W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > October to December 2003

RE: portals (CMS Client Area Issues)

From: Geoff Deering <gdeering@acslink.net.au>
Date: Fri, 10 Oct 2003 09:01:48 +1000
To: "Charles McCathieNevile" <charles@sidar.org>
Cc: "FOX, Jake" <FOXJ@NORWICH-UNION.CO.UK>, <w3c-wai-ig@w3.org>
Message-ID: <NBBBJPNFCLNLAADCLFJBKEJMFAAA.gdeering@acslink.net.au>

As Tom has said, Plone are addressing these issues.

Drupal (http://drupal.org/) also have plugins/modules to address this.

Movable Type (http://movabletype.org/), which is meant to come out in a Pro
version soon, will address workflow, etc, has a series of plugins

Possible Solutions??

Write a WAI compliant version of HTMLarea
(http://www.interactivetools.com/products/htmlarea/).  Not too difficult (I
don't think).  Only solves some problems.

Many CMSs / Bloggs now accept content via email.  Another possible solution
better than textarea editing.

Many CMSs now work with the Blogger API (WebDave/SOAP I think).  But the
Blogger clients I have seen are not W3C/WAI compliant.  So there is a need
to write one according to the ATAG.  This would be a good solution if they
interfaced with the actual contents style.

Even though I have been working with building and using CMSs in large scale
projects since 1996, and I don't consider myself an expert on these issues,
as I can be proved wrong about issues I thought there was no solution too,
but for my mind I see no real way until web publishing somehow meets the
traditional publishing model.  What I mean by that is that authors were only
responsible for content, editors and designers laid out the document format.
In a purists world, I am against content authors having any access to
anything other than markup tools that have no capacity to impact design.

I really think that the XML publishing format comes closest to separating
content from the document structure and design.  I think this is the only
way to allow users to put in content and the system put a valid markup
wrapper around it with correct semantic information, without the content
author having to concern themselves with any of these issues.  Systems like
Apache/Cocoon/Lenya and BitFlux (and some other XML publishing systems) are
focused in this direction.  What I also like about the Apache/Cocoon
approach is that it is built around TCN (transparent content negotiation),
something that is sadly underused, this gives the publisher more control
over document delivery.

The other thing I find very strange in this area is the lack of use of
Dublin Core metadata as a repository in any of these systems to manage the
versioning of content and also for gathering semantic information to
populate documents.  This is something sadly missing in most / all CMS
(probably I am missing some that actually do it).

You need all the above information if you are going to have enough data
repository to generate the likes of tables with full semantic information on
the fly out of XML/XSLT/etc or any other method where the content author
only has knowledge and responsibility of the content (no knowledge of markup

I'd offer to supply some feedback to the ATAG but it would only add to the
things on my wish list and things I have promised or intended to do, but
never seem to have the time to do.  Maybe later.


-----Original Message-----
From: Charles McCathieNevile

Hi Geoff,

Ah yes, this is going beyond "content" to create an "application", and
an authoring application at that.

And now that you mention it, there are the "Authoring Tool
Accessibility Guidelines", a W3C recommendation published in early
2000, with an active working group - see http://www.w3.org/WAI/AU - and
they talk about many of the issues you raise, as well as some others.
It is true that doing this right is a bit more complex than just making
a web page.

If you (or anyone) happen to have time to review those guidelines,
evaluate a tool against them or provide some techniques, I presume the
working group would be interested in the feedback (they're a small
group taking on a large and complex task), and I would be myself.


Received on Thursday, 9 October 2003 19:08:44 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:25 UTC