- From: Charles McCathieNevile <charles@w3.org>
- Date: Thu, 15 Mar 2001 08:01:14 -0500 (EST)
- To: "Bailey, Bruce" <Bruce_Bailey@ed.gov>
- cc: <w3c-wai-gl@w3.org>, "'jim@jimthatcher.com'" <jim@jimthatcher.com>
Although a server side map is just a collection of points, so in theory could be broken into a big pile of 1x1 squares, there are practical reasons for having them in some cases. For example, mapping applications will generate a map on the fly from a much larger store of information, send it down, and then when a point is selected on the map generate a new one based on that point. The practical value of a client-side image map is in part that it can have alternatives associated with the destination of each hotspot. But in fact is not always very easy to provide useful alternatives - for example in the mapping scenario. An alternative technique for doing this might be as follows: Use SVG to create the maps. Make each street a "hot" object, with the default activation behaviour to get a new map based on that point. The basic algorithm can be based on the client side on DOM changes or animations to a use element. On the server side you could implement the same algorithms, but you might choose to use a different processing language (or not...) Charles McCN On Wed, 14 Mar 2001, Bailey, Bruce wrote: Even allowing for deprecated equipment, can anyone think of real-life examples where the exception provided for in 9.1 would be needed? I can image that a server side imagemap could have regions defined by, for example, ellipsis's. Does any one know of configurations where this is true in actual practice? And, as Jim points out, a polygon can get arbitrary close (at least for the need of screen resolution) to an ellipsis, so why allow the exception any way? Are servers side imagemaps mentioned in WCAG 2.0?
Received on Thursday, 15 March 2001 08:01:23 UTC