- From: Jukka K. Korpela <jukka.k.korpela@kolumbus.fi>
- Date: Mon, 09 Dec 2013 14:23:42 +0200
- To: HTMLWG WG <public-html@w3.org>
- CC: HTML Accessibility Task Force <public-html-a11y@w3.org>
2013-12-09 13:36, Steve Faulkner wrote: > HI all, have added some advice to html 5.1 text alternatives for <img> > elements when they are used as image maps > > http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#image-maps-0 > > At your leisure, feedback welcome and appreciated. The addition seems to be a copy of the example at http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#image-map except that alt="Four shapes are available: a red hollow box, a green circle, a blue triangle, and a yellow four-pointed star" has been replaced by the shorter alt="Four shapes are available:". This is an improvement, since color is not part of shape. But wouldn't it be better to improve the example where it is now and just link to from the lengthy discussion of the alt attribute? Besides, the example would still be odd. The alt attributes of the area elements mention colors, and the first one does not really describe the pattern: alt="Red box." should be alt="Square with a square hollow in its center", if the intent is to specify the shapes. (I don't see any good reason for the periods at the ends of alt attributes here. They don't cause harm I suppose, except that they make the reader wonder whether I should really use periods there, and why.) The example as a whole is rather artificial. It is difficult to imagine a meaningful context where you would present such a choice of shapes. And it would be more accessible (and simpler to code, hence safer against authoring mistakes) if presented simply as four image links. But, admittedly, it is difficult to find good examples of image maps. How about a geographic map? Say, a map of Australia (relatively simple, with not too many states), acting simply as an "active map" where you can click on any state to visit a page about that state. The coords attributes would have many numbers, but that would not really add much to the structural complexity. The added text before the copy of the example says: "If an|img <http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#the-img-element>|element has a|usemap|attribute which references a|map <http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#the-map-element>|element containing|area <http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#the-area-element>|elements that have|href|attributes, the|img <http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#the-img-element>|is considered to be interactive content. In such cases, provide a text alternative that acts as a group label for the linked regions of the image." The end of the sentence is then repeated after the image, followed by the text "The|alt|attribute on each of the|area <http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#the-area-element>|elements provides a label for each linked region". The formulation could probably be made simpler (more accessible), e.g.: If an img element has a usemap attribute and is thus associated with an image map, the alt attribute value describes the image as a whole in the context. This means that it acts as a caption for the collection of the alt attributes of the area elements, so that all these attributes together can be used as a captioned menu of choices, in a situation where the image is not seen. (I would prefer "caption", or maybe "heading", to "label" here, to avoid confusion with labels of controls defined via <label> markup.) -- Yucca, http://www.cs.tut.fi/~jkorpela/
Received on Monday, 9 December 2013 12:24:14 UTC