Internet-Draft H. Alvestrand draft-alvestrand-lang-tag-v2-02.txt EDB Maxware Target Category: Standards Track June 2000 Obsoletes: RFC 1766 Expires: December 2000 Tags for the Identification of Languages Status of this Memo The file name of this memo is draft-alvestrand-lang-tag-v2-0.txt This document is an Internet-Draft and is in full conformance with all provisions of Section 10 of RFC 2026. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet- Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet- Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html. Comments on this draft should be sent to the mailing list Abstract This document describes a language tag for use in cases where it is desired to indicate the language used in an information object. It also defines a "Content-language:" header, for use in the case where one desires to indicate the language of something that has RFC-822-like headers, like MIME body parts or Web documents, and a new parameter to the Multipart/Alternative type, to aid in the usage of the Content- Language: header. Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 1. Introduction There are a number of languages presently or previously used by human beings in this world. A great number of these people would prefer to have information presented in a language which they understand. In some contexts, it is possible to have information available in more than one language, or it might be possible to provide tools (such as dictionaries) to assist in the understanding of a language. In other cases, it may be desirable to use a computer program to convert information from one format (such as plaintext) into another (such as computer-synthesized speech, or Braille, or high-quality print renderings). A prerequisite for any such function is a means of labelling the information content with an identifier for the language that is used in this information content. This document specifies an identifier mechanism, and one possible use for it. The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC 2119]. 2. The Language tag 2.1 Language tag syntax The language tag is composed of one or more parts: A primary language tag and a (possibly empty) series of subtags. The syntax of this tag in RFC 2234 ABNF is: Language-Tag = Primary-tag *( "-" Subtag ) Primary-tag = 1*8ALPHA Subtag = 1*8ALPHA All tags are to be treated as case insensitive; there exist conventions for capitalization of some of them, but these should not be taken to carry meaning. For instance, ISO 3166 recommends that country codes are capitalized (MN Mongolia), while ISO 639 recommends that language codes are written in lower case (mn Mongolian). draft-alvestrand-lang-tags-v2-01.txt [Page 2] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 2.2 Language tag sources The namespace of language tags is administered by the IANA according to the rules in section 5 of this document. The following registrations are predefined: In the primary language tag: - All 2-letter tags are interpreted according to assignments found in ISO standard 639, "Code for the representation of names of languages" [ISO 639], or subsequently made by the standardÆs registration authority. (Note: A revision is underway, and is expected to be released as ISO 639-1:2000) - All 3-letter tags are interpreted according to assignments found in ISO 639 part 2, "Codes for the representation of names of languages - - Part 2: Alpha-3 code [ISO 639-2] , or subsequently made by the standardÆs registration authority. - The value "i" is reserved for IANA-defined registrations - The value "x" is reserved for private use. Subtags of "x" shall not be registered by the IANA. - Other values shall not be assigned except by revision of this standard. The reason for reserving all other tags is to be open towards new revisions of ISO 639; the use of "i" and "x" is the minimum we can do here to be able to extend the mechanism to meet our immediate requirements. In the first subtag: - All 2-letter codes are interpreted as ISO 3166 alpha-2 country codes denoting the area in which the language is used. - Codes of 3 to 8 letters may be registered with the IANA, according to the rules in chapter 5 of this document. The information in the subtag may for instance be: - Country identification, such as en-US (this usage is described in ISO 639) - Dialect or variant information, such as no-nyn (nynorsk) or en-scouse - Languages not listed in ISO 639 that are not variants of any listed language, which can be registered with the i-prefix, such as i- cherokee - Script variations, such as az-Arab and az-Cyrl (Azerbaijani in Arabic or Cyrillic script - these script codes are suggested by the pending script code standard ISO/DIS 15924) draft-alvestrand-lang-tags-v2-01.txt [Page 3] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 This document does not place any restriction on what values one can register here, as long as they conform to the rules in section 5. NOTE IN DRAFT: It has been suggested that subtags of 4 characters be reserved for ISO/DIS 15924 codes. Opinions for and against are sought. ISO 639 defines a registration authority for additions to and changes in the list of languages in ISO 639. This authority is: International Information Centre for Terminology (Infoterm) P.O. Box 130 A-1021 Wien Austria Phone: +43 1 26 75 35 Ext. 312 Fax: +43 1 216 32 72 ISO 639-2 defines a registration authority for additions to and changes in the list of languages in ISO 639-2. This authority is: Library of Congress Network Development and MARC Standards Office Washington, D.C. 20540 USA Phone: +1 202 707 6237 Fax: +1 202 707 0115 URL: http://www.loc.gov/standards/iso639 The registration agency for ISO 3166 (country codes) is: ISO 3166 Maintenance Agency Secretariat c/o DIN Deutsches Institut fuer Normung Burggrafenstrasse 6 Postfach 1107 D-10787 Berlin Germany Phone: +49 30 26 01 320 Fax: +49 30 26 01 231 ISO 3166 reserves the country codes AA, QM-QZ, XA-XZ and ZZ as user- assigned codes. 