On Mon, 19 Jan 2004, Michael Paul Johnson wrote: > The xml:lang attribute is only supported for English texts, right now, > since most of the languages I'm interested in have no two-letter > codes. (There are too many of them. You can't represent over 6,000 > languages with two letters.)
I'm not sure where you got the idea that xml:lang only holds an ISO 639-1 (or some other two-letter) value, but the actual standard can easily refer to any human language. xml:lang's contents are defined by RFC 1766/3066. By the former, ISO 639-1, IANA, and x- codes are valid. The latter adds ISO 639-2 codes. I think OSIS preference (though this isn't in writing yet) is to use ISO 639-1, ISO 639-2, IANA, then x- codes, according to what is available for a given language, in that order of preference. SIL/Ethnologue codes should be of the form x-SIL-<code>. LINGUIST List codes should be of the form x-LINGUIST-<code>. That covers all human languages, delegating code assignment to the proper organizations (SIL & LINGUIST), as well as permitting locale-specific dialects and other subtags, according to RFC 1766/3066 conventions. And if additional codes are needed, x- codes without the SIL or LINGUIST subtag can be used. > Footnote start anchors use generic milestone markers. This may change > if OSIS starts really supporting these. This really isn't a violation > of the OSIS 2.0 spec, but if you use this feature beware that it may > change. Has this been suggested to OSIS yet? > Hebrew Psalm book titles are rendered as text (poetry <l> or prose <p> > elements) rather than <title> elements, because <title> elements > couldn't handle the appropriate "italics" markup for KJV "added" words > in the current OSIS 2.0.1 schema. I'll point that out at the next meeting. --Chris _______________________________________________ sword-devel mailing list [EMAIL PROTECTED] http://www.crosswire.org/mailman/listinfo/sword-devel
