Namespace Registration for International Standard Book Number (ISBN) ISO 2108:2017 Namespace ID: ISBN Requested of IANA. Version: 2 (revision of earlier registration, RFC 3187). Date: 2017-06-12 Registrant: Name: Stella Griffiths E-mail: stella&isbn-international.org Affiliation: Executive Director, The International ISBN Agency Ltd Address: International ISBN Agency Ltd c/o EDItEUR, United House, North Road, London, N7 9DP, U.K. Web URL: http://www.isbn-international.org/ Requesting entity is a standards development organization revising the namespace registration which was based on an earlier version of the ISBN standard. Purpose: The ISBN (International Standard Book Number) is a unique machine-readable identifier, which identifies unambiguously any edition of a text-based monographic publication that is available to the public. The identifier is defined in ISO Standard 2108. The identifier has been in use for over 45 years and has revolutionised the international book trade. 150 countries and territories are officially ISBN members, and more are applying to join the system. The administration of the ISBN system is carried out on three levels: • International ISBN agency, • ISBN registration agencies, • Registrants. The International ISBN agency is located in London, UK. The main functions of the International ISBN Agency are: • To promote, co-ordinate and supervise the world-wide use of the ISBN system. • To define registration groups and the areas to which they relate • To allocate ranges of unique registrant elements to specific registration groups for assignment by ISBN registration agencies • To advise on the establishment and functioning of ISBN registration agencies • To advise ISBN registration agencies on the allocation of registrant elements • To ensure that an accurate and complete register of GS1 elements, registration group elements and registrant elements is available. Information about ISBN usage in general can be found on the International ISBN Agency website http://www.isbn-international.org/. Syntax: The namespace-specific string of 'ISBN' URNs is either a 13 digit number (representing an ISBN-13) or a 10 digit number (representing an ISBN-10).In either 13 or 10 digit form, the ISBN may present with hyphens which are used to indicate the separate constituent parts of the number and for human readability. Hyphens are not semantically significant and are ignored during lexical comparison. In addition, the final "digit" of the 10-digit format may instead be an upper-case "X". All new ISBN assignments are based on ISBN-13. If a 10 digit ISBN is found on the resource, it should be converted to a 13 digit number, following the rules set out later in this section, before being encoded into the URN framework. According to the rules of the ISBN standard, such conversion does not create a new ISBN for the book, but a new representation of the existing ISBN. Editions of ISO 2108 up to and including the 1992 edition defined ISBN as a ten digit number (the last character could be an upper case "X" as well) which was divided into four parts (three of which were of variable length) usually separated by optional hyphens when printed. The parts were defined as follows and in this order: • a group identifier specifying a group of publishers, based on national, geographic, language or some other criteria; • the publisher identifier; • the title identifier; and • a modulo 11 check digit, using upper-case “X” instead of 10; the details of the calculation being specified in ISO 2108-1992 and earlier editions. ISBN in 10 digit form is deprecated, but it remains possible to incorporate ISBN-10s in URN: Example 1: URN:ISBN:951-0-18435-7 (ISBN-10) ISBN-10 was in use from the 1970s until ISBN-13 replaced it in January 2007. The ISBN in thirteen digit form is defined by the ISO Standard 2108-2005 and later editions. It was previously referred to as “ISBN-13” to distinguish it from “ISBN-10”, but since all ISBNs are now valid only in the 13 digit format and ISBN-10 is deprecated, ISBN-13 should be referred to as “ISBN”, although in this document ISBN-13 is used for the sake of clarity. The ISBN-13 is a thirteen digit number which is divided into five parts which may be separated by optional hyphens when printed. The first and the last parts have a fixed length, but the other parts have variable length. The parts of ISBN-13 are defined as follows and in this order: • the GS1 element which is a 3 digit prefix (at the time of writing, legal values were 978 and 979; additional values shall be defined and confirmed at the appropriate time) specified by the International ISBN Agency; • a registration group element that specifies the registration group; it identifies the national, geographic, language, or other such grouping to which a publisher belongs; • the registrant element; • the publication element; and • a modulo 10 check digit. Example 2: URN:ISBN:978-951-0-18435-6 (Example 1 expressed as ISBN-13) Variable length elements have been allocated in such a way that the hyphens are not necessary for machine to machine transmission since only certain registrant element ranges are valid in any given registration group. The registration group and registrant element rules are devised and maintained by the International ISBN Agency. They are available from https://www.isbn-international.org/range_file_generation. Q-component semantics is undefined at the time of writing. When an ISBN identifies a book in a single product form (e.g. a book in PDF format), F-component may be used. However, if an ISBN identifies a publication or series that is comprised of multiple volumes such as The Lord of the Rings, f-component is not applicable. R-component semantics is undefined at the time of writing. Rules for Lexical Equivalence: Since ISBN-13 is the canonical form for ISBN, all equivalence checking should be performed using that format. Checking for lexical equivalence should be performed as follows: • If the NSS is an ISBN-10, first convert it to ISBN-13: the GS1 element 978 should be inserted before the first 9 digits of the 10-digit ISBN which are then followed by a revised check digit calculated using the modulus 10 check digit formula as described in the ISBN Users’ Manual. • Remove all hyphens from the NSSs • If the modified NSSs are equivalent on a character-by-character basis, the URNs are equivalent, otherwise not. Rules for Semantic Equivalence Where an invalid or incorrect ISBN has subsequently been corrected, each of the incorrect and correct ISBNs can be regarded as