2.3 Choice of language tag One may occasionally be faced with several possible tags for the same body of text. Interoperability is best served if all users send the same tag, and use the same tag for the same language for all documents. Exact requirements may need to vary by application area; if so, the application protocol specification MUST specify how the procedure varies from the one given here. draft-alvestrand-lang-tags-v2-01.txt [Page 4] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 The text below is based on the set of tags known to the tagging entity. 1. Use the most precise tagging known to the sender that can be ascertained. 2. When a language has both an ISO 639-1 2-character tag and an ISO 639- 2 3-character tag, you MUST use the ISO 639-1 2-character tag. 3. When a language has no ISO 639-1 2-character tag, and the ISO 639-2/T (Terminology) tag and the ISO 639-2/B (Bibliographic) tag differ, you MUST use the Terminology tag. NOTE: At present, all languages for which there is a difference have 2-character tags, and the displeasure of developers about the existence of 2 tag sets has been adequately communicated to ISO. So this situation will hopefully not arise) 4. When a language has both an IANA-registered tag (i-something) and an ISO registered tag, you MUST use the ISO tag. NOTE: When such a situation is discovered, the IANA-registered tag SHOULD be deprecated as soon as possible. 5. You SHOULD NOT use the UND (Undetermined) tag unless the protocol in use forces you to give a value for the language tag, even if the language is unknown. Omitting the tag is preferred. 6. You MUST NOT use the MUL (Multiple) tag if the protocol allows you to use multiple languages, as is the case for the Content-Language: header. NOTE: In order to avoid versioning difficulties in applications such as that of RFC 1766, the ISO 639 RA-JAC has agreed on the following policy statement: "After the publication of ISO/DIS 639-1 as an International Standard, no new 2-letter code shall be added to ISO 639-1 unless a 3-letter code is also added at the same time to ISO 639-2. In addition, no language with a 3-letter code available at the time of publication of ISO 639-1 which at that time had no 2-letter code shall be subsequently given a 2-letter code." This will ensure that, for example, a user who implements "hwi" (HawaiÆian), which currently has no 2-letter code, will not find his or her data invalidated by eventual addition of a 2-letter code for that language. 2.4 Meaning of the language tag The language tag always defines a language as spoken (or written, signed or otherwise signalled) by human beings for communication of information to other human beings. Computer languages such as programming languages are explicitly excluded. draft-alvestrand-lang-tags-v2-01.txt [Page 5] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 There is no guaranteed relationship between languages whose tags begin with the same series of subtags; specifically, they are NOT guaranteed to be mutually intelligible, although it will sometimes be the case that they are. Applications should always treat a language tag as a single token; the division into main tag and subtags is an administrative mechanism, not a navigation aid. The relationship between the tag and the information it relates to is defined by the standard describing the context in which it appears. Accordingly, this section can only give possible examples of its usage. - For a single information object, it should be taken as the set of languages that is required for a complete comprehension of the complete object. Example: Plain text documents. - For an aggregation of information objects, it should be taken as the set of languages used inside components of that aggregation. Examples: Document stores and libraries. - For information objects whose purpose is to provide alternatives, it should be regarded as a hint that the material inside is provided in several languages, and that one has to inspect each of the alternatives in order to find its language or languages. In this case, multiple languages need not mean that one needs to be multilingual to get complete understanding of the document. Example: MIME multipart/alternative. - In markup languages, such as HTML, it is possible to define a construct embedding a language tag to indicate that contained text is written in this language, such that one could write
C'est la vie
