- From: Matthew Raymond <mattraymond@earthlink.net>
- Date: Wed, 12 Jan 2005 23:54:03 -0500
Jonas Sicking wrote:
> I don't really know enough about how screenreaders or speech UAs work to
> say if this would be usefull to them, or how it would work.
Is there a use case for this? I suppose it's possible for a
screenreader could be affected by the transparency being transparent to
mouse events, but that would depend on how the screenreader works. Does
it even use the same media? If not, web authors could simply specify
different CSS properties for different medias.
> Note though that transparent parts can be tricky when it comes to text.
> You probably don't want just the actual drawn pixels of the text to be
> clickable, but rather the general area that is covered by text. This
> 'general area' needs to be defined though.
That's why I'd rather see a solution that involves the CSS3 |crop|
property. You could use the background image's transparency to define
the area of the element...
| div.myClass { crop: background-image; }
...Or you could have a file specifically to define that area...
| div.myClass { crop: url(area.png); }
In such a way, the web authors define what the area of an element is
rather than having the browser arbitrarily choose for them.
> Also, I'm not sure if the WHATWG mailing list is the appropriate forum.
> This more sounds like something for the w3c CSS WG or the DOM-Events WG.
> Unfortunatly I don't think the DOM-Events WG (nor any other DOM WG)
> exists any more.
This probably isn't the best forum for this. I'll probably right up
a short spec and post it in www-style within the next few days.
Received on Wednesday, 12 January 2005 20:54:03 UTC