Digital object identifier

The Digital Object Identifier (DOI) System is a managed system for persistent identification of content-related entities on digital networks. These entities may be content items (digital files, physical objects, abstract works), or any related entities in a content transaction (e.g. licenses, parties, etc.). "DOI" is sometimes used to mean the identifiers within this system; hence the use of the term alone is deprecated unless the meaning is sufficiently clear from an earlier mention or the specific context: instead it should always be used in conjunction with a specific noun. The DOI name is the identifier string that specifies a unique object (the referent) within the DOI System; the DOI syntax is the form and sequence of characters comprising any DOI name, specifically the prefix element, separator, and suffix element; and the DOI System is the functional deployment of DOI names as identifiers in computer sensible form through assignment, resolution, referent description, administration, etc.

The DOI System can be used to identify physical, digital, or abstract entities; these names resolve to data specified by the registrant, and use an extensible metadata model to associate descriptive and other elements of data with the DOI Name. The DOI System is an implementation of the Handle System and of the indecs Content Model and so inherits the design principles and features of each.

The DOI System is implemented through a federation of DOI Registration Agencies, under policies and common infrastructure provided by the International DOI Foundation, which developed and controls the system. The DOI System has been developed and implemented in a range of publishing applications since 2000; by early 2009 approximately 40 million DOI names had been assigned.

International DOI Foundation (IDF)

The International DOI Foundation (IDF), a non-profit organisation created in 1998, is the governance body of the DOI System. It safeguards all intellectual property rights relating to the DOI System, manages common operational features, and supports the development and promotion of the DOI System. The IDF ensures that any improvements made to the DOI System (including creation, maintenance, registration, resolution and policymaking of DOI names) are available to any DOI registrant, and that no third party licenses might reasonably be required to practice the DOI standard.

IDF is controlled by a Board elected by the members of the Foundation, with an appointed Managing Agent who is responsible for co-ordinating and planning its activities. Membership is open to all organizations with an interest in electronic publishing and related enabling technologies. The IDF holds annual open meetings on the topics of DOI and related issues: the 2009 meeting will be held in San Francisco in October

Applications

A DOI name can be assigned to any object that is a form of intellectual property. The term object is used with a specific sense within the DOI system: in the ontology sense of any entity, like the common meaning of the word "thing" (rather than in any computer science sense e.g. Object-oriented programming). So "DOI" is parsed as "digital identifier of an object", rather than "identifier of a digital object". As well as identifying digital media manifestations of intellectual property, DOI names can also identify physical manifestations, performances and abstract works. For example, they can be used to identify: e-texts; images; audio or video items and software, etc. DOI names can also be assigned to related entities in a content transaction (e.g. licenses, parties, etc.).

An entity can be identified at any arbitrary level of granularity. This means that, for instance, DOI names can identify a journal, an individual issue of a journal, an individual article in the journal or a single table in that article. The choice of granularity is left to the assigner, but in the DOI System it must be declared as part of the accompanying metadata; where an application is highly reliant on knowledge of granularity and relationships, the accompanying metadata specified as a requirement by the DOI Registration Agency will normally describe this, using a data dictionary based on the indecs Content Model.

Applications of the DOI System are provided by DOI Registration Agencies (RAs), appointed by the IDF, whose primary role is to provide services to DOI registrants: allocating DOI prefixes, registering DOI names and providing the necessary infrastructure to allow registrants to declare and maintain metadata and state data. RAs are also expected to actively promote the widespread adoption of the DOI System, to cooperate with the IDF in the development of the DOI System as a whole and to provide services on behalf of their specific user community. A list of current RAs is maintained by the International DOI Foundation.

Currently, most applications use a single redirection to a managed URL. It is expected that more applications will begin to make use of additional features in the DOI System, such as multiple resolution (the return as output of several pieces of current information related to a DOI-identified entity — specifically at least one URL plus defined data structures allowing management) and the provision of structured metadata in machine-readable form.