inside a Norwegian document; the Norwegian-speaking user could then access a French-Norwegian dictionary to find out what the marked section meant. If the user were listening to that document through a speech synthesis interface, this formation could be used to signal the synthesizer to appropriately apply French text-to-speech pronunciation rules to that span of text, instead of misapplying the Norwegian rules. 2.5 Language-range Since the publication of RFC 1766, it has become apparent that there is a need to define a term for a set of languages that share some common property. The following definition of language-range is derived from RFC 2616 (HTTP/1.1). language-range = ( ( 1*8ALPHA *( "-" 1*8ALPHA ) ) / "*" ) draft-alvestrand-lang-tags-v2-01.txt [Page 6] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 A language-range matches a language-tag if it exactly equals the tag, or if it exactly equals a prefix of the tag such that the first tag character following the prefix is "-". The special range "*" matches any tag. A protocol which uses language ranges may specify additional rules about the semantics of "*"; for instance, HTTP/1.1 specifies that it only matches languages not matched by any other range within an "Accept-Language:" header. NOTE: This use of a prefix matching rule does not imply that language tags are assigned to languages in such a way that it is always true that if a user understands a language with a certain tag, then this user will also understand all languages with tags for which this tag is a prefix. The prefix rule simply allows the use of prefix tags if this is the case. 3. The Content-language header The "Content-Language" header is intended for use in the case where one desires to indicate the language(s) of something that has RFC-822-like headers, such as MIME body parts or Web documents. The RFC-822 EBNF of the Content-Language header is: Content-Language = "Content-Language" ":" 1#Language-tag Or in RFC 2234 ABNF: Content-Language = "Content-Language" CFWS ":" Language-List Language-List = Language-Tag [ CFWS "," CFWS Language-List ] The Content-Language header may list several languages in a comma- separated list. The CFWS construct is intended to function like the whitespace convention in RFC 822, which means also that one can place parenthesized comments anywhere in the language sequence, or use continuation lines. A formal definition is given in the update to the RFC 822 grammar, currently a work in progress. 3.1 Examples of Content-language values Norwegian official document, with parallel text in both official versions of Norwegian. (Both versions are readable by all Norwegians). Content-Type: multipart/alternative; differences=content-language Content-Language: no-nyn, no-bok Voice recording from Liverpool downtown draft-alvestrand-lang-tags-v2-01.txt [Page 7] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 Content-type: audio/basic Content-Language: en-scouse Document in Mingo, an American Indian language which does not have an ISO 639 code: Content-type: text/plain Content-Language: i-mingo An English-French dictionary Content-type: application/dictionary Content-Language: en, fr (This is a dictionary) An official European Commission document (in a few of its official languages) Content-type: multipart/alternative Content-Language: da, de, el, en, fr, it An excerpt from Star Trek Content-type: video/mpeg Content-Language: i-klingon (All the tags used in these examples were registered with IANA after the publication of RFC 1766) 4. IANA registration procedure for language tags Any language tag shall begin with an existing tag, and extend it. The registration form given here must be used by anyone who wants to use a language tag not defined by ISO or IANA. ---------------------------------------------------------------------- LANGUAGE TAG REGISTRATION FORM Name of requester : E-mail address of requester: Tag to be registered : English name of language : Native name of language (transcribed into ASCII): Reference to published description of the language (book or article): Any other relevant information: ---------------------------------------------------------------------- draft-alvestrand-lang-tags-v2-01.txt [Page 8] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 The language form must be sent to for a 2- week review period before it can be submitted to IANA. (This is an open list. Requests to be added should be sent to .) When the two week period has passed, the language tag reviewer, who is appointed by the IETF Applications Area Director, either forwards the request to IANA@ISI.EDU, or rejects it because of significant objections raised on the list. Note that the reviewer can raise objections on the list himself, if he so desires. The important thing is that the objection must be made publicly. The applicant is free to modify a rejected application with additional information and submit it again; this restarts the 2-week comment period. Decisions made by the reviewer may be appealed to the IESG. All registered forms are available online in the directory ftp://ftp.isi.edu/in-notes/iana/assignments/languages/ Updates of registrations follow the same procedure as registrations. The language tag reviewer decides whether to allow a new registrant to update a registration made by someone else; in the normal case, objections by the original registrant would carry extra weight in such a decision. There is no deletion of registrations; when some registered tag should not be used any more, for instance because a corresponding ISO 639 code has been registered, the registration should be amended by adding a remark like "DO NOT USE: use instead" to the "other relevant information" section. 