semantically equivalent because they both refer to the same product. However, resolving these URN:ISBNs may produce different results. For instance, the publisher’s bibliographic database may contain only the incorrect ISBN assigned by the publisher, while the national bibliography usually contains both ISBNs. ISBN assignment: ISBNs are assigned according to the rules of the ISBN Standard and the policies and procedures of the International ISBN Agency. These rules are specified in the ISBN Standard (ISO 2108), in the ISBN Users’ Manual (https://www.isbn-international.org/content/isbn-users-manual) and other guidelines published from time to time by the International ISBN Agency. All publishers must apply for ISBNs for their books from the appropriate ISBN registration agency. Some publishers with sizable publishing programmes may receive a dedicated block of ISBNs from the ISBN registration agency and may then assign individual ISBNs from these blocks to their publications as necessary. In other cases ISBN registration agencies may assign individual ISBNs directly to the publisher for a specific publication. The basic guideline for assigning ISBNs is: • Format/means of delivery is irrelevant to the decision whether a product needs an ISBN or not. If the content meets the scope of the Standard, it qualifies to be assigned an ISBN, no matter what the format of the delivery system. Each separate publication should have a separate ISBN. The definition of a separate or new edition is normally based on one or more of the following six criteria: • A change in the product form: for example, the hard cover edition, the paperback edition, the library-binding edition and each e-book edition would each get a separate ISBN. • A change in the product form details: for example, the jacketed hard cover edition would get a separate ISBN from the unjacketed hard cover edition if both are available at the same time. • A change in the text, excluding packaging or minor changes such as correcting a spelling error. • A change in the language of the text. • Specifically in the case of a digital publication a change in the usage constraints, e.g. adjusting the usage settings so that printing is allowed in one, but not in another. • A change in the imprint name or publisher. Although these rules are clear, their interpretation may sometimes vary. As RFC 2288 points out, The choice of whether to assign a new ISBN or to continue to use the existing one when publishing a revised printing of an existing edition of a publication or even a revised edition of a publication is somewhat subjective since to qualify for a new ISBN the textual changes to the publication concerned should be ”significant”. The use of ISBNs within the URN framework simply reflects existing practices and interpretations of these rules. Note that it is likely that an ISBN URN may resolve to many instances of the publication (many URLs). These instances may be fully identical, or there may be some minor differences between them. Publishers have also, on occasion, incorrectly re-used the same ISBN for another book. This reasonably rare kind of human error does not threaten or undermine the value of the ISBN system as a whole. Neither does it pose a serious threat to the URN resolution service based on ISBNs. An error such as this should only lead to the retrieval of two or more bibliographic records describing two different monographic publications. Based on the information in the records, a user can choose the correct record from the result set. Most national bibliographies and especially Books in Print services correct ISBN mistakes. Library systems often provide cross references ("incorrect ISBN -> correct ISBN"). Other systems, such as those used by publishers and the book trade, may not always be aware of both ISBNs, which may have an impact on the URN resolution process. Further details on the process of assigning ISBNs can be found in the section ”ISBN assignment” above. Security and Privacy: This document proposes the means of encoding ISBNs within the URN framework. An ISBN-based URN resolution service is depicted here both for ISBN-10 and ISBN-13, but only at a fairly generic level; thus questions of secure or authenticated resolution mechanisms are excluded. It does not deal with means of validating the integrity or authenticating the source or provenance of URNs that contain ISBNs. Issues regarding intellectual property rights associated with objects identified by the ISBNs are also beyond the scope of this document, as are questions about rights to the databases that might be used to construct resolvers. Interoperability: ISBN is a unique and persistent identifier. An ISBN, once it has been assigned, must never be re-used for another book. Moreover, a single product form of a book must never get a new ISBN unless there are significant changes to content or other changes such as a different publisher. There may be multiple product forms of a single publication such as a novel. In this case each product form shall receive a different ISBN. ISO standard ISTC (International Standard Text Code, ISO 21047-2009) enables identification of abstract, textual works. See http://www.istc-international.org/ for more information. There are no characters in either ISBN-10 or ISBN-13 which would require percent-encoding. Resolution: General Resolution means the act of supplying services related to the identified entity such as translating the identifier into one or more current locators for the entity, or delivering metadata about the entity in an appropriate format. For the time being, there is no decision on what resolution services should be supported in the URN:ISBN namespace. Locating the appropriate resolver In the URN:ISBN namespace, locating the appropriate resolver (resolvers) is based on the ISBN semantics. All ISBN elements except the check digit (0-9 for ISBN-13, 0-9 or X for ISBN-10) must be taken into account. The registration group and registrant element assignments are managed in such a way that the hyphens are not needed to parse the ISBN unambiguously into its constituent parts. However, the ISBN is normally transmitted and displayed with hyphens to make it easy for humans to recognize these elements without having to make reference to, or have knowledge of, the assignments of registration group and registrant elements. In ISBN-10, registration group element codes such as 91 for Sweden were unique. In ISBN-13 only the combination of GS1 element and registration group elements is guaranteed to be unique. 