for which can be viewed at Slideshare or via the EclipseCon session page. The basic idea of the presentation was to propose set of essential services that might one day form the basis for an integrated modeling workbench at Eclipse. Which "things" would you have included? Can you think of other services which ought to be on our list?