5. Security Considerations The only security issue that has been raised with language tags since the publication of RFC 1766, which stated that "Security issues are believed to be irrelevant to this memo", is a concern with language ranges used in content negotiation - that they may be used to infer the nationality of the sender, and thus identify potential targets for surveilllance. This is a special case of the general problem that anything you send is visible to the receiving party; it is useful to be aware that such concerns can exist in some cases. The exact magnitude of the threat, and any possible countermeasures, is left to each application protocol. 6. Character set considerations Codes may always be expressed using the US-ASCII character repertoire (a-z), which is present in most character sets. The issue of deciding upon the rendering of a character set based on the language tag is not addressed in this memo; however, it is thought draft-alvestrand-lang-tags-v2-01.txt [Page 9] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 impossible to make such a decision correctly for all cases unless means of switching language in the middle of a text are defined (for example, a rendering engine that decides font based on Japanese or Chinese language may fail to work when a mixed Japanese-Chinese text is encountered) 7. Acknowledgements This document has benefited from many rounds of review and comments in various fora of the IETF and the Internet working groups. Any list of contributors is bound to be incomplete; please regard the following as only a selection from the group of people who have contributed to make this document what it is today. In alphabetical order: Tim Berners-Lee, Nathaniel Borenstein, Sean M. Burke, Jim Conklin, John Cowan, Dave Crocker, Martin Duerst, Michael Everson, Ned Freed, Tim Goodwin, Dirk-Willem van Gulik, Paul Hoffman, Olle Jarnefors, John Klensin, Keith Moore, Masataka Ohta, Keld Jorn Simonsen, Rhys Weatherley, Misha Wolf, Francois Yergeau and many, many others. Special thanks must go to Michael Everson, who has served as language tag reviewer for almost the complete period since the publication of RFC 1766, and has provided a great deal of input to this revision. 8. Author's Address Harald Tveit Alvestrand EDB Maxware Pirsenteret N-7462 TRONDHEIM NORWAY EMail: Harald.Alvestrand@maxware.no Phone: +47 73 54 57 97 9. References [ISO 639] ISO 639:1988 (E/F) - Code for the representation of names of languages - The International Organization for Standardization, 1st edition, 1988-04-01 Prepared by ISO/TC 37 - Terminology (principles and coordination). Note that a new version (ISO 639-1:2000) is in preparation at the time of this writing. [ISO 639-2] draft-alvestrand-lang-tags-v2-01.txt [Page 10] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 ISO 639-2:1998 - Codes for the representation of names of languages -- Part 2: Alpha-3 code - edition 1, 1998-11-01, 66 pages, prepared by ISO/TC 37/SC 2 [ISO 3166] ISO 3166:1988 (E/F) - Codes for the representation of names of countries - The International Organization for Standardization, 3rd edition, 1988-08-15. [ISO 15924] ISO/DIS 15924 - Codes for the representation of names of scripts (under development by ISO TC46/SC2) [RFC 1327] Kille, S., "Mapping between X.400(1988) / ISO 10021 and RFC 822", RFC 1327, University College London, May 1992. [RFC 1521] Borenstein, N., and N. Freed, "MIME Part One: Mechanisms for Specifying and Describing the Format of Internet Message Bodies", RFC 1521, Bellcore, Innosoft, September 1993. [RFC 2119] Key words for use in RFCs to Indicate Requirement Levels. S. Bradner. March 1997. [RFC 2234] Augmented BNF for Syntax Specifications: ABNF. D. Crocker, Ed., P. Overell, November 1997. [RFC 2616] Hypertext Transfer Protocol -- HTTP/1.1. R. Fielding, J. Gettys, J. Mogul, H. Frystyk, L. Masinter, P. Leach, T. Berners-Lee. June 1999. Appendix A: Language Tag Reference Material The Library of Congress, maintainers of ISO 639-2, has made the list of languages registered available on the Internet. At the time of this writing, it can be found at http://www.loc.gov/standards/iso639-2/langhome.html The IANA registration forms for registered language codes can be found at http://www.isi.edu/in-notes/iana/assignments/languages/ draft-alvestrand-lang-tags-v2-01.txt [Page 11] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 Appendix B: Changes from RFC 1766 . Email list address changed from ietf-types@uninett.no to ietf- languages@iana.org . Updated author's address . Added language-range construct from HTTP/1.1 . Added use of ISO 639-2 language codes . Added reference to Library of Congress lists of language codes . Changed examples to use registered tags . Moved Multipart/Alternative-related stuff to appendix C . Added "Any other information" to registration form . Added description of procedure for updating registrations Appendix C: Use of Content-Language with Multipart/Alternative NOTE: This appendix details an idea that was proposed in RFC 1766 to deal with a particular kind of