W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > January to March 1999

Re: Tooltips - RE: feedback from DHTML developer

From: Al Gilman <asgilman@iamdigex.net>
Date: Wed, 13 Jan 1999 17:23:07 -0500
Message-Id: <199901132221.RAA4265268@relay.interim.iamworld.net>
To: w3c-wai-gl@w3.org
CO::
>  I agree.  There needs to be a CSS attribute to control the rendering of the
>  ALT and TITLE attributes.  Possible values would be:  TOOLTIP or NONE.
That
>  way, the author can control the visual appearance, and if required, the
user
>  can override it.

CmCN::

>Tooltips are an implementation issue, and therefore in the purview of UA,
>not PAGL. And for what it's worth, my personal view is that rather than
>being a core CSS attribute it should be a preference in a UA which
>implements them.

AG::

Chuck's approach, where the author proposes and the user disposes, is
better than trying to say "this is document and that is presentation."
Human communication just doesn't divide up that neatly.  We should have
learned as much from the mess involving ALT, TITLE, tool-tips and image box
or none around the ALT.  The best choice of text to put in ALT given the
Netscape 3 image show/hide protocol is different text from the best choice
given what Lynx 2.7 is going to do with it.

Regardless of what the W3C document says, the ALT or TITLE text string is
defined, in the user's perception, by how it is presented (transient or
persistent, boxed or not) and we can't define document elements and
attributes in a way which is totally independent of how these components
get presented.  Things we relegate to style will blur or outright change
the way that readers and authors peceive the [document fragment] structures
that we are talking about.

CSS is on the right track in that it realizes we need to have a framework
where the author can be thoroughly anal about how her document looks and
the user can still re-map properties as required to make the information
accessible.  I am not saying that the user shouldn't have a UI in the
browser to control this, but that the author is going to care and should be
encouraged to fine tune the default behavior by working with well-posed
flexible structures that can be re-bent under circumstances the author did
not anticipate without breaking the essential connections and distinctions
in the content.

Al
Received on Wednesday, 13 January 1999 17:21:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:46:59 GMT