- From: Jeroen van der Gun <noreplytopreventspam@blijbol.nl>
- Date: Wed, 2 Dec 2009 23:44:26 +0100
- To: "Tab Atkins Jr." <jackalmage@gmail.com>
- Cc: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>, public-html <public-html@w3.org>
2009/12/2 Tab Atkins Jr. <jackalmage@gmail.com>: > True, *but there is still no need to*. The image+caption is still a > valid figure according to the spec. There is no reason to pursue > workarounds, as no workarounds are necessary. Just mark it up as a > <figure> and be done with it. There indeed is currently no need to. However, requiring figure elements to be movable instead of just saying it's a common use, attaches more semantic information to the element that is useful for special user agents. Speech synthesizers may for example skip reading the figures to get faster to the main content. The presence of a figure element then indicates: you do not necessarily have to read this now to be able understand the text after it. So you can read the main content before the figure, if you want to. Take this Wikipedia page as an example: http://en.wikipedia.org/wiki/Obsession_%28film%29 By marking the large infobox on the right as a figure, the user agent knows it can jump over it. (If it does jump over it and then finds a reference to it in the text, then it can always jump back later.) The same applies to the photo of the actors and the DVD cover: these do not have to be viewed before the main content can be read. A speech synthesizers might e.g. simply beep, indicating that a figure is available, and skip it unless the user presses a key indicating that it wants to read the figure now. The above is only possible if the spec guarantees that figure elements are movable. So I think it should. (If the browser jumps over critical parts of the main content with a beep, it will be very annoying.) Jeroen van der Gun http://www.jeroenvandergun.nl
Received on Wednesday, 2 December 2009 22:45:08 UTC