Tags for the identification of languages (fwd)

Albert Lunde (Albert-Lunde@nwu.edu)
Sun, 26 Feb 95 15:13:21 EST

This may be of interest in the context of the issues of multilingual documents:

>The IESG has approved the Internet-Draft "Tags for the identification
>of languages" <draft-ietf-mailext-lang-tag-02.txt> as a Proposed
>Standard. This document is the product of the Mail Extensions Working
>Group. The IESG contact persons are John Klensin and Erik Huizer.
>
>
>Technical Summary
>
> This specification describes a language tag, and MIME syntax and
> semantics for applying it. The language tags is for use in cases
> where it is desired to indicate the language used in an information
> object, e.g., when presentations in multiple languages are available
> in a single MIME document. The specification also includes
> suggestions about the relationship of the language tag mechanism to
> other standards and work in progress, e.g., RFC1327 on conversions
> between X.400 and RFC 822-format mail and its successors.
>
>Working Group Summary
>
> The header option and structure result from a suggestion made during
> the original development of MIME. Earlier versions were favorably
> reviewed on the 822 Extensions mailing list. The proposal was then
> reviewed by the Mail Extensions WG, and consensus was reached on its
> soundness after minor additional revisions.
>
> Another component of this proposal adds a new optional parameter,
> "differences", to the definition of multipart/alternative in MIME.
> This parameter provides additional information that a receiving user
> agent might use to determine which alternative section to present.
>
> While there has been clear consensus that the technique and definition
> are sound, there is equally clear consensus that the method outlined
> in this specification does not solve all possible problems that one
> might associate with the identification of the language used in a MIME
> body part. In particular, a textual body part that contains an
> arbitrary mixture of languages might require labelling within the text
> itself as the language changed. This more complex case should not
> constitute an impediment to the use of the specification as described
> in this document.

---
    Albert Lunde                      Albert-Lunde@nwu.edu