- From: Lisa Seeman <lisa@ubaccess.com>
- Date: Wed, 31 Mar 2004 15:15:58 +0300
- To: 'Chris Ridpath' <chris.ridpath@utoronto.ca>, 'Michael Cooper' <michaelc@watchfire.com>, w3c-wai-er-ig@w3.org
The technique may be : if the image has information and message that is not in the linked text -that should be alt text of the image I do not think it is an automated check (except that we are looking into incorporating character recognition for V@ of SWAP - maybe we will get it automated after all :) All the best Lisa Seeman Visit us at the UB Access website UB Access - Moving internet accessibility > -----Original Message----- > From: w3c-wai-er-ig-request@w3.org > [mailto:w3c-wai-er-ig-request@w3.org] On Behalf Of Chris Ridpath > Sent: Tuesday, March 30, 2004 5:51 PM > To: Michael Cooper; w3c-wai-er-ig@w3.org > Subject: Re: [check 7] ALT text can't be empty (null or all > spaces) if image is used as an anchor > > > > Thanks for the update. Sounds like this requirement should > appear in the next draft. > > This has got me thinking that check #7 [1] is not quite > right. What we really want is simply that an anchor must > contain some text. An image used in an anchor can have empty > ALT text as long as there is a text equivalent somewhere > (ALT, TITLE or link text). > > The different placements of text have different meaning, > semantically and in terms of specification purpose. But does > this make much of a practical difference to the user? If link > text is present, do we need another check to make sure it's > put in the proper place? > > [1] http://checker.atrc.utoronto.ca/servlet/ShowCheck?check=7&lang=eng > > Chris > > > ----- Original Message ----- > From: "Michael Cooper" <michaelc@watchfire.com> > To: <w3c-wai-er-ig@w3.org> > Sent: Monday, March 29, 2004 5:03 PM > Subject: RE: [check 7] ALT text can't be empty (null or all > spaces) if ima ge is used as an anchor > > > > > > I think you've got your finger on a problem with the techniques and > > guidelines. The HTML techniques provide a few techniques designed to > prevent > > null link text, but none of them say in so many words that's the > objective. > > I think we decided to move some of that to the Gateway techniques, > > which would describe the general case of hyperlinks independent of > > technology. > But > > that isn't in the current draft of the Gateway either. Also, when we > mapped > > HTML techniques to WCAG 2.0 I think link text was one of the holes, > > where > it > > didn't seem the guidelines spoke clearly enough about. > > > > My personal perspective is that hyperlinks should never > have null link > text. > > How you achieve the link text, from text or image alt text > inside the > link, > > doesn't matter, but it needs to be present. There should be a > > guideline > that > > says so in so many words (along with forbidding meaningless link > > text). > The > > Gateway techniques would probably expound on that, and the HTML > > techniques would cover the cases Chris has been describing > - various > > combinations of images (with and without alt text) and text inside > > links. > > > > Michael > > >
Received on Wednesday, 31 March 2004 08:20:32 UTC