W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 2008

Re: Serving the Most Appropriate Content to Multiple User Agents from a Single Document Source: XHTML Media Types

From: Christophe Strobbe <christophe.strobbe@esat.kuleuven.be>
Date: Wed, 05 Nov 2008 19:21:11 +0100
Message-Id: <>
To: w3c-wai-gl@w3.org
Cc: wai-xtech@w3.org


At 17:02 5/11/2008, Gregory J. Rosmaita wrote:
>as many of you are aware, XHTML™ 1.0, The Extensible HyperText Markup
>Language (A Reformulation of HTML 4 in XML 1.0 - Second Edition)
>    * [http://www.w3.org/TR/xhtml1]
>included an informative Appendix C: "HTML Compatibility Guidelines"
>    * [http://www.w3.org/TR/xhtml1/#guidelines]
>now, the XHTML2 Working Group has clarified how to use XHTML 1.0 to
>serve the most appropriate content to multiple user agents from a
>single document source, in a new document:
>"XHTML Media Types: Serving the Most Appropriate Content to Multiple
>User Agents from a Single Document Source"; current drafts of XHTML
>Media Types are available from
>   * [http://www.w3.org/MarkUp/Drafts/Overview.html#xhtmlmime]
>currently, the latest editor's draft is located at:
>   * [http://www.w3.org/MarkUp/2008/ED-xhtmlmime-20081024/]
>explanation: this document supersedes the informative Appendix C of
>the original XHTML 1.0 Recommendation, entitled: "HTML Compatibility
>Guidelines", (...) as a normative document.

As far as I can see, the implications for WCAG 
2.0 Techniques and Failures would be the following:

  * A.1: remove the XML processing instructions 
and the XML declaration from the code examples in Techniques and Failures;
  * (A.2: I think we always used minimized tag 
syntax for elements with an EMPTY content model);
  * (A.3: I think we never used minimized tag 
syntax for elements that don't have an EMPTY content model);
  * A.4: scripts that contain ampersands are not 
embedded, except in technique ARIA1 (which uses a CDATA section);
    there are no scripts that contain ']]>';
    some scripts contain a less-than character (SCR19, ARIA1, ARIA4, SCR22);
  * A.5: I haven't checked if code examples 
contain line breaks in attribute values (it's verly unlikely);
  * A.7: the use of lang or xml:lang in the 
examples depends on the DOCTYPE: only lang in 
HTML4, both lang and xml:lang in XHTML 1.0, only xml:lang in XHTML 1.1;
  * A.8: the only code snippet with name and 
without id for a fragment identifier is in techniques G58.
  * A.9: charset values in the code snippets are 
normally utf-8; iso-8859-1 is used in SVR3 (and 
in a param element in SM11 and SM12);
  * A.10: we always used the full form for 
checked="checked"; I haven't found other examples of boolean attributes;
  * A.11: not checked but should be OK;
  * A.12: there are ampersands that need escaping in H50 and H73
  * (A.13: no issues found;)
  * (A.14: no processing instructions found in the examples;)
  * (A.15: not checked; it's unlikely that U+000C 
is used in Techniques and Failures;)
  * (A.16: &apos; not found in Techniques and Failures;)
  * (A.17: no DTD internal subsets in Techniques and Failures;)
  * (A.18: CDATA sections are only used as described in A.4;)
  * A.19: there are no explicit tbody elements in 
H4, H43, H51, H63, H73 (both examples), H79, C29, 
F32, F46, F49 (both examples), F63;
  * (A.20: no examples with base or xml:base in Techniques and Failures;)
  * (A.21: document.write is not recommended by 
WCAG 2; see technique SCR21, which recommends DOM funcions instead;)
  * (A.22: same comment as above;)
  * A.23: see A.19;
  * (A.24: no CSS examples that style the HTML element;)
  * A.25: noscript is used in SCR19;
  * A.26: WCAG 2 does not prohibit iframe; 
technique H64 is about the title attribute on iframe elements;
  * A.27: example code in Techniques and Failures 
only uses createElement, not createElementNS.

Best regards,


Christophe Strobbe
K.U.Leuven - Dept. of Electrical Engineering - SCD
Research Group on Document Architectures
Kasteelpark Arenberg 10 bus 2442
B-3001 Leuven-Heverlee
tel: +32 16 32 85 51
Please don't invite me to LinkedIn, Facebook, 
Quechup or other "social networks". You may have 
agreed to their "privacy policy", but I haven't.

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
Received on Wednesday, 5 November 2008 18:22:31 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:34:04 UTC