W3C home > Mailing lists > Public > w3c-wai-er-ig@w3.org > October 2001

Re: Where does the EARL go?

From: Jim Ley <jim@e-media.co.uk>
Date: Fri, 19 Oct 2001 18:47:45 -0000
Message-ID: <04f301c158ce$8aa59820$ca969dc3@emedia.co.uk>
To: <w3c-wai-er-ig@w3.org>
> 3. Tools that don't support EARL and therefore don't generate the
granular
> claims we find so exciting.  In this case an author will have to use
their
> authoring tool to produce the content then go someplace else to spit
out a
> conformance claim in EARL.  How likely is it that they will do this for
> each page if it is not required by WCAG 2.0?

The assumption here though is that authors use _an authoring tool_ I
don't believe this is true at all, especially within groups particularly
interested in accessibility, those groups, use a suite of tools,
especially due to the low level of accessibility support in the majority
of authoring tools, especially when combined with cost (so older tools
are in use.)  In that enviroment I think most people would, as an site
manager you could easily have a robot at least test the normative parts
of WCAG 2.0 for all pages even if the author didn't.

> How horrible would it be if
> they made a granular claim (e.g., checkpoint specific) that applies to
a
> large set of pages?

I think that would be fine, although it would presumably exclude some
users from a subset of those pages, e.g. if the granular report said it
failed in an area that was only relevant to a couple of the pages, then
the EARL report would still mark them all as inaccessible to its user.  I
see nothing wrong with it refering to all, how does that work though
within the current EARL framework, what wouldearl:testSubject be?

> This is true.  Although, the Bobby stamps don't provide granularity and
are
> not machine readable.

Are they not? there's a reasonably clear set of methods of adding them to
pages, which does make them in effect readable.

> Yes, the event horizon.  WCAG is looking at a conformance scheme today
and
> what they should require of authors.  The questions that came up:
> 1. do we require all conformance claims in markup?

Require, with difficulty I think unless the EARL can be generated
automatically, so as not put an undue burden, otherwise I feel the EARL
would just be added, as you say we need to trust the author, if it's a
struggle to create EARL, it won't be done.

> They are good questions.  My personal answers are:
>
> 1. I would like to, but this only seems possible if all WYSIWYG
authoring
> tools are able to generate the EARL conformance claim for the author.

I don't see WYSIWYG as necessary a single seperate program (or suite)
would be better, rather than trying to generate new ones for each
authoring tool, or encouraging authoring tools to do it. (which obviously
don't exist, I'm slightly confused as to what you refer to here, I'm
assuming just "authoring tool")  If we concentrate too much on authoring
tools it increases the complexity and difficulty and time it would take
to develop them which would be better served concentrating on the content
itself.

> The upshot of all of this is a wonderful amount of metadata for us to
> search through.  The scenarios for using it keep growing. It's very
exciting!

but a long way off.

Jim.
Received on Friday, 19 October 2001 14:49:41 GMT

This archive was generated by hypermail 2.2.0 + w3c-0.30 : Thursday, 9 June 2005 12:10:39 GMT