- From: Al Gilman <asgilman@access.digex.net>
- Date: Sat, 8 Nov 1997 10:39:44 -0500 (EST)
- To: w3c-wai-hc@w3.org (HC team)
to follow up on what Hakon Lie said: > ... When you refer to a fixed, rendered version of the document it > certainly makes sense to use page numbers as a reference. We then face > two problems: > > 1 how are the page references encoded? > 2 how is the rendered version referred to? > > The first problem can be solved e.g. with an INDEX attribute in HTML > (or XML, of course). We (HC) should review that. Sounds good. > One possible solution to the second problem is to > use the "ALTERNATE" value for the "REL" attibute on LINK [1]. Most of the time the print version will not be accessible by any URL. One place the print version will most likely be identified is in the content of the Braille document in a section dealing with copyright or acknowledgements. An emerging scenario which automates this more would be that the print version is identified using the information elements of the Dublin Core schema and that RDL is used to associate these information items with the resource containing the etext. > As you note, CSS is not the main responsible. What CSS must offer is > a way to insert values from attributes into the presentation of the > document. This is currently no in CSS2, but we hope it will be there > soon. That is the main point. We have scenarios, both the print page number scenario and table readout scenarios, where it seems to us that we would like to be able to invoke the value of an attribute from the document markup as an item to be included in the rendered version in the target medium. Warning: In the table readout case, we also want to be able to call up the content of a element associated with the current element. Specifically, a body cell in a table will be associated with some set of header cells and we want to be able to lift the text [or ABBR: attribute if present] of these header cells as content to apply in a SPEAKCELL, SPEAKROW, SPEAKCOL or SPEAKTAB generated texts. These generated texts are approximately at the same programming level as page headers and footers. Warning2: In the above case the association from body cell to header cell can be explicit by HEADERS attribute on the body cell or implicit by SCOPE attribute on header cell. We probably need to define a property which is derived as the resultant of SCOPE and HEADERS explicit markup. -- Al
Received on Saturday, 8 November 1997 10:40:16 UTC