Content-Script- and Content-Style-Type

Dear HTML Working Group,

  <http://www.w3.org/TR/html401/present/styles.html#default-style>:

[...]
  Documents that include elements that set the style attribute but which
  don't define a default style sheet language are incorrect.
[...]

and <http://www.w3.org/TR/html401/interact/scripts.html#h-18.2.2.1>:

[...]
  Documents that do not specify default scripting language information
  and that contain elements that specify an intrinsic event script are
  incorrect.
[...]

I guess it does not suprise you that most authors are not aware of this
rule, among all the documents directly linked from <http://www.w3.org/>
only two

  * http://www.keio.ac.jp/
  * http://www.w3.org/TR/2004/CR-css3-ui-20040511/

specify a default scripting language and/or style sheet language. Of
course, not all of them are required to, but the following are and do
not specify this mandatory information:

  * http://www.w3.org/2000/xp/Group/
  * http://www.w3.org/2001/sw/WebOnt/
  * http://www.w3.org/2002/mmi/
  * http://www.w3.org/2002/mmi/ink
  * http://www.w3.org/2004/02/mmi-workshop-cfp.html
  * http://www.w3.org/2004/03/w3c-track04.html
  * http://www.w3.org/Consortium/Translation/
  * http://www.w3.org/DOM/
  * http://www.w3.org/Graphics/PNG/
  * http://www.w3.org/Graphics/WebCGM/
  * http://www.w3.org/International/
  * http://www.w3.org/International/ws/
  * http://www.w3.org/Jigsaw/
  * http://www.w3.org/Mail/
  * http://www.w3.org/Mobile/CCPP/
  * http://www.w3.org/P3P/
  * http://www.w3.org/Search/Mail/Public/
  * http://www.w3.org/TR/2004/WD-i18n-html-tech-bidi-20040509/
  * http://www.w3.org/TR/2004/WD-i18n-html-tech-char-20040509/
  * http://www.w3.org/TR/2004/WD-i18n-html-tech-lang-20040509/
  * http://www.w3.org/Voice/
  * http://www.w3.org/XML/Schema
  * http://jigsaw.w3.org/css-validator/
  * http://lists.w3.org/Archives/Public/www-validator/
  * http://www2004.org/
  * http://www.ercim.org/

That's about 25% of the documents linked from the W3C homepage or 93% of
the documents linked from there that use either the style attribute or
one or more intrinsic event attributes. That's already a lot, but it
gets worse once you leave W3C web space...

Most of these documents pass the Validator <http://validator.w3.org/> as
these rules (and many more) are currently not enforced, same goes for
most if not all similar tools.

<http://www.w3.org/TR/html401/present/styles.html#default-style> states

[...]
  Authors must specify the style sheet language of style information
  associated with an HTML document.
[...]

where the "must" is to be interpreted as definded in RFC 2119 according
to <http://www.w3.org/TR/html401/conform.html>. This is a violation of
RFC 2119 which requires

[...]
  Imperatives of the type defined in this memo must be used with care
  and sparingly.  In particular, they MUST only be used where it is
  actually required for interoperation or to limit behavior which has
  potential for causing harm (e.g., limiting retransmisssions)  For
  example, they must not be used to try to impose a particular method
  on implementors where the method is not required for
  interoperability.
[...]

The specification of a default style sheet language is not mandatory for
interoperation as user agents must determine the default style sheet
language to be "text/css" if no language is specified and lack of such a
specification does not have the potential of causing harm.

For intrinsic event attributes it could indeed be considered necessary
for interoperation that a default scripting language is specified, but
it seems obvious from common practise that it is not.

Considering that most authors, even those who care a lot about web
standards, are unaware of these rules and the number of documents that
are rendered incorrect due to these rules, I feel strongly that an
enforcement of these rules in the various conformance tools I help to
maintain and develop would be damaging.

It would frustrate too many authors for no good reason. And it would
frustrate too many people involved with these tools or with web
standards in general. If we change the W3C MarkUp Validator, I see
endless complains on the mailing list, in blogs, articles about
"changing the rules" and evangelists would have to argue about
"ever-changing" standards... It does not make your web site any better
if you specify these defaults.

Thus, I ask you to propose a correction to the HTML 4.01 Recommendation
(and, if necessary, sub-sequent Recommendations) that removes these
requirements.

regards.

Received on Friday, 28 May 2004 01:49:33 UTC