- From: Bjoern Hoehrmann <derhoermi@gmx.net>
- Date: Mon, 12 Jul 2004 06:14:06 +0200
- To: www-html-editor@w3.org
Dear HTML Working Group, Appendix C.13 of the XHTML 1.0 Second Edition Recommendation states: [...] 3. Within the XHTML namespace, user agents are expected to recognize the "id" attribute as an attribute of type ID. Therefore, style sheets should be able to continue using the shorthand "#" selector syntax even if the user agent does not read the DTD. 4. Within the XHTML namespace, user agents are expected to recognize the "class" attribute. Therefore, style sheets should be able to continue using the shorthand "." selector syntax. [...] It is not clear to me what you mean by "Within the XHTML namespace". Do you mean on elements in the XHTML namespace as defined in section 3.1.1 or do you mean attributes in the XHTML namesapce as defined in section 3.1.1 on arbitrary elements? It is also not clear to me why user agents are expected to do as the section suggests, section 3.2 does not cite such a requirement. Do you mean that e.g. if an XHTML 1.0 user agent encounters an XHTML 1.1 document that uses ruby elements, it must not only process the content of the unrecognized attributes, but also process id, class, xml:lang attributes? That would seem inconsistent with section 3.2 which clearly states [...] When a user agent processes an XHTML document as generic XML, it shall only recognize attributes of type ID (i.e. the id attribute on most XHTML elements) as fragment identifiers. [...] as it does not know whether such attributes are ID attributes unless it processes the document type definition which it is not required to do, as far as I understand. But then, this part of section 3.2 does not make much sense to me. I cannot think of a situation where an XHTML user agent would process an XHTML document as "generic XML" and still be considered an XHTML user agent. It seems that XHTML user agents would always process XHTML documents as XHTML documents or they are not XHTML user agents. But then, maybe my confusion is caused by the lack of definition for "generic XML". In either case, if there is such an expectation, you need to clearly state this somewhere, not in the informative Appendix C. That said, it seems that the entire Appendix C.13 seems misplaced in Appendix C as it does not seem to contain guidelines for XHTML authors who wish to deliver their documents to legacy user agents. Could you please clarify what requirements they have to meet in order to do so? If there is no requirement, the appropriate section for the section seems to be section 4, "Differences with HTML 4". regards.
Received on Monday, 12 July 2004 00:14:50 UTC