- From: Jason White <jasonw@ariel.ucs.unimelb.EDU.AU>
- Date: Sat, 1 Nov 1997 13:08:40 +1100 (AEDT)
- To: w3c-wai-gl@w3.org
I have read the quick summary for web authors which comprises part of the Trace unified guidelines, version 8. I like it very much. A few points could however be noted: 1. Recently in the HTML review process, the issue of quotations was discussed. As an aside to that discussion, mention was made of the practice among certain HTML authors of using BLOCKQUOTE as a means of indenting text, in circumstances where no quotation is actually intended. This may convey a misleading impression to readers, especially in the future, when audio and braille formatting software is used instead of a conventional screen reader. Style sheets will alter voice characteristics and insert braille quotation marks as required whenever a BLOCKQUOTE is encountered (unless, of course, the HTML author supplies her/his own braille or audio style sheet to override the user's default settings in this respect). 2. The WAI HTML/CSS working group has requested that LONGDESC be added to both FRAME and IFRAME. If the initial content of the frame, specified by the SRC attribute, is an image or a script, a textual alternative should be provided and referred to by means of LONGDESC. 3. I think that the HTML/CSS WG has now completed its review of the table markup and it should now be possible to include relevant usage suggestions in the guidelines. 4. Another issue relates to the treatment of navigation bars which, on some web sites, appear at the start of every document. It is hoped that in the future, style sheet mechanisms will enable the reading order of the text to be changed so that such aids to navigation appear at the bottom of the page instead of the top. At present, a person who is using a screen reader must move the cursor past all of the links in the navigation bar prior to selecting any of the links that appear in the main text of the document. I have found this to be rather inconvenient at times. However, there does not appear to be a convenient solution that would apply to today's HTML user agents and access software. For reference, I am using Lynx version 2.7.1 together with a screen reader. I hope to have the opportunity to experiment with T.V. Raman's Emacspeak software at some stage in the next six months, but this may not be possible. I think that T.V. Raman's approach is the way of the future and will supplant conventional screen readers in the years to come.
Received on Friday, 31 October 1997 21:09:01 UTC