- From: Bjoern Hoehrmann <derhoermi@gmx.net>
- Date: Sat, 20 Oct 2001 22:24:00 +0200
- To: www-style@w3.org, www-html@w3.org
Hi, HTTP/1.1 offers a Content-Language entity header to identify the language of the relevant entity. HTML4 uses this header for language identification (see section 8.1.2). XML uses higher-level protocol information only for character encoding determination, not for language, thus you can only use the xml:lang attribute in XML documents to identify language information. While I think it was a bad idea in general to add meta data features to core XML (even worse as attribute, leaving of an opportunity to define a language for attribute values), we are stuck with it. Proper language identification is vital for the :lang() pseudo-class in CSS. It is currently defined [1], that the document language has to define means for language identification of elements. Considering the case of XHTML documents not having any xml:lang attribute, would the root element inherit the language from higher-level protocol information and e.g. :lang(...) would select anything? Are semantics inherited from HTML4 for XHTML 1.0 and XHTML M12N based XHTML family document types? If so, would it be possible to note this somewhere? Or is this an error in XML 1.0 not considering higher-level protocol information in this case? Should the W3C Selectors module be fixed to consider higher-level protocol information? [1] http://www.w3.org/TR/selectors#lang-pseudo -- Björn Höhrmann { mailto:bjoern@hoehrmann.de } http://www.bjoernsworld.de am Badedeich 7 } Telefon: +49(0)4667/981028 { http://bjoern.hoehrmann.de 25899 Dagebüll { PGP Pub. KeyID: 0xA4357E78 } http://www.learn.to/quote/
Received on Saturday, 20 October 2001 16:25:04 UTC