This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
In email sent 26 Feb 2009 and on the server at http://lists.w3.org/Archives/Public/www-xml-schema-comments/2009JanMar/0151.html Janina Sajka writes on behalf of the W3C WAI PF working group: Re: Part 2, http://www.w3.org/TR/2009/WD-xmlschema11-2-20090130/ (2) Diagram with built-in datatypes (http://www.w3.org/TR/2009/WD-xmlschema11-2-20090130/#built-in-datatypes) Our reviewers found this diagram very useful (even more than the corresponding one of the last version), and expect it will be frequently consulted by XML developers. However, the information provided in the diagram (in particular inheritance) is not available in a long text description, as required by WCAG 2.0 guideline 1.1 (http://www.w3.org/TR/WCAG20/#text-equiv). Since this is a last-call document, PF encourages the editors to provide a long description of the diagram that includes inheritance information of the built-in datatypes. PF thanks Joseph Scheuhammer <clown@utoronto.ca> and Gottfried Zimmermann <zimmermann@accesstechnologiesgroup.com> for reading the XSD 1.1 Last Call document on behalf of WAI-PF.
Created attachment 656 [details] draft long description for Built-in Datatypes Hierarchy diagram I attach a draft long description of the built-in datatypes hierarchy diagram, for examination and comment by those better versed in current practice than I am. I have attempted to describe the diagram as I would if it occurred in a book I was reading for Recording for the Blind and Dyslexic, but (a) it is some years since I did any reading for RFB, so I'm out of practice, and (b) I do not know if this is considered the appropriate level of detail for long descriptions of images in HTML documents. I also don't know the answers to several questions that arose while drafting. On any and all of them, I would welcome input from participants in WAI, or from potential users or readers of the description. 1 There are no headings in the longdesc document. Should it have an HTML "h1" element at the beginning of the page? Should there be "h2" heading elements at the beginning of the overview and at the beginning of the detailed description? (There are div elements, in the hopes that they may be useful for navigation.) 2 Every occurrence of a box label is hyperlinked to the corresponding section of the spec. This is simple and consistent and means I don't have to decide which occurrence of a label is the 'main' occurrence. If links are distracting to readers, however, then I can imagine readers preferring that fewer phrases in the description be hyperlinked. What does good style in the formulation of long descriptions prescribe here? 3 Should the longdesc document have a navbar of its own? Any other suggestions which would improve the utility of this long description for readers of the spec would be gratefully accepted.
Adopt comment #1
The spec publication system for XSD 1.1 has now been modified to publish a long description for the type hierarchy diagram; the current text is essentially that shown in the attachment above, minus the note about it being a draft not yet reviewed by the XML Schema WG and plus h1 headings for the two main sections. Following some of the links to other resources given at http://www.w3.org/TR/2008/NOTE-WCAG20-TECHS-20081211/H45 I found some useful discussions of longdesc. In an attempt to make the description useful both to users of screen readers and to users of screen magnifiers, I have linked to the long description using both the 'longdesc' attribute on the img element and using a text link in a (new) image caption. I did not locate a longdesc attribute or anything similar on the 'object' element; further tweaks may be required. The longdesc should be integrated into the W3C-internal copy of the status quo draft of the spec, at http://www.w3.org/XML/Group./2004/06/xmlschema-2/datatypes.html as soon as the new version is checked in (any minute now). Input from readers with experience with longdesc and related issues continues to be very welcome and vigorously solicited.
I've now verified that the integration of the long description into the editorial production system appears to have been successful (at least, it works in the status-quo documents), so I'm marking this issue RESOLVED / FIXED. The WAI PF WG has been informed of the disposition of this and their other comments in email archived at http://www.w3.org/mid/F6DBA530-0B5D-4FF2-9CC4-2C02002694A1@blackmesatech.com