Network Working Group N. Walsh Internet-Draft Sun Microsystems, Inc. Expires: August 14, 2001 J. Cowan Reuters Health Information P. Grosso Arbortext, Inc. February 13, 2001 A URN Namespace for Public Identifiers draft-walsh-urn-publicid-00 Status of this Memo This document is an Internet-Draft and is in full conformance with all provisions of Section 10 of RFC2026. 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. This Internet-Draft will expire on August 14, 2001. Copyright Notice Copyright (C) The Internet Society (2001). All Rights Reserved. Abstract This document describes a URN namespace that is designed to allow Public Identifiers to be expressed in URI syntax. 1. Introduction XML[1] external entities have two identifiers: a public identifier and a system identifier. The system identifier is a URI, by definition, but the public identifier is simply a string. Walsh, et. al. Expires August 14, 2001 [Page 1] Internet-Draft A URN Namespace for Public Ids February 2001 Historically, the system identifier of an external entity has been a local, or system-specific identifier while the public identifier has been a more global, persistent name. Unfortunately, public identifiers do not fit neatly into the existing web architecture because they are not legal URIs. Many new specifications (XSLT, XML Schema, etc.) have the implicit or explicit requirement that all external identifiers be URIs. Any string which consists only of the public identifier characters (defined by Production 13 of Extensible Markup Language (XML) 1.0 Second Edition[1]) is a legal public identifier. But SGML[3] defines a restricted subset of public identifier called a "Formal Public Identifier" (FPI). For the purpose of this document, the significant difference between public identifiers and FPIs is that FPIs have internal structure and may have registered owner identifiers. This document describes a scheme for representing public identifiers as URNs by introducing a public identifier namespace, "publicid". This namespace specification is for a formal namespace. 2. Specification Template Namespace ID: "publicid" requested. Registration Information: Registration Version Number: 1 Registration Date: 2001-02-13 Declared registrant of the namespace: Norman Walsh Sun Microsystems, Inc. One Network Drive MS UBURO2-201 Burlington, MA 01803-0902 Norman.Walsh@East.Sun.COM Declaration of structure: The purpose of this namespace is to allow public identifiers to be encoded in URNs in a reliable, comparable way. To that end, this document mandates that public identifiers be Walsh, et. al. Expires August 14, 2001 [Page 2] Internet-Draft A URN Namespace for Public Ids February 2001 normalized before encoding them into URNs. As described in ISO 8879[3], a public identifier is normalized by removing all leading and trailing whitespace and replacing all remaining sequences of two or more whitespace characters with a single space. For public identifiers that are not FPIs, the Namespace Specific String (NSS) for URNs in the "publicid" namespace has the following structure: urn:publicid:{public-identifier-text} The character set of public identifiers is constrained by XML[1]. Most of the legal public identifier characters are also legal characters in URNs. Unless otherwise noted, the characters in the {public-identifier-text} are directly transcribed from the corresponding character in the public identifier. The following exceptions are made: + Spaces in the public identifier are transcribed as "+" characters. Whitespace normalization must be performed before constructing a URN in the "publicid" namespace, therefore the sequence of characters "++" should never occur in such URNs. + Literal "+" characters in the public identifier must be %-encoded. + Literal ":" characters in the public identifier must be %-encoded. + The reserved characters that may appear in public identifiers, "%", "/", "?", and "#", must be %-encoded. Formal Public Identifiers are a subset of public identifiers. They are strings composed from the same range of characters, but have an explicit internal structure. The structure of Formal Public Identifiers is normatively described in SGML[3], we review it here for convenience. Most Formal Public Identifiers consist of the following fields, in this order: an owner identifier, a public text class, a public text description, a public text language or public text designating sequence, and an optional public text display version. Owner identifiers may begin with "-//" or "+//", otherwise "//" is used to delimit fields in the