Network Working Group                                          J. Hakala
Request for Comments: 3188                   Helsinki University Library
Category: Informational                                     October 2001


                 Using National Bibliography Numbers as
                         Uniform Resource Names

Status of this Memo

   This memo provides information for the Internet community.  It does
   not specify an Internet standard of any kind.  Distribution of this
   memo is unlimited.

Copyright Notice

   Copyright (C) The Internet Society (2001).  All Rights Reserved.

Abstract

   This document discusses how national bibliography numbers (persistent
   and unique identifiers assigned by the national libraries) can be
   supported within the URN (Uniform Resource Names) framework and the
   syntax for URNs defined in RFC 2141.  Much of the discussion is based
   on the ideas expressed in RFC 2288.

1. Introduction

   As part of the validation process for the development of URNs the
   IETF working group agreed that it is important to demonstrate that
   the current URN syntax proposal can accommodate existing identifiers
   from well established namespaces.  One such infrastructure for
   assigning and managing names comes from the bibliographic community.
   Bibliographic identifiers function as names for objects that exist
   both in print and, increasingly, in electronic formats.  RFC 2288
   [Lynch] investigated the feasibility of using three identifiers
   (ISBN, ISSN and SICI) as URNs.

   This document will analyse the usage of national bibliography numbers
   (NBNs) as URNs.  The need to extend analysis to new identifier
   systems was briefly discussed in RFC 2288 as well, with the following
   summary: "The issues involved in supporting those additional
   identifiers are anticipated to be broadly similar to those involved
   in supporting ISBNs, ISSNs, and SICIs".

   A registration request for acquiring a Namespace Identifier (NID)
   "NBN" for national bibliography numbers has been written by the
   National Library of Finland on the request of the Conference of
   Directors of National Libraries (CDNL) and the Conference of the
   European National Librarians (CENL).  Chapter 5 contains a URN
   namespace registration request modeled according to the template in
   RFC 2611.

   The document at hand is part of a global co-operation of the national
   libraries to foster identification of electronic documents in general
   and utilisation of URNs in particular.  Some national libraries,
   including the national libraries of Finland, Norway and Sweden, are
   already assigning NBN-based URNs for electronic resources.

   We have used the URN Namespace Identifier "NBN" for the national
   bibliographic numbers in examples below.

2. Identification vs. Resolution

   As a rule the national bibliography numbers identify finite,
   manageably-sized objects, but these objects may still be large enough
   that resolution to a hierarchical system is appropriate.

   The materials identified by a national bibliography number may exist
   only in printed or other physical form, not electronically.  The best
   that a resolver will be able to offer in this case is bibliographic
   data from a national bibliography database, including information
   about where the physical resource is stored in a national library's
   holdings.

   The URN Framework provides resolution services that may be used to
   describe any differences between the resource identified by a URN and
   the resource that would be returned as a result of resolving that
   URN.  However, NBNs will be used for instance to identify resources
   in digital Web archives created by harvester robot applications.  In
   this case, NBN will identify exactly the resource the user expects to
   see.

3. National bibliography numbers

3.1 Overview

   National Bibliography Number (NBN) is a generic name referring to a
   group of identifier systems utilised by the national libraries and
   only by them for identification of deposited publications which lack
   an identifier, or to descriptive metadata (cataloging) that describes
   the resources.  In many countries legal (or voluntary) deposit is
   being extended to electronic publications.

   Each national library uses its own NBN strings independently of other
   national libraries; there is no global authority which controls them.
   For this reason NBNs are unique only on national level.  When used as
   URNs, NBN strings must be augmented with a controlled prefix such as
   country code.  These prefixes guarantee uniqueness of the NBN-based
   URNs on the global scale.

   NBNs have traditionally been given to documents that do not have a
   publisher-assigned identifier, but are cataloged to the national
   bibliography.  NBNs can be seen as a fall-back mechanism: if no
   other, better established identifier such as ISBN can be given, an
   NBN is assigned.  In principle, NBN usage enables identification of
   any Internet document.  Local policies may limit the NBN usage to a
   much smaller subset of documents.

