- From: Leif Halvard Silli <lhs@malform.no>
- Date: Fri, 16 Jan 2009 21:34:53 +0100
- To: chrisreeve15@yahoo.com
- CC: public-html@w3.org, bzbarsky@mit.edu
Chris Reeve 2009-01-16 18.15: > Boris, and Leif, in response to WCAG 1.4.5, is there a rule on > what is a good/bad "alt" tag for images of text. It is some times since this thread (ore even I myself ...) last was active ... But supposing you ask how to apply WCAG 1.4.5 [1] for the below scenarios ... > Scenario A: > > A photographer takes a picture, and surroundings have images of > text in it. But the intent and writeen document only shows the > integrity of one image (no text is inside it) I guess you have to look at the photo and make a judgement about how important the text bits are, af they are you must somehow quote those text bits in the alt text. > Scenario B: > > A photographer takes a picture, and surroundings have images of > text in it. This time the intent of the image and the document > suggests that multiple areas in the image have image of text > (1.4.5) and the site suggests that multiple areas should be > used. You could use the OBJECT element and present the the complete "story" of the image as fallback, in a more structured way. > I also have a text inside a logo and WCAG 2.0 suggests that > text inside logos should comply with 1.4.5. When 1.0 was out it > used to be "Company logo"? What is the problem here? You could use <img src=image alt="ACME Corp. The company that makes everything!" >. I'll add a Scenario C: An online library of scanned books, where each page contains the scanned image of a page, plus the OCR version of its content below the scanned image. See for example Project Runeberg [1], which uses alt="scanned image" for each scanned image, which seems very meaningful to me. [1] http://www.w3.org/TR/UNDERSTANDING-WCAG20/visual-audio-contrast-text-presentation.html [2] http://runeberg.org/display.pl?mode=facsimile&work=scanbrit -- leif halvard silli
Received on Friday, 16 January 2009 20:35:37 UTC