Re: WCAG 2.0 -> 3 - Bottom Layer (Server Side Techniques)

At 01:26 AM 2003-08-26, Geoff Deering wrote:

>I feel there is a section here that needs addressing, and yes, if others
>feel it is relevant and warranted, I'll draft the material on it.  This is
>addressing "server side techniques".

This is an important topic and it should go forward.

It is an open question how much of this should be published as an
"accessibility" document or as a "quality" document, a companion to
the CUAP and CHIP documents.  Failure of server operators to employ
the functionality in the specifications or to give content authors
control over the necessary [HTTP header] metadata have repeatedly
frustrated the intentions of Web architecture.

I don't think we are going to see an HTML 4.02 out of W3C with the DTD
changes that you suggest.  But I would definitely go to bat for a policy
statement that it is an OK and sometimes constructive thing to do to scrub
your content (on posting to the server) with a stricter DTD than the
specification requires.  Interoperability failures happen when the server
assumes user agent respect for rules that the UA does not understand or
process.  Examples here are altered content models for TABLE as you suggest
and html:a.name (make the content be NMTOKEN rather than CDATA).

<http://www.w3.org/2002/02/mid/5.1.0.14.2.20021129085107.01e3f080@pop.iamdigex.net;list=www-qa>

Al

>http://www.w3.org/TR/WCAG20/#how-to
>
>see
>
><WCAG2 version="W3C Working Draft 24 June 2003">
>3 - Bottom layer - Technology-specific application information
>The Techniques Documents will include code examples, screen shots, and other
>information specific to a technology.  These documents will be
>non-normative. They will contain different strategies for meeting the
>requirements as well as the current preferred approaches where they exist.
>Examples include:
>
>* Hypertext Markup Language (HTML) and Extensible Hypertext Markup Language
>(XHTML) Techniques
>* Cascading Style Sheets (CSS) Techniques
>* Server-side scripting Techniques
>* Client-side scripting Techniques
>* Scalable Vector Graphics (SVG) Techniques
>* Synchronized Multimedia Integration Language (SMIL) Techniques
>* Extensible Markup Language (XML) Techniques
>
>(These will become active links as the corresponding working drafts are
>published)
></WCAG2>
>
>I am not too sure what defines "Server-side scripting Techniques", but as a
>web developer I understand that as anything on the server side that has to
>do with programming or markup.  What that does not refer to is server
>management and configuration techniques that address WCAG (and other
>issues).  I feel there is a real need to add "Server Management Techniques"
>into this list.
>
>Why? Web sites can greatly improve accessibility and usability addressing
>server configuration and management issues.
>
>For instance; server side redirects/redirection.  See
>
>http://httpd.apache.org/docs/mod/mod_alias.html#redirect
>
>If the redirection is marked as permanent, and if the user requested the URL
>for redirection from a bookmark, the user agent is supposed to update the
>bookmark (I haven't looked at UAAG thoroughly, but this was always meant to
>be the case in handling HTTP).
>
>There are heaps of other server side techniques, like using mod_speling,
>mod_rewrite, etc.
>
>There is also the hugely underused and undervalued "Transparent content
>negotiation" (TCN), see
>
>http://httpd.apache.org/docs/content-negotiation.html
>
>It's true that Netscape 2.0 destroyed this feature by getting that UA to
>market as quickly as possible, and sending back any requests for browser
>configs with "*.*" (ie - everything including the kitchen sink is installed
>on this UA), but it is reasonably well supported these days.  Well enough to
>build Web Services.
>
>TCN becomes very important and far more sophisticated in XML content
>delivery systems such as Apache/Cocoon (http://cocoon.apache.org/).  This is
>something I am spending more time with and would be happy to try and deliver
>some Techniques, etc for, later down the track.
>
>Is there any value or point to addressing these as part of WCAG2 Techniques?
>
>Geoff Deering

Received on Tuesday, 26 August 2003 10:45:14 UTC