- From: Internationalization Core Working Group Issue Tracker <sysbot+tracker@w3.org>
- Date: Mon, 26 Mar 2012 11:49:46 +0000
- To: public-i18n-core@w3.org
I18N-ISSUE-155: [Bug 16520] New: Don't indicate that XML MIME types *requires* xml:lang [Adhoc-HTML] http://www.w3.org/International/track/issues/155 Raised by: Richard Ishida On product: Adhoc-HTML Bugzilla: https://www.w3.org/Bugs/Public/show_bug.cgi?id=16520 Raised by: Leif Halvard Silli The HTML+RDFa spec says: ]] If an author is editing an HTML fragment and is unsure of the final encapsulating MIME type for their markup, it is suggested that the author specify both lang and xml:lang [[ NIT: "If an author is [snip] for their markup". Correct: "If an author is [snip] for his/her markup". ISSUE: The advice proliferate the belief that XML mime types *need* xml:lang. But it is only if they don't understand XHTML that they *need* xml:lang. They might very well not understand XHTML. But is that related to the MIME type? The only use case I have heard for xml:lang is XML authoring tools - thus, not exactly "the final encapsulating MIME type". XML parsers of the Web browser kind (IE/Webkit/Opera/Gecko) do understand the @lang attribute. (Though there might be legacy versions which don't.) And e.g. the XHTML+RDFa DOCTYPE supports both @lang and @xml:lang. Are there any *real* reasons for using both attributes - unrelated to authors' fears and feelings? Such as legacy RDFa parsers? Or specific XML authoring tools? Or specific consumers?
Received on Monday, 26 March 2012 11:49:53 UTC