- From: Jason White <jasonw@ariel.ucs.unimelb.EDU.AU>
- Date: Tue, 1 Jul 1997 09:41:59 +1000 (AEST)
- To: dd@w3.org
- cc: WAI Working Group <w3c-wai-wg@w3.org>
It is also necessary to distinguish how the alt attribute should be defined in the HTML specification from the usage suggestions that would be included in style guidelines. However, the treatment of alt by user agents could at least be partially explained within the HTML specification itself rather than in UI guidelines. HTML 2.0 and 3.2 already offer suggestions as to how certain features of HTML should be presented to the user. Thus, a recommendation that the value of alt should be defaulted to the name of the image file, would be reasonable to include within the HTML specification, provided that the WAI Working Group were to reach a consensus in favour of that approach. My personal opinion is that authors should be strongly encouraged to devise their own labels for images, and that in the case of purely decorative graphics such descriptive contents should be kept as brief as possible. However, this suggestion belongs within the province of style guidelines. Another possible point of confusion for authors is that HTML offers the potential for both an alt attribute and an advisory title whenever an image appears as the content of an anchor element. In most instances that I have encountered of proper use of the alt attribute, the link target is described within the alt text. Assuming that readers would not benefit significantly from both an advisory title and a brief description of the image, the value of one of the two attributes would become redundant in such cases. Again, this is probably an issue that should be addressed in style guidelines. I am also in favour of making provision for long descriptions of images, which could be used whenever the meaning conveyed by a graphic can not be stated in a brief description, as is required by the alt attribute. I would rather wait to read the action item before making further comments on this matter. Jason White.
Received on Monday, 30 June 1997 19:42:09 UTC