FPI with the exception of the public text class which is delimited from the public text description by a space. In other words, most FPIs look like this: Walsh, et. al. Expires August 14, 2001 [Page 3] owner//class description//language//version and most owners begin with "+//" or "-//", although they are not required to. Here are some example FPIs: +//IDN python.org//DTD XML Bookmark Exchange Language 1.0//EN//XML -//OASIS//DTD DocBook XML V4.1.2//EN -//ArborText::prod//DTD Help Navigation Document::19970708//EN ISO/IEC 10179:1996//DTD DSSSL Architecture//EN ISO 8879:1986//ENTITIES Added Latin 1//EN An algorithm for correctly identifying a Formal Public Identifier and determining the various fields within it is out of scope for this document. We begin our discussion of the representation of FPIs in our URN namespace under the assumption that these steps have already been taken. The Namespace Specific String (NSS) for the URNs in the "publicid" namespace that represent Formal Public Identifiers have the following structure: urn:publicid:{owner-identifier}:{text-class} :{text-description}:{language|designating-sequence} {:display-version}? Where: {owner-identifier} is derived from the owner identifier in the FPI. Owner identifiers in FPIs have one of three forms: "+//" followed by a string, "-//" followed by a string, or a string that does not contain "//". The following rules apply to derive a URN {owner-identifier} from the owner identifier in an FPI: - Owner identifiers that begin "+//" are transcribed into the URN {owner-identifier} by replacing "+//" with "+:" and transcribing the remaining string. - Owner identifiers that begin "-//" are transcribed into the URN {owner-identifier} by replacing "-//" with "-:" and transcribing the remaining string. - All other {owner-identifiers} are transcribed directly from the owner identifier in the FPI. {text-class} is the public text class from the FPI. The public text class of FPIs is constrained by SGML[3] to the following 13 strings: "CAPACITY", "CHARSET", "DOCUMENT", "DTD", "ELEMENTS", "ENTITIES", "LPD", "NONSGML", "NOTATION", "SHORTREF", "SUBDOC", "SYNTAX", or "TEXT". The "publicid" URN namespace explicitly relaxes this Walsh, et. al. Expires August 14, 2001 [Page 4] Internet-Draft A URN Namespace for Public Ids February 2001 constraint. Any string may be used. {text-description} is the public text description transcribed from the FPI. {language} is the public text language transcribed from the FPI. The {language} codes used in "publicid" URNs should be drawn from RFC 3066[6]. {designating-sequence} is the public text designating sequence transcribed from the FPI. Formal Public Identifiers that describe character sets may use the designating sequence (a string defined by ISO 2022[2]) to identify the character set. {display-version} is the public text display version transcribed from the FPI. Most of the legal public identifier characters are also legal characters in URNs. Unless otherwise noted, the characters in the {owner-identifier}, {text-class}, {text-description}, {language}, {designating-sequence}, and {display-version} are directly transcribed from the corresponding character in the Formal Public Identifier. The following exceptions are made: + Spaces in the FPI are transcribed as "+" characters. Whitespace normalization must be performed before constructing a URN in the "publicid" namespace, therefore the sequence of characters "++" should never occur in such URNs. + Literal "+" characters in the FPI, except at the beginning of {owner-identifier}s for FPIs that have the "+//"-form of owner identifier, must be %-encoded. The "+" characters at the beginning of {owner-identifier}s for FPIs that have the "+//"-form of owner identifier, must not be %-encoded. + The sequence "::" in the owner identifier or public text description is transcribed as "::"; all other uses of a literal ":" in the FPI must be %-encoded. + The reserved characters that may appear in FPIs, "%", "/", "?", and "#", must be %-encoded. A small subset of Formal Public Identifiers cannot be represented by this namespace. An FPI cannot be represented if either of the following conditions applies: + After transcription, the {owner-identifier}, {text-class}, {text-description}, {language}, or {designating-sequence} would be empty. Allowing any of these fields to be empty could introduce ambiguous "::" sequences into the URN. Walsh, et. al. Expires August 14, 2001 [Page 5] Internet-Draft A URN Namespace for Public Ids February 2001 + The FPI uses the optional unavailable text