- From: Jason White <jasonw@ariel.ucs.unimelb.EDU.AU>
- Date: Sat, 13 Sep 1997 11:24:00 +1000 (AEST)
- To: WAI Working Group <w3c-wai-wg@w3.org>
Given the urgency of the HTML 4.0 review, there are a few points that I think could be clarified. Firstly, on the Formats and Protocols working group page, it is stated that there are issues concerning alternate style sheets and media types which are yet to be resolved. I would like to know what problems remain to be dealt with in these areas. Secondly, should this working group recommend that a TITLE attribute be added to the FRAME element so that authors can specify a meaningful label for each frame? I understand the present situation in this respect to be as follows. The NAME attribute is intended to define a target to which reference can be made via an URL fragment, for example in an anchor. NAME is an optional attribute, in that it is implied in the published DTD. Clearly, the value of NAME is not intended for human consumption, and nor is the value of SRC, which gives the URL of the document that comprises the content of the frame. This situation appears to be unsatisfactory. Lynx version 2..7.1 provides an interesting illustration of how a user agent can attempt to make the best of it. Lynx displays a document in which frames are defined as a series of links. The text comprising each link is derived as follows: if the corresponding frame includes the NAME attribute, then that name is displayed as the text of the link (E.G. "FRAME: INDEX", where "INDEX" is the value of the NAME attribute); but if NAME is not specified, then the URL of the target document, as given in SRC, is presented (E.G. "FRAME: "sidebar.htm"). As suggested above, a TITLE attribute could be added so as to enable a more adequate label to be associated with each frame. It would be best if TITLE were required, but assuming this is not possible in the short time available to us, I would suggest that a default behaviour be defined, such as that exemplified above, for representing frames in a non-visual environment. The foregoing suggestions are only tentative and I would appreciate comments from other members of the working group.
Received on Friday, 12 September 1997 21:24:13 UTC