- From: Charles McCathieNevile <charles@w3.org>
- Date: Mon, 14 Jul 2003 18:47:47 -0400 (EDT)
- To: Gregg Vanderheiden <gv@trace.wisc.edu>
- cc: <w3c-wai-gl@w3.org>
On Mon, 14 Jul 2003, Gregg Vanderheiden wrote: > >Good point. > >And the person can just point to the dictionary they choose in their list >(since they have to look at one anyway to see if the ABBV is in it.) > >Hmmmm. I hate to keep adding things to the header of each pages. But if we >just do it on the root home page (that is the root of the URI) then that >should make it easy. This is not as easy as it seems - what is the root of the guidelines internal drafts? What about a site that is available in 11 languages such as http://www.ascii.be or http://www.europa.int ? Not all people with sites hosted by tripod will use the same dictionary, but some will. Actually this is one of the things the semantic web is meant to help with. You can associate information with your page, including what dictionary you checked it against. I recently made an Xform to allow people to create EARL evaluations (conformance statements_ - I wanted to enhance the real-world functionality a bit before announcing it, but anyone keen to get an xforms capable browser and play around (http://www.xsmiles.org is one, and for people with flash the DENG browser produced by Sebastian Schnitzenbaumer and his colleagues at mozquito provides the functionality apparently inside their normal browser) there is some bare-bones stuff at http://www.w3.org/2001/sw/Europe/200305/axforms Similarly, we could use EARL or something like it to assert that the language used in a page or group of pages conforms to the list of terms in a particular vocabulary - whether that is the wordnet dictionary available in RDF, some other dictionary, or a particular controlled vocabulary used in some industry sector. with the use of technologies like Annotea, the focus moves from the author having to put everything in the page, to somebody (it could well be the author, but it could be a third party) putting it on the web in machine-treatable RDF, and then using an RDF-aware search engine to fin the kind of data you are looking for - optionally including information about whose information you are prepared to trust. cheers Chaals
Received on Monday, 14 July 2003 18:47:48 UTC