indicator defined in SGML[3] but rarely used in practice. Relevant ancillary documentation: Extensible Markup Language (XML) Version 1.0 Second Edition[1] Standard Generalized Markup Language (SGML)[3] Registration procedures for public text owner identifiers[4] Identifier uniqueness considerations: The identifier uniqueness considerations for URNs in the "publicid" namespace are the same as the identifier uniqueness considerations for public identifiers. Formal Public Identifiers with registered owner identifiers are required to be unique. For unregistered owner identifiers and informal public identifiers, they may or may not be unique, no enforcement policy can be asserted. Identifier persistence considerations: The persistence of URNs in the "publicid" namespace is the same as the persistence of the corresponding public identifier. Process of identifier assignment: Identifiers in the "publicid" namespace may be assigned by the same policies and procedures as public identifiers. Process of identifier resolution: Identifiers in the "publicid" namespace may be resolved by the same policies and procedures as public identifiers. Rules for Lexical Equivalence: Whitespace normalization is performed before constructing a URN in the "publicid" namespace, so such URNs are lexically equivalent if they are lexically identical. Conformance with URN Syntax: No special considerations. Validation mechanism: None specified. Walsh, et. al. Expires August 14, 2001 [Page 6] Internet-Draft A URN Namespace for Public Ids February 2001 Scope: Global 3. Examples The following examples are not guaranteed to be real. They are listed for pedagogical reasons only. "ISO/IEC 10179:1996//DTD DSSSL Architecture//EN" becomes "urn:publicid:ISO%2FIEC+10179%3A1996:DTD:DSSSL+Architecture:EN" "ISO 8879:1986//ENTITIES Added Latin 1//EN" becomes "urn:publicid:ISO+8879%3A1986:ENTITIES:Added+Latin+1:EN" "-//OASIS//DTD DocBook XML V4.1.2//EN" becomes "urn:publicid:-:OASIS:DTD:DocBook+XML+V4.1.2:EN" "+//IDN python.org//DTD XML Bookmark Exchange Language 1.0//EN//XML" becomes "urn:publicid:+:IDN+python.org:DTD:XML+Bookmark+Exchange+Language+1.0:EN:XML" "-//ArborText::prod//DTD Help Navigation Document::19970708//EN" becomes "urn:publicid:-:ArborText::prod:DTD+Help+Navigation+Document::19970708:EN" "foo" becomes "urn:publicid:foo" "3+3=6" becomes "urn:publicid:3%2B3=6" "-//Acme, Inc.//DTD General Book Markup Version 1.0" becomes "urn:publicid:-%2F%2FAcme,+Inc.%2F%2FDTD+General+Book+Markup+Version+1.0" because it is not an FPI (it has no public text language or designating sequence). 4. Security Considerations There are no additional security considerations other than those normally associated with the use and resolution of URNs in general. References [1] W3C, XML WG, "Extensible Markup Language (XML) 1.0 Second Walsh, et. al. Expires August 14, 2001 [Page 7] Internet-Draft A URN Namespace for Public Ids February 2001 Edition", February 1998, . [2] JTC 1, SC 2, "ISO (International Organization for Standardization) ISO 2022:1994 Information technology -- Character code structure and extension techniques (fourth edition).", 1994. [3] JTC 1, SC 34, "ISO 8879:1986 Information processing -- Text and office systems -- Standard Generalized Markup Language (SGML)", 1986. [4] JTC 1, SC 34, "ISO/IEC 9070:1991 Information technology -- SGML support facilities -- Registration procedures for public text owner identifiers", 1991. [5] Moats, R., "URN Syntax", RFC 2141, May 1997. [6] Alvestrand, H., "Tags for the Identification of Languages", RFC 3066, January 2001. Authors' Addresses Norman Walsh Sun Microsystems, Inc. One Network Drive MS UBURO2-201 Burlington, MA 01803-0902 US EMail: Norman.Walsh@East.Sun.COM John Cowan Reuters Health Information 1700 Broadway, 31st Floor New York, NY 10019 US EMail: jcowan@reutershealth.com Paul Grosso Arbortext, Inc. 1000 Victors Way Ann Arbor, MI 48108-2744 US EMail: pgrosso@arbortext.com Walsh, et. al. Expires August 14, 2001 [Page 8] Internet-Draft A URN Namespace for Public Ids February 2001 Full Copyright Statement Copyright (C) The Internet Society (2001). All Rights Reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Internet organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by the Internet Society or its successors or assigns. This document and the information contained herein is provided on an "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Acknowledgement Funding for the RFC editor function is currently provided by the Internet Society. Walsh, et. al. Expires August 14, 2001 [Page 9]