[html4all] distinguishing text equivalents: alt text and descriptive text

Robert J Burns rob at robburns.com
Sat Aug 23 15:15:15 PDT 2008


Hello 4all,

Incidentally, on my proposal that we separate text equivalents into  
two categories — alt text and descriptive text — my preferred way to  
handle the descriptive text is through the immanent metadata  
properties of the media files[1]. If we had HTML UAs that could  
retrieve and present such metadata properties to users upon user  
request, I think that would be idea. I'm not sure the existing  
metadata standards are sufficient for this purpose, but we might  
piggyback something on XMP metadata now part of Adobe[2]. My thinking  
is that since XMP places XML structured metadata into the media file  
format, we could add specific W3C named properties that used XHTML5 /  
XHTML2 syntax within the property.

The short-term approach I'm advocating is to extend the longdesc  
attribute (or another similar attribute like textdesc) to the other  
embedded media elements (e.g., OBJECT). It may be that this short term  
approach is unnecessary since it isn't a feature of the language  
authors have come to expect yet. However, so far Ian has rejected both  
longdesc and UAs providing user access to media metadata[3].

Take care,
Rob

[1]: <http://esw.w3.org/topic/HTML/UANormAndDOMForMediaPropeties>
[2]: <http://en.wikipedia.org/wiki/Extensible_Metadata_Platform>
[3]: <http://www.w3.org/Bugs/Public/show_bug.cgi?id=5755>


More information about the List_HTML4all.org mailing list