- From: Daniel Dardailler <danield@w3.org>
- Date: Tue, 10 Oct 2000 15:17:39 +0200
- To: wai-tech-comments@w3.org
Daniel: Al, do you want to recap your recent post [about XHTML 2]? Al: er, not sure I can do that. Let's try this: XHTML 2.0 is our first chance to make functional upgrades in HTML. We need to be stepping back and looking for those functions that aren't working [in HTML 4 + WCAG] as well as we think they need to, and see if we can come up with a better solution in the format. Two major themes in my mind at the moment in this regard: top down decomposition, and unarticulated context dependencies. Top-down decomposition issues come to a head in the current debate in UAWG over the orientation and navigation checkpoints. Would like to have better signposts for the User Agent to follow in constructing and exposing an intra-page table of contents than there are now. Unarticulated context dependencies are like what we addressed by adding HEADERS attribute to directly link TD to its related TH cells. The absense of this kind of clear identification of the immediate logical context of active elements comes up again and again, from "click here" for link text to the overly cryptic field labels that failed to communicate the questions to Kelly Ford's friends in the football pool (see post from earlier in the day). The solution to this problem seems to involve a structure with a graph topology, as opposed to the tree topology of the top-down decomposition. ... Daniel: Is there going to be a "architecture" developed to re-engineer and replace the 'object' element type? Something like X-Link where the basic framework is in place but people can declare their own specialized versions based on what is given? Mimasa: The HTML Working Group in their July face to face looked at just taking the SWITCH construct from SMIL. But it has very multimedia-specific features like system-bitrate. Not acceptable as is. But a solution for content control is sought. Daniel: But Murray Altheim seemed to be hot on the architecture approach when we met back in January. Mimasa: Murray is no longer working on the HTML Working Group. Al: Whatever is worked out for content control, WAI is very interested in it; we depend a lot on the ability to provide equivalent alternatives for medium-specific or sense-specific data.
Received on Tuesday, 10 October 2000 09:17:40 UTC