The W3C CSS Working Group met in San Diego on August 13-15. Vincent Hardy, Steve Zilles and myself were representing Adobe. Quite a lot was accomplished in three short days. The full meeting notes will be available soon from the CSS WG Blog.
There is always a flurry of activity in the working group in the weeks leading up to the face to face meetings. On Adobe’s side we prepared demos of implementation progress on all of our existing specifications and introduced a new pseudo-elements proposal on the mailing list. We also saw a new overflow proposal from Mozilla based on an idea that Adobe and Opera had presented at the previous meeting in Hamburg.
This is my summary of the results of the San Diego meetings as they apply to Adobe’s contributions to the web platform.
CSS Filter Effects
The group decided (along with the SVG WG) to publish the first public working draft of CSS Filter Effects, with Adobe’s Dirk Schulze added as an editor. This document combines SVG Filters, CSS filter shorthands and CSS Shaders and is updated with the current security model that’s been agreed on for shaders.
CSS Exclusions and Shapes
If you have been following the working group discussion on CSS Exclusions and Shapes, you will have noticed that there’s one issue that has come up in every recent meeting. The concern boils down to the fact that implementing exclusions makes absolute positioning more useful, and absolute positioning has issues of its own in adaptive layout. Some members of the working group would rather not implement exclusions without solving the problems of absolute positioning first.
Florian Rivoal from Opera came up with a proposal to add an optional collision avoidance mechanism that could solve some of the problems with absolute positioning, and thus could pave the way for exclusions to move forward. The group resolved to work on Florian’s proposal as part of CSS Positioning, and to revise the issue text in CSS Exclusions and Shapes to reflect the underlying concern.
We quickly resolved to publish a new working draft of CSS Regions to reflect the changes from the last three months of implementation feedback. There have been many updates to the CSSOM section of the editor’s draft, so the new working draft will align more closely with the current thinking of what’s implementable and useful.
I introduced our new proposal for multiple before and after pseudo-elements. We got some good feedback on our syntax choices, and the room was split on whether the use cases in the draft are appropriate for CSS to solve. The people for the proposal are backing the overwhelmingly positive author feedback we’ve gotten on the ideas in the draft. The people against the proposal are also against how these authors are using the existing before and after pseudo-elements. In the end, the group decided that our proposal is something they want to work on, so we will be checking in an editor’s draft (with all of the group’s concerns noted) to the official CSS WG space.
Other Interesting Notes
- The new CSS Overflow proposal from Mozilla dovetails nicely with CSS Regions. The overflow:fragments value is particularly useful for a region chain with a fixed number of boxes. There is some overlapping functionality in how boxes are sized and content is fragmented, so we will be working with Mozilla to determine what can be shared between these specifications.
- Discussions on a number of specifications uncovered a shared desire to lift the restriction on what can follow a pseudo-element in a selector. This may result in more compact selector syntax in several new CSS features.