- From: T.V Raman <raman@google.com>
- Date: Wed, 8 Oct 2008 06:21:56 -0700
- To: bhawkeslewis@googlemail.com
- Cc: raman@google.com, aleventh@us.ibm.com, wai-xtech@w3.org, wai-xtech-request@w3.org
Algorithm == good --- mandatory --- at this early stage == asking for disaster. Benjamin Hawkes-Lewis writes: > T.V Raman wrote: > > In general all landmark roles should be allowed to appear > > multiple times. Unclear that we can even insist uniqueness for > > role=main --- especially given that content tends to get > > aggregated --- so something that started off as role=main on a > > Web site when used in a gadget is unlikely to be the one and > > only role=main on the containing page. > > I think if more than one role="main" is allowed, there needs to be a > mandatory algorithm for determining to what ancestor the landmarked > region stands in a "main" relationship. > > Just for example: > > 1. First try to assign the relationship to the first ancestor with > role="application" or role="document" (or native equivalents). > > 2. If no such ancestor exist, assign to the root element. > > 3. If the found ancestor already has an assigned element with > role="main" (or native equivalent), flag this second role="main" as an > error. > > With this algorithm in place, you could create a user interface for > users to move to the main area of their current context, rather than > just the next main area in the page. > > In your gadget example, the gadget's role="main" would be defined as > relative to the role="application" enclosing the gadget. > > -- > Benjamin Hawkes-Lewis -- Best Regards, --raman Title: Research Scientist Email: raman@google.com WWW: http://emacspeak.sf.net/raman/ Google: tv+raman GTalk: raman@google.com, tv.raman.tv@gmail.com PGP: http://emacspeak.sf.net/raman/raman-almaden.asc
Received on Wednesday, 8 October 2008 13:23:01 UTC