   Some national libraries (e.g., Finland, Norway, Sweden) have
   established Web-based URN generators, which enable authors and
   publishers to fetch NBN-based URNs for their network documents.  At
   least national libraries of Sweden and Finland are harvesting and
   archiving domestic Web documents (and a number of other libraries
   plan to start this activity), and long-time preservation of these
   materials requires persistent and unique identification.  NBNs can be
   and are in fact already used as internal identifiers in these Web
   archives.

   Both syntax and scope of NBNs can be decided by each national library
   independently.  Typically, an NBN consist of one or more letters
   and/or digits.  This simple syntax makes NBNs infinitely extensible
   and very suitable for e.g., naming of the Web documents.  For
   instance the application used by the national library of Finland for
   Web harvesting creates NBNs which are based on the MD5 checksum of
   the archived resource.

3.2 F-code

   F-code is the NBN used by the National Library of Finland.

   F-codes have been used since early 20th century to identify catalogue
   cards and later MARC records in the national bibliography.  In 1998
   the national library decided to enable the Finnish authors and
   publishers to assign F-codes to their Internet documents, if these
   documents do not qualify for other identifiers such as ISBN.  F-
   codes, embedded into URNs, can be fetched from the URN generator
   (http://www.lib.helsinki.fi/cgi-bin/urn.pl) developed in co-operation
   between the national library of Finland and the Lund University
   library, NETLAB unit.  Attached to the generator there is a user
   guide (http://www.lib.helsinki.fi/meta/URN-opas.html; only in
   Finnish), which tells the users how to use URNs.

   F-codes are also used within the Web harvesting and archiving
   software (http://www.csc.fi/sovellus/nedlib/), which has been built
   for the Networked European Deposit Library (NEDLIB) project (see
   http://www.kb.nl/nedlib).  NEDLIB harvester calculates MD5 checksum
   for each archived resource, and then builds an NBN-based URN from the
   checksum.  The URN serves then as a unique identifier to the archived
   resource.  Traditional identifiers can not be used for this purpose,
   since there may for instance be several variants of a book which
   (quite rightly so) all have the same ISBN.  Moreover, identifiers
   embedded into a document do not necessarily belong to the document
   itself; thus the Web archiving application can not trust the
   identifiers embedded into the body of the document.

   The F-code built by the URN generator consist of:

   Prefix (for example fe)
   Year (YYYY; for example 1999)
   Number (for example 1055)

   The generator also adds namespace identifier "NBN" and ISO 3166
   country code.  Thus a URN based on F-code would in this case be for
   instance urn:nbn:fi-fe19991055.

   URNs created by the Web archiving application have similar overall
   structure, except that prefix (which may be defined by the operator)
   is fea and year is not used.  An example:  urn:nbn:fi-fea-
   5c5875e6e49ae649cad63e5ee4f6c346.

   F-codes never need any special encoding when used as URNs, since they
   consist of alphanumeric codes only (0-9, a-z).  This is often the
   case for other national libraries' NBN systems as well.

3.3 Encoding Considerations and Lexical Equivalence

   Embedding NBNs within the URN framework usually presents no
   particular encoding problems, since all of the characters that can
   appear in commonly used NBN systems can be expressed in special
   encoding, as described in RFC 2141 [MOATS].

   When an NBN is used as a URN, the namespace specific string will
   consist of three parts: prefix, consisting of either a two-letter ISO
   3166 country code or other registered string, delimiting character
   which is either hyphen (-) or colon (:), and NBN string assigned by
   the national library.  Delimiting characters are not lexically
   equivalent.

   Hyphen is always used for separating the prefix and the NBN string.

   Colon is used as the delimiting character if and only if a country
   code-based NBN namespace is split further in smaller sub-namespaces.
   If there are several national libraries in one country, these
   libraries can split their national namespace into smaller parts using
   this method.

