- From: Loretta Guarino Reid <lorettaguarino@google.com>
- Date: Thu, 17 Jun 2010 11:36:15 -0700
- To: PFWG Public Comments <public-pfwg-comments@w3.org>
i accept this response. > > Comment 115: Rendered text > Date: 2009-04-16 > Archived at: http://lists.w3.org/Archives/Public/public-pfwg-comments/2009AprJun/0063.html > Relates to: Accessible Rich Internet Applications (WAI-ARIA) 1.0 - 4.2.7.3. Text Equivalent Computation <http://www.w3.org/TR/2009/WD-wai-aria-20090224/#textequivalentcomputation> > Status: Alternate action taken > > ------------- > Your comment: > ------------- > 4.2.7.3 Text Equivalent Computation > > Is it really a good idea to use rendered text instead of the DOM? This > creates yet another interpretation of the html, and seems like it could > lead to subtle bugs. > > -------------------------------- > Response from the Working Group: > -------------------------------- > --- NEW RESPONSE TO COMMENT REJECTION --- > > It's come to our attention that a related note was out of place. That > error has been corrected. See the following note, now associated with the > text in question. > > "Note: Though a user agent may make efforts to compute a text alternative > from CSS-generated text in the absence of text content determinable from > the DOM, authors should not provide text through a style sheet, as a user > agent may incorrectly determine the text alternative." > > http://www.w3.org/WAI/PF/aria/roles#textalternativecomputation > > Although we encourage authors to comply with WCAG principles, the ARIA > spec is not intended to be used as an enforcement of WCAG. Our end goal is > to provide an accessible solution to users, so we encourage UAs or ATs to > attempt to bridge any gaps in inaccessible content. The note is to > discourage authors from relying on that technique, but because > CSS-generated text is visible, if an AT or UA can include that text in the > accessible name, it should. > > > > > > --- PREVIOUS RESPONSE TO COMMENT --- > > > > Consider these style rules and fragment of HTML markup: > > p[title]:before { content: attr(title); } > > <p title="Step 1:">Put up the tent.</p> > > A user agent (browser) will render this as: > > ... > Step 1: Put up the tent. > ... > > Because the title attribute is only included if there is no other text > content, the DOM representation of the text equivalent is "Put up the > tent." > > Though a user agent should make its best effort to compute a text > equivalent from CSS-generated text in the absence of text content > determinable from the DOM, authors should not provide text through a style > sheet, as a user agent may incorrectly determine the text equivalent. > > Authors need to be aware how the label text will be computed by the user > agent, so as to structure markup and CSS appropriately. Since they are > creating the CSS content rules, they must have an idea of how the entire > string will look. > > In the case of a text equivalent checker tool that authors might use to > confirm that their labels are properly encoded: the checker tool must, > like user agents, look for any CSS content and insert/append it as > necessary. >
Received on Thursday, 17 June 2010 18:36:46 UTC