- From: Carlos A Velasco <Carlos.Velasco@fit.fraunhofer.de>
- Date: Mon, 13 Feb 2006 21:04:38 +0100
- To: Gregg Vanderheiden <gv@trace.wisc.edu>
- Cc: w3c-wai-gl@w3.org
Hi Gregg, Gregg Vanderheiden wrote: > Hi Carlos > > > The problem with the approach you suggest below is that you would have to > specify each delivery unit that conformed. If you had a directory with > pages and all their resources, you can't specify that all the content in > that directory is accessible. You would have to list the individual > delivery units that were accessible and keep updating the list every time > you changed or added a page. Not very realistic for a large web site. (or > even a small one?) Yes, you must do specify these resources. These raises two issues: - It is realistic? Yes, it is, and I know what I mean because we do that almost on a daily basis, for customers with sites in the order of several thousands of resources (or DUs as well). You just need appropriate tools to do the job. - Why such an effort? Because conformance claims must be done with care, and not 'á la WCAG 1.0', when you are dealing with legal environments like BITV in Germany. To avoid liability issues, we must specify what we tested, and how we tested it. Furthermore, there is a movement in Europe towards certification of accessibility, whether we like it or not (I personally, don't). If you certify something, you must make clear what you certify. You cannot get along saying "I certified this site." How to do this? - You might want to look at this very, very, rough draft of expressing HTTP in RDF [1], a coming ERT WG note. - You might also want to look at previous work on regard to scope and sampling of Web sites from UWEM [2]. > With regard to your second question: >> For example, what if I claim the German version of my site (available via > language negotiation) is accessible, but the english (default) is not? > > The working group decided that you can use content negotiation to get an > alternate inaccessible version but the default had to be accessible since > many user agents and users don't know how to do content negotiation. So > the answer to your question is no. I hate to spell out such sentences, but the WG is *plain wrong*. As pointed out by Johannes, content negotiation happens whether you are aware of it or not. The fact that you are ignoring it is not making things more clear. Also think of internationalization where content negotiation is a daily business. I know it is not typical of US sites, but it is highly frequent in Europe. regards, carlos [1] http://www.w3.org/WAI/ER/HTTP/WD-HTTP-in-RDF-20060131 [2] <http://www.wabcluster.org/uwem05/uwem_0_5.html#Scope_of_a_Web_site_and_methods_for_sampling> -- Dr Carlos A Velasco - http://access.fit.fraunhofer.de/ Fraunhofer-Institut für Angewandte Informationstechnik FIT [Fraunhofer Institute for Applied Information Technology (FIT)] Barrierefreie Informations- und Kommunikationstechnologie für Alle Schloss Birlinghoven, D53757 Sankt Augustin (Germany) Tel: +49-2241-142609 Fax: +49-2241-1442609
Received on Monday, 13 February 2006 20:05:14 UTC