Re: log of chat: EARL and TDPL

> 21:54:47 <wendy> we should have thought about users as well as devs, but
we
> focused on devs and now jim ley is upset.

I should've always explained myself better, what exasperates me, is you've
all put a lot of effort into EARL, and come up with what seems a good
technical solution (I'm not completely up on RDF, but can pretty much grok
what seems to be going on.)  The problem was that I suddenly discover that
you don't seem to be too sure what EARL is the technical solution to.  Too
much technical work has gone on, without the user implications being fully
thought through.  I've been in projects like this, I'm technical and I've
learnt from my hard learnt mistakes, this has led to me wasting lots of my
time, I hope none of EARL has been wasted.

By the way, in case people wonder why I'm reading/posting, I have an
Evaluation tool (on the use of scripts),
http://jibbering.com/accessibility/snork.html, and am also interested in
creating an extension to my browser that uses EARL reports.

> 22:17:36 <sbp> well, for a start: bug with 1.0: how do we make
conformance
> claims about a whole site?

Definition of site needed? I'm assuming domain, but I don't think that's
safe, too many "sites" are widely spread with what are really subsites
within them, for example Government sites have often only made part of
their sites fully accessible and different departments with different
criteria are still running off the same www.ukonline.gov.uk so this by
domain wouldn't be appropriate.

By startsWith, I assume you mean that to be a url fragment? so
log:startsWith "http://example.org/chickens/"
would refer to http://example.org/chickens/bantam.html but not
http://example.org/dogs/  ?

> 22:30:40 <wendy> the user could also say, "don't show me sites that
don't
> make a claim, since they'll likely be useless to me."

My own personal direction with this would not be in the search engine, but
would be in the user agent, so is more generic, I can apply it to any page
of links, (for example a list of where to learn more about HTML authoring,
I can prioritise those to the ones that have provided EARL.)  I already
have something similar, I currently check that links aren't 404's and also
get the file size and content type, and even this minimal information is
improving my search efficiency.  EARL would open this up to further areas.
It may be nice to have it as wide as possible, so it therefore is as
useful to as many people, whatever their accessibility interest.

Okay I've also thought of some "Opportunities" for EARL's links to HTML:

I have a site, it provides presentations online - Generally it only
provides Graphic Slides (without description) and Audio, so is in general
pretty inaccessible - this is for simple cost reasons and there's a
limited audience (it's a hosted service we generally have nothing to do
with the production, they just phone the computer.)  So I'd have to say
http://www.domain.invalid/conference/ "inaccessible"  However on some of
them we have the material available to provide descriptions on the images,
closed captions on the audio etc. etc. and would want to let the users
know, but would still not want to go to the effort of individually stating
the checks on all pages (I assume it's not a proposal that you have to
manually check each page when they are all generated from a similar
template?)

What solutions could there be in EARL for this?

My own thoughts were an EARL "testPage" that just referred to a "template"
or "sample" or "class" or something I don't know the word, but this class
of testpage could then be referred to throughout the site and linked to
with a LINK element.  This would let you create a series of "classes"
being different Accessibility successes, and then reference them from
anywhere on your site, without worrying about the urls in the EARL.  I
realise this is slightly away from EARL's current use, but I think it
would be a solution to using it on sites.

Then of course there's non HTML resources, how do we link EARL to them,
here I think a robots.txt idea whereby EARL specifies URL fragments and
types say "PDF's in /documents/"  and has the report for these.  Unless we
go down the compact privacy solution and there's an HTTP option for
specifying it here.

Okay that's about it on my thoughts at the moment...

Jim.

Received on Friday, 19 October 2001 20:01:38 UTC