- From: Boris Zbarsky <bzbarsky@MIT.EDU>
- Date: Wed, 24 Nov 2004 23:49:11 -0600
- To: www-svg@w3.org
14: Is there a reason to introduce <xa> instead of simply using <a> with xlink:type set accordingly? That would allow UAs that implement multi-target xlink to reuse their standard xlink implementation for this functionality. The case when there is an xlink:title attribute and a <title> element is underspecified -- what is a UA to do? What are the conditions under which a descendant <title> should be looked at by <loc>? Clearly, if <switch> is the only thing between them. Can there be any other cases? If so, they should be documented. If not, it should be specified that the only titles to be looked at are those that are direct children of the <loc> or direct children of a <switch> that is a child of the <loc>. The section describing the behavior must mean that this is what happens when the link is activated, right? It should say so. The behavior for "2 or more" demands that the size be such that the longest string not be clipped. When the longest string does not fit on the user's display device (monitor, braille reader, whatever), this requirement cannot be fulfilled. I think it would make more sense to require that the UA present the options to the user in the most reasonable way it can... The exact behavior is already not defined enough for pixel-level interoperability, so the additional wiggle room this adds, if any, should not be a big issue. -Boris
Received on Thursday, 25 November 2004 05:51:23 UTC