- From: Al Gilman <asgilman@access.digex.net>
- Date: Sat, 18 Apr 1998 17:22:20 -0400 (EDT)
- To: w3c-wai-gl@w3.org
> commenting on http://www.w3.org/WAI/GL/WD-WAI-PAGEAUTH-0414.html > > _________________________________________________________________ > > 1. Style and Structure > > 1. [Required] > Use elements and attributes that comply with an HTML 4.0 > Document Type Definition (DTD) and CSS-1. > The W3C offers an HTML validation service at > [35]http://validator.w3.org/ that can be used to determine if a > site complies with one of the HTML 4.0 Document Type Definitions > (there are three: strict, transitional, and frameset - see the > validation service for more information). It is recommended, but > not currently required, that pages comply with the strict > definition. > 3. [Required] > Nest headings properly. > Since some users skim through a document by navigating its > headings, it is important to increment heading levels correctly > (H1 followed by H2, rather than H1 followed by H3). Headings used > for other purposes, such as formatting text in a larger font size, > may disorient users; use style sheets for formatting. Note. See > items 9 and 10 in this section. > 4. [Required] > Encode list structure and list items properly. > The HTML list elements (DL, UL, OL, LI) should only be used to > create lists. Avoid using list elements for presentation effects > such as indentation. > [New] Use style sheets rather than HTML attributes to control item > spacing. Note. See item 7 in this section. ASG:: I have difficulty understanding the "required" rating of these three. The first one is too broad to be all required. Suppose someone uses HTML 3.0 <UL PLAIN>. Who is going to find that page inaccessible as a result? Yes, we want to _recommend_ that people use the W3C Recommendation for HTML. But it is hard to defend the choice of HTML 4 specifically as a make-or-break threshold. Having headers skip levels, or building a single element list, is not necessarily improper, at least as I see it. Calling this improper is taking a somewhat narrow view of document structure. Rating these rules as "required" says that if these rules are broken, somebody's ability to access the page _will be broken_. Can we really make that argument in this case? Al
Received on Saturday, 18 April 1998 17:26:26 UTC