   A national library may also assign a trusted organisation(s) its own
   sub-namespace.  For instance, the national library of Finland has
   given Statistics Finland (http://www.stat.fi/index_en.html) a sub-
   namespace "st" (e.g., urn:nbn:fi:st:).  The Finnish Council of State
   (http://www.vn.fi/vn/english/index.htm) will use sub-namespace "vn"
   (e.g., urn:nbn:fi:vn).

   Non-ISO 3166-prefixes, if used, must be registered on the global
   level. The Library of Congress will maintain the central register of
   reserved codes.  This register will be available to the national
   libraries and other users in the Web.

   Sub-namespace codes beneath a country-code-based namespace need to be
   registered on the national level by the national library which
   assigned the code.  The national register must be available in the
   Web and should also be linked to the global register maintained by
   the Library of Congress.

   Two-letter codes may not be used as non-ISO prefixes, since all such
   codes are reserved for existing and possible future ISO country
   codes. If there are several national libraries in one country who use
   the same prefix - for instance, a country code -, they need to agree
   on how to split the namespace between them.

   Models:
   URN:NBN:<ISO 3166 country code>-<assigned NBN string>
   URN:NBN:<ISO 3166 country code>:<sub-namespace code>-<assigned NBN
   string>
   URN:NBN:<non-ISO 3166 prefix>-<assigned NBN string>

   Examples:
   URN:NBN:fi-fe19981001 (A "real" URN assigned by the National Library
   of Finland).

3.4 Resolution of NBN-based URNs

   The (usually) country code-based prefix part of the URN namespace
   specific string will provide a guide to where to find a resolution
   service, and the NBN register will identify the assigning agency.
   Once the NBN-based URN resolution is in global usage, the number of
   prefixes will slowly approach and may eventually exceed the number of
   national libraries.

   If NBN assignment for a given country is limited to the national
   bibliography database, then all NBN-based URNs for that country will
   be resolved there.  In one model these databases contain detailed
   resource descriptions including URLs, which will point both to the
   copy of the document in the Internet and to the copy in the national
   library's (legal) deposit collection.  Due to the limitations in the
   usage of legal deposit documents it is possible that the deposited
   electronic materials can not be delivered in electronic form outside
   the premises of the national library.

   If it is possible for the authors and publishers to retrieve NBNs to
   Web documents and there is no obligation to deposit thus identified
   documents to the national library, URN resolution service is not
   possible without a national Web index and archive, maintained by the
   national library or other organisation(s).  A Web index/archive will
   also resolve machine-generated URNs to the archived Web documents.

3.5 Additional considerations

   Guidelines adopted by each national library define when different
   versions of a work should be assigned the same or differing NBNs.
   These rules apply only if identifier assignment is done manually.  If
   identifiers are allocated programmatically, the only criteria that
   can be used is that two documents which are identical on the bit
   level (have the same MD5 checksum) are deemed identical and should
   receive the same NBN.  The likelihood of this happening to dissimilar
   documents is about 2^64, according to the RFC 1321.

   The rules governing the usage of NBNs are less strict than those
   specifying the usage of ISBN or other, better established
   identifiers. Since the NBNs have up to now been given only by the
   personnel (cataloguers) working in the national libraries, the
   identifier assignment has in practice been well co-ordinated.

   A NBN-based URN will resolve to single instance of the work if
   identifier assignment has been automatic.  Given the nature of NBNs
   it is also likely that different versions of the same work will
   receive different NBNs even if the identifier is given manually.

4. Security Considerations

   This document proposes means of encoding several existing
   bibliographic identifiers within the URN framework.  This document
   does not discuss resolution except at a very generic level; thus
   questions of secure or authenticated resolution mechanisms are out of
   scope.  It does not address means of validating the integrity or
   authenticating the source or provenance of URNs that contain
   bibliographic identifiers.  Issues regarding intellectual property

   rights associated with objects identified by the various
   bibliographic identifiers are also beyond the scope of this document,
   as are questions about rights to the databases that might be used to
   construct resolvers.

5. Namespace registration

   URN Namespace ID Registration for the National Bibliography Number
   (NBN)

   Namespace ID:

   NBN

   This Namespace ID has been in production use in demonstrator systems
   since summer 1998; thousands of URNs from this namespace have already
   been delivered in Finland, Sweden and Norway.