alternative content. However, this has not found use in practice, and is therefore not suitable for the IETF standards track. It is being preserved here as a non-normative appendix only. When using the Multipart/Alternative body part of MIME, it is possible to have the body parts giving the same information content in different languages. In this case, one should put a Content-Language header on each of the body parts, and a summary Content-Language header onto the Multipart/Alternative itself. The differences parameter to multipart/alternative As defined in RFC 1541, "Multipart/Alternative" only has one parameter: boundary. The common usage of "Multipart/Alternative" is to have more than one format of the same message (f.ex. PostScript and ASCII). The use of language tags to differentiate between different alternatives will certainly not lead all MIME UAs to present the most meaningful, understandable or significant body part as default. Therefore, a new parameter is defined, to allow the configuration of MIME readers to handle language differences in a sensible manner. Name: Differences Value: One or more of Content-Type Content-Language draft-alvestrand-lang-tags-v2-01.txt [Page 12] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 Further values can be registered with IANA; these shall refer to the name of a header for which a definition exists in a published RFC. If not present, "Differences=Content-Type" is assumed. The intent is that the MIME reader can look at these headers of the message component to make an intelligent choice of what to present to the user, based on knowledge about the user preferences and capabilities. (The intent of having registration with IANA of the fields used in this context is to maintain a list of usages that a mail UA may expect to encounter, not to reject usages.) (NOTE: The MIME specification [RFC 1521], section 7.2, states that headers not beginning with "Content-" are generally to be ignored in body parts. People defining a header for use with "differences=" should take note of this.) The mechanism for deciding which body part to present is outside the scope of this document. MIME EXAMPLE: Content-Type: multipart/alternative; differences=Content-Language; boundary="limit" Content-Language: en, fr, de --limit Content-Language: fr Le renard brun et agile saute par dessus le chien paresseux --limit Content-Language: de Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-encoding: quoted-printable Der schnelle braune Fuchs h=FCpft =FCber den faulen Hund --limit Content-Language: en The quick brown fox jumps over the lazy dog --limit-- When composing a message, the choice of sequence may be arbitrary. However, non-MIME mail readers will show the first body part first, meaning that this should most likely be the language understood by most of the recipients. Appendix X1: Changes from draft -00 to -01 This appendix is to be deleted by the RFC Editor before publication as RFC. Changes from draft-00: draft-alvestrand-lang-tags-v2-01.txt [Page 13] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 - Fixed up the language tag table - Moved multipart/alternative stuff to appendix - Changed examples to use registered tags - Added * in languagte tag table to indicate B/T conflicts - Considered, but did not adopt, changing from recommending T codes to recommending B codes. At the moment, the only argument that appeals to the author is that the T codes look more like the 639-1 codes than the B codes do. - Added procedures for updating a registration Here is the list of changes that need to be done to this doc before advancing it to Draft or reissuing it. - Decide whether or not to write anything about use of country codes in other places than the first subtag, or region codes, or script codes - Decide whether it is worth it to try to write down any more guidelines for what language tags people should register Appendix X2: Changes from draft -01 to -02 This appendix is to be deleted by the RFC Editor before publication as RFC. - Minor updates - Added reference to Library of Congress code lists instead of including code values - Changed grammars to use RFC 2234 ABNF - Used MUST and SHOULD in label choice algorithm TODO list Consider whether to include Accept-Language: as a generic. If so, decide whether to use the HTTP standard's "weight factor" or the nonstandard, but commonly used "best language first". Suggested language, to be placed as section 3.2: The "Accept-Language" header is intended for use in the case where a user or a process desires to indentify the language(s) he prefers when RFC-822-like headers, such as MIME body parts or Web documents are used. The RFC-822 EBNF of the Accept-Language header is: Accept-Language = "Accept-Language" ":" Language-List draft-alvestrand-lang-tags-v2-01.txt [Page 14] Tags for the names of languages Harald Alvestrand draft-alvestrand-lang-tag-v2-02.txt Expires December 2000 The Accept-Language header may list several languages in a comma-separated list. The leftmost is the most preferred. Note that HTTP uses the same header name with a different syntax; see RFC 2616 for the details. Consider whether more guidance on "appropriate" tags is needed. Consider whether we need to allow numbers in language tags. draft-alvestrand-lang-tags-v2-01.txt [Page 15]