I’m in Nürnberg, Germany for the SVG working group meeting, which just finished last night. I have to say (without trying to sound too self-congratulatory) that we really put in a lot of effort, finishing way into the evening on some days, with encouraging results that will stand us in good stead for republication soon. The majority of the time was spent working on improving the SVG Tiny 1.2 spec and test suite — something not nearly as fun as discussing new work, but it really had to be done. (Note that there’s now a single page version of the spec, which should help facilitate review, searching, printing, etc.)
In response to implementor feedback, the Connection interface of SVG Tiny 1.2 is gone. Connection always was a controversial interface, with many commentors pointing out that it shouldn’t have been something that SVG defined. At the time it was introduced, long before I was a member of the working group, there really wasn’t another forum that was appropriate for a network interface to be specced, so I can understand why it was, especially as, despite the views of some people, SVG really was (and is) being used an application platform, and not just a static graphics language. The Network API spec in the WebApps WG isn’t really progressing, but there is some new, similar work happening in HTML 5, so I hope that these become unified.
The arrangement of the Global and SVGGlobal interfaces in the SVG Tiny 1.2 uDOM is being cleaned up to fit in better with other specs. So instead of SVGGlobal inheriting from this empty Global interface, we’re going to state that the default AbstractView object (i.e., the global ECMAScript object) also implements SVGGlobal, and not make SVGGlobal inherit from anything. This parallels the way HTML 5 defines Window.
The ‘static’ property from the old SVG 1.2 Full draft will be making a comeback in 1.2 Tiny in some form, in response to requests from implementors.
We had an interesting discussion on layout functionality that we’d like to see in the new Layout module (which is the primary reason I’m here). There seems to be support in the group for some high level layout container support in conjunction with some lower level layout control specifications, such as expression-based positioning of objects.
John Daggett from Mozilla dropped in for a day and gave a presentation on the CSS3 Fonts draft, including an engaging overview of the background and technical issues surrounding web fonts. (And I must say that that spec is one of the most visually pleasing documents I’ve seen come out of W3C for a while.)
SVG Open 2008 begins tomorrow, and for a change, I’m not presenting anything. Here’s hoping attendance is up on last year!