   Registration Information:

   Version: 3
   Date: 2001-01-30
   The first registration of the NID "NBN" was done via the URN WG in
   1998. The second, slightly edited registration request was done in
   1999.

   Declared registrant of the namespace:

   Name: Juha Hakala
   E-mail: juha.hakala@helsinki.fi
   Affiliation: Helsinki University Library - The National Library of
   Finland, Conference of European National Librarians (CENL) and
   Conference of Directors of National Libraries (CDNL)
   Address: P.O.Box 26, 00014 Helsinki University, Finland

   Both CENL and CDNL made decisions to foster the usage of URNs during
   1998.  The latter organisation has set up a working group for this
   purpose.  One item in the common work plan is utilisation of national
   bibliography numbers as URNs for identification of grey literature
   published in the Internet.  The NBN namespace will be available for
   free for all national libraries in the world.

   Declaration of syntactic structure:

   The namespace specific string will consist of three parts:

   prefix, consisting of either a two-letter ISO 3166 country code or
   other registered string and sub-namespace codes,

   delimiting characters (colon (:), or hyphen (-), and

   NBN string assigned by the national library.

   Colon is used as a delimiting character only within the prefix,
   between ISO 3166 country code and sub-namespace code, which splits
   the national namespace into smaller parts.  This technique can be
   used when there are several national libraries, which all need their
   own namespaces, or when the national library allows trusted partners
   to set up their own sub-namespaces within the national NBN namespace.

   Dividing non-ISO 3166-based namespaces further with sub-namespace
   codes is not allowed.

   Hyphen is used as a delimiting character between the prefix and the
   NBN string.  Within the NBN string, hyphen can be used for separating
   different sections of the code from one another.

   Non-ISO prefixes used instead of the ISO country code must be
   registered.  A global registry, maintained by the Library of
   Congress, will be created and made available via the Web.  Contact
   information: nbn.register@loc.gov.us.

   All two-letter codes are reserved for existing and possible future
   ISO country codes and may not be used as non-ISO prefixes.

   Sub-namespace codes must be registered on the national level by the
   national library which assigned the code.  The register must be
   available via the Web, and it should be accessible via the global
   registry set up by the Library of Congress.

   Models:

   URN:NBN:<ISO 3166 country code>-<assigned NBN string>
   URN:NBN:<ISO 3166 country code:sub-namespace code>-<assigned NBN
   string>
   URN:NBN:<non-ISO 3166 prefix>-<assigned NBN string>

   Example:

   A country code-based URN: URN:NBN:fi-fe19981001 (A URN assigned by
   the National Library of Finland).

   Relevant ancillary documentation:

   National Bibliography Number (NBN) is a generic name referring to a
   group of identifier systems used by the national libraries for
   identification of deposited publications which lack an identifier, or
   to descriptive metadata (cataloguing) that describes the resources.
   Each national library uses its own NBN system independently of other
   national libraries; there is no global authority which controls
   syntax of these identifier systems.

   Each national library can decide freely which resources will receive
   NBNs.  These identifiers have traditionally been assigned to
   documents that do not have a publisher-assigned identifier, but are
   nevertheless catalogued to the national bibliography.  Typically
   identification of grey publications have largely been dependent on
   NBNs.

   Some national libraries (Finland, Norway, Sweden) have established
   Web-based URN generators, which enable authors and publishers to
   fetch NBN-based URNs for their network documents.

   Both syntax and scope of NBNs is decided by each national library
   independently.  Typically, a NBN consist of one or more letters and a
   number.

   Identifier uniqueness considerations:

   NBN strings assigned by two national libraries may be identical.  For
   this reason usage of a controlled prefix in the namespace specific
   string is obligatory in order to guarantee global uniqueness of NBN-
   based URNs.

   In the national level, libraries utilise different policies for
   guaranteeing uniqueness.  A national library may automate the
   delivery of NBN-based URNs.  In this case, the NBNs are assigned
   sequentially by a program (URN generator).

