W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 1998

Re: results: implementing 'invisible' tags

From: <nir.dagan@econ.upf.es>
Date: Thu, 5 Nov 1998 22:57:39 +0100
Message-Id: <H0000e2200d2fe44@MHS>
TO: c.g.colwell@herts.ac.uk
CC: w3c-wai-gl@w3.org
The guidelines should mention that HTML is an information rich 
language that implementation is typically media dependent. For example, 
that TABLE's summary and longdesc is markup that is typically implemented 
better in aural browsers than in graphical ones. That META is better 
implemened by robots compared to Netscape, but still authors write it
since *some* user agents make use of it, and it doesn't harm the others.

This applies to most of HTML4.0 new features. 

I disagree that there are "accessibility only" features in HTML. 
And I think that the accessibilty guidelines over-emphasize 
accessibility.
Longdesc is a link to a document that enables search engines 
to index your important images. TABLE's summary or the association of 
headers with cells is part of the information of the page. 
If a cell is associated with a header cell it is true for 
all users, although some users may infer the association 
from a two dimentional layout if they happen to use a medium
that provides such layout.

The guidelines cannot say that something will not be visible to the 
page-author, as the guidelines cannot assume that authors use 
graphical browsers. How about "Note that the SRC attribute in IMG 
is required. Although the ALT attribute is presented by default 
in most media, and probably so in your browser, some browsers for people 
with special needs (e.g., a need for a better browser) render a broken 
image icon instead of the ALT text when a SRC attribute is not present".

It should be noted that some browsers do support longdesc, so it is not 
"will be available".

Regards,
Nir Dagan.
http://www.nirdagan.com

> Some participants have been confused when they have used particular tags
> suggested by the Guidelines and then not been able to see or test the
> result with their browser (usually Netscape or IE).
> 
> This occurred particularly with tags that were perceived to be 'special'
> for accessibility.  For example:
> LONGDESC,
> TABLE summary,
> ID / headers (for table cells)
> 
> I suggest that the Guidelines (techniques) include notes that point out
> that particular techniques will not be 'visible' to them, but will be
> available to future browsers and/or assistive technology.
> 
> Chetz
> PS the tags mentioned above are the ones that were highlighted - there may
> be others.
> 
> 
> 
> 
> Chetz Colwell <c.g.colwell@herts.ac.uk>
> Sensory Disabilities Research Unit
> Department of Psychology
> University of Hertfordshire
> Hatfield AL10 9AB, UK
> 
> Tel: +44 (0) 1707 284630
> Fax: +44 (0) 1707 585059
> 
Received on Thursday, 5 November 1998 16:57:59 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:46:58 GMT