Major applications currently include:

• persistent citation in scholarly materials (journal articles, books, etc.) through CrossRef;
• scientific data sets, through a consortium of leading research libraries and technical information providers, building on work by the German National Library of Science and Technology (TIB);
• European Union official publications, through the EU publications office.

An illustration of an application making good use of DOI System functionality is OECD's publication service SourceOECD: each table or graph in an OECD publication containing a DOI name leads to an Excel file of data underlying the tables & graphs. Further development of such services is planned..

A multilingual European DOI RA activity, mEDRA and a Chinese RA, Wanfang Data, are active in non-English language markets. Expansion to other sectors is planned by the International DOI Foundation.

The DOI System is currently being standardised through the International Organization for Standardization, in its technical committee on identification and description TC46/SC9. In April 2008 the Committee Draft prepared by an international Working Group was approved, after voting by ISO's national bodies, for distribution as a Draft International Standard (DIS). In February 2009, ISO provided further editorial comments on that draft. A further revised draft was submitted to ISO on 2 April 2009. Depending upon the result of ISO activities and further voting, the final standard may be published in 2009 or 2010.

DOI names may be used with other appropriate technology to provide added services, e.g., the OpenURL for context sensitive linking: the DOI directory is OpenURL-enabled so can recognize a user with access to an OpenURL link resolver. Hence on resolving, metadata can be pulled from the DOI agency CrossRef to create an OpenURL targeting the current local link resolver. Such an OpenURL link that contains a DOI name is persistent; publishers who use the CrossRef DOI System to identify their content make their products OpenURL-aware.

Features and benefits

DOI names were developed with the key intended benefits of:

• Persistent identification: each DOI name unequivocally and permanently identifies the object to which it is associated
• Network actionability: each DOI name resolves to one or more web pages or other data assigned by the publisher
• Semantic interoperability: metadata can be provided which allows unambiguous communication to any user, from any place, at any point of a distribution chain, with relevant pieces of information about the identified objects and their relationships

The DOI System uses two underlying technologies plus a social infrastructure to achieve this. The technical infrastructure inherits the features and capabilities of the two underlying technologies: the Handle System and the indecs content model.

The Handle System ensures that the DOI name:

• is not based on any changeable attributes of the entity (location, ownership, or any other attribute that may change without changing the referent's identity);
• is opaque (preferably a "dumb number": a well known pattern invites assumptions that may be misleading, and meaningful semantics may not translate across languages and may cause trademark conflicts);
• is unique within the system (to avoid collisions and referential uncertainty);
• has optional, but nice to have, features that should be supported (human-readable, cut-and-paste-able, embeddable; fits common systems, e.g., URI specification).

And that the DOI name's resolution mechanism:

• is reliable (using redundancy, no single points of failure, and fast enough to not appear broken);
• is scalable (higher loads simply managed with more computers);
• is flexible (can adapt to changing computing environments; useful to new applications);
• is trusted (both resolution and administration have technical trust methods; an operating organization is committed to the long term);
• builds on open architecture (encouraging the leverage efforts of a community in building applications on the infrastructure);
• is transparent (users need not know the infrastructure details).

The Handle System's ability to provide administrative granularity, multiple resolution, and data typing were key to its selection for the DOI System. The Handle System is part of a Digital Object Architecture which relates to digital objects in a computer science sense, as an identifiable item of structured information in digital form within a network-based computer environment. Any object in the more general sense (the ontology sense, the word "thing") may be represented as a digital object, so there is no inconsistency in this use in the DOI System.

The indecs Content Model is the basis of the DOI System's approach to assigning metadata to define a referent and its relationships. This approach places importance on:

• unique identification;
• functional granularity;
• appropriate access;
• designated authority; and
• independence of specific business model or legal framework.

The International DOI Foundation (IDF) oversees the integration of these technologies and operation of the system through a technical and social infrastructure. The social infrastructure of a federation of independent registration agencies offering DOI services was modelled on existing successful federated deployments of identifiers such as GS1 and ISBN.

Comparison with other identifier schemes

A DOI name