   Identifier persistence considerations:

   Persistence of the NBNs as identifiers is guaranteed by the
   persistence of national libraries and information systems, such as
   national bibliographies, maintained by them.  NBNs have been used for
   several centuries for printed materials.  NBN-based identification of
   electronic documents is a recent practice, but it is likely to
   continue for a very long time.

   Process of identifier assignment:

   Assignment of NBN-based URNs is always controlled on national level
   by the national library / national libraries.  The Conference of
   Directors of National Librarians (CDNL) has established in 1999 a
   task force, which will co-ordinate the URN usage in all national
   libraries.

   National libraries may choose different strategies in assigning NBN-
   based URNs.  One option is assignment by the library personnel only.
   This is done when the document is catalogued into the national
   bibliography.  Thus in this case the national bibliography database
   will serve as the URN resolution service.

   A national library may also set up a URN generator (generators), and
   allow publishers and authors to retrieve NBN-based URNs from there.
   In this case there is no guarantee that the identified resource will
   ever be catalogued into the national bibliography, and URN resolution
   is dependent on Web index/archive.

   Process for identifier resolution:

   URNs based on NBNs will be primarily resolved via the national
   bibliography databases.  In one model these databases contain
   detailed resource descriptions including URLs, which will point both
   to the copy of the document in the Internet and to the copy in the
   national library's (legal) deposit collection.  Due to the
   limitations in the usage of legal deposit documents it is possible
   that the deposited materials can not be delivered outside the
   premises of the national library.

   For those documents not catalogued into the national bibliography
   database URN resolution may take place via national or international
   Web indexes and/or archives.  Nordic national libraries have
   established in autumn 2000 a joint initiative called Nordic Web
   Archive (NWA), which aims at creating a national Web archive into all
   Nordic countries. Indexes to these archive systems will be able to
   act as URN resolution services of any document which a) is or has
   been available via the Web, and b) had an URN embedded into it.

   Country code and additional sub-namespace information will provide a
   guide to where to find appropriate resolution services.  For
   instance, if the country code is "fi", the primary resolution service
   is the national bibliography database.  Secondary resolution service
   is the Web archive.

   Generally, there will be one or more resolution services specified
   for each country, depending on the assignment policy and services of
   the national library.  If NBN assignment is limited to the national
   bibliography database, then all NBN-based URNs for that country will
   be resolved there.  If the authors and publishers have been allowed
   to retrieve NBNs to their Web resources, URN resolution services
   require a national Web archive.  If other organisations have been
   allowed to assign NBNs, they may also set up their own URN resolution
   services.

   Rules for Lexical Equivalence:

   None in the global level.  Any national library may provide its own
   rules, on the basis of its NBN syntax.

   Conformance with URN Syntax:

   All NBNs we know of are ASCII strings consisting of letters (a-z) and
   numbers (0-9).  If NBN contains characters that are reserved in the
   URN syntax, this data must be presented in hex encoded form as
   defined in RFC 2141.  A national library may limit the full scope of
   its NBN strings in URN usage in such a way that there are no reserved
   characters in the URN namespace specific strings.

   Validation mechanism:

   None specified on the global level.  A national library may use NBNs,
   which contain a checksum and can therefore be validated, but this is
   for the time being not a common practice.

   Scope:

   Global.

6. References

   [Daigle] Daigle, L., van Gulik, D., Iannella, R. and P. Faltstrom,
            "URN Namespace Definition Mechanisms", RFC 2611, June 1999.

   [Lynch]  Lynch, C., Preston, C. and R. Daniel, "Using Existing
            Bibliographic Identifiers as Uniform Resource Names", RFC
            2288, February 1998.

   [Moats]  Moats, R., "URN Syntax", RFC 2141, May 1997.

7. Author's Address

   Juha Hakala
   Helsinki University Library - The National Library of Finland
   P.O. Box 26
   FIN-00014 Helsinki University
   FINLAND

   EMail: juha.hakala@helsinki.fi

8. 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.