Re: DC2002 and Accessibility Metadata

I think this is taking an overly pessimistic and defeatist approach. My
experience with large producers of metadata is that they try to take it
seriously. Interestingly, those are the organisations I see taking
accessibility most seriously. Often they are large, slow-moving institutions,
and they don't rush out to implement every new idea the day they hear of it.
But they are often the type of organisation who has a real commitment to
making their services accessible (for a variety of motives) and they are
starting to get it right.

Dublin Core is not necessarily author-supplied. It can be provided by anyone.
I think one of the intersting features of EARL, thanks in large part ot Sean
Palmer, is that it requires a statement of who is making a claim - whether
that is a person, a template (i.e. an RDF equivalent of "the author forgot to
change the default title" in HTML pages) or an automatic tool.

The availability of accessibility testing tools (imperfect though they
currently are) makes me think that in fact this data might be expected to be
a little more reliable than other metadata when it comes from "ordinary
authors". (Most good data comes from people who are professionals in
producing the data - improving what is produced by people editing their pet
dog's web page is a challenge for all fields of the Web).

So I think that this is a fairly sensible initiative, and if anything
slightly late. Don't forget that people on this list have been looking for
web content and services accessible to them for several years already, and
are suffering becuase the right discovery tools are one of the big gaps in
what is currently available. (Pace ableFish and similar attempts to close
that gap).

Cheers

Chaals

DISCLAIMER: Liddy is my Mum, so maybe I am just backing her up instinctively.
I don't think so - in my judgement there are sound reasons for supporting the
work she is doing.

If you think my technical judgement is really clouded because it is my mother
please let me know (off-line).

(If you think I am permanently talking rubbish, I assume I already know, or
will find out in due course...)



On Sat, 14 Sep 2002, David Woolley wrote:

>
>> seem to be widespread support for development by DCMI of
>> accessibility metadata. What is required is a little more than
>> interoperable discovery metadata.
>>
>
>I though Dublin Core was about author provided metadata.  It seems
>to me that such data for accessibility would simply further ghettoize
>accessible web sites.  Ordinary authors will ignore it, or, if forced
>to include it, provide token "safe" values, thus degrading the value
>of the data.  If may even be seen as an extra cost in making a site
>accessible, making accessibility weaker in a cost benefit analysis.
>

-- 
Charles McCathieNevile  http://www.w3.org/People/Charles  tel: +61 409 134 136
SWAD-E http://www.w3.org/2001/sw/Europe ------------ WAI http://www.w3.org/WAI
 21 Mitchell street, FOOTSCRAY Vic 3011, Australia  fax(fr): +33 4 92 38 78 22
 W3C, 2004 Route des Lucioles, 06902 Sophia Antipolis Cedex, France

Received on Saturday, 14 September 2002 18:55:58 UTC