W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2001

Re: Transparency and linked objects was Re: javascript dhtml and browser variations

From: David Woolley <david@djwhome.demon.co.uk>
Date: Mon, 2 Jul 2001 23:33:11 +0100 (BST)
Message-Id: <200107022233.f62MXCI06215@djwhome.demon.co.uk>
To: w3c-wai-ig@w3.org
> The screen buffer would tell one the current colour, and this can be
> compared with the 'transparent colour' for a given object.

The screen buffer colour would be the background colour (including
any content underlying the image).  The fact that there is a colour
associated with transparency in GIF is an implementation artefact;
that colour can actually be the same as a non-transparent colour, as it
isn't the colour that matters but the entry number in the colour table.
Any use of the GIF transparent colour is far too implementation dependent
for a graphic formats independent standard.

PNG doesn't have this artefact, as it has a proper alpha (transparency)
channel.  PNG can actually do intermediate levels of transparency,
which is where things get really ambigous if you associate links and
transparent material.  The feeling when this was discussed on the SVG
(www-svg) and/or styles (www-style) list was that PNG with a zero alpha
channel was an extreme case of partial transparency, and therefore ought
to still hide material below it.  On the other hand, the view was that
areas set transparent with style="color: transparent" ought to ignore
mouse clicks and let them through to the underlying layer.

I'd suggest you have a look at the discussions in the archives of those
lists.
Received on Monday, 2 July 2001 19:12:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:13:55 GMT