978-951 and 978-952 both mean Finland, but 979-951 and 979-952 almost certainly will not; registration group element(s) for Finland are not yet known for ISBNs starting with 979. The Finnish URN registry is maintained by the national library. The service is capable of resolving ISBN-based URNs. URNs starting with URN:ISBN:978-951 or URN:ISBN:978-952 are mapped into appropriate URL addresses in a table maintained within the registry. Applications, such as the national bibliography or the open archive of a university, can use the URN as the address of the resource. There is just one place (the registry) where the location information must be kept up to date. ISBN-13 GSI element/registration group element combinations (and the corresponding ISBN-10 registration group identifiers, if any) usually designate a country, but occasionally a single combination/ISBN-10 group identifier is used to indicate a language area. For instance, "978-3" (or "3" in ISBN-10) is used in Germany, Austria, and the German speaking parts of Belgium and Switzerland. Note that the GS1 element and registration group element combination "979-3" has not yet been assigned, and indeed may not be valid in the future; there is no intention to allocate the registration group elements in the same way as was done with ISBN-10. Also at the time of writing, there are two regional registration groups: "978-976" is used by the Caribbean community and "978-982" by the South Pacific region (see https://www.isbn-international.org/range_file_generation). The registrant element may or may not be used for resolution purposes, depending on whether individual publishers have set up their resolution services. The publication element shall enable targeting the individual publication. Due to the lack of URN support in, e.g., web browsers, the URNs are usually expressed as HTTP URIs when embedded in documents. The Finnish URN registry is located at http://urn.fi, and URNs are therefore expressed in the form http://urn.fi/. For example, the URN http://urn.fi/URN:ISBN:978-952-10-3937-9 identifies Sami Nurmi's doctoral dissertation "Aspects of Inflationary Models at Low Energy Scales". The Finnish URN registry cannot resolve URN:ISBNs with non-Finnish registration group element values until other countries establish their registries, and all these services become aware of each other and their respective registration group responsibility domains and are able to communicate with each other. Thus the Finnish registry can deal with URN:ISBN instances with “978-91” (indicating Sweden) if and only if the Swedish registry exists, its address is known to the Finnish peer and the Swedish service is capable of receiving and processing requests from other registries. If a registration group element does not identify a single country but a language area, there are at least two means for locating the correct national bibliography. First, it is possible to define a cascade of URN registries - for instance, German, Austrian and Swiss national registries, in this order - which should collectively be aware of resolution services such as national bibliographies for ISBNs starting with "978-3". If the German registry is not able to find an authoritative resolution service, the request could be passed to the Austrian one, and if there are still no hits, finally to the Swiss service. Second, the registrant element ranges assigned to the publishers in Germany, Austria and Switzerland by the ISBN Agencies could be defined directly into the national registries. This method would be more efficient than cascading, since the correct resolution service would be known immediately. The choice between these two and any other possible options should be made when the establishment of the European network of URN registries reaches this level of maturity. In some exceptional cases - notably in the US and in the UK, where international companies carry out a significant proportion of publishing - the information provided by the registration group identifier may not always be fully reliable. For instance, some monographs published by international publishing companies may get an ISBN that has the first two elements "978-3". This may be technically appropriate when the headquarters of the publisher is located in Germany. Information about such a book may not always be available in the German national bibliography, but may be available through others such as the Library of Congress systems. Unfortunately, the German/Austrian/Swiss URN registries that should be contacted in the case of ISBNs that begin “978-3” may not be aware of the appropriate resolution service. However, the problem posed by international publishers may well be less severe than it looks. Some international publishers (Springer, for example, which has its headquarters in Germany) give their whole production to the national library of their home country as legal deposit. Thus everything published by Springer should be resolvable via the German national bibliography. A large union catalogue, such as WorldCat (http://worldcat.org) maintained by OCLC could be used to complement the resolution services provided at the national level, or as the default service, if no national services exist or are known to the registry from which the query originates. Due to the semantic structure of ISBN, even the registrant element can be used as a "hint". Technically, it is possible to establish a large number of URN:ISBN resolution services maintained by different kinds of organizations. For instance, "978-951-0" is the unique ISBN registrant element of the largest publisher in Finland, WSOY. Resolution requests for ISBNs starting with "978-951-0" could be passed to, and dealt with, by the publisher's URN resolver, if and when such service is established. Resolvers maintained by e.g. publishers can be linked to other resolvers, such as the one maintained by the national library, in order to improve reliability and to support additional resolution services. Persistence of URN:ISBN resolution services is largely dependent on the persistence of publishers, libraries and other organizations providing those services. Thus some services, independent of the base technology chosen, may disappear or their content may change much sooner than some peer solutions. Documentation: ISBN Users’ Manual https://www.isbn-international.org/content/isbn-users-manual Additional Information: Web URL: http://www.isbn-international.org/ Revision Information: This registration is based on ISBN standard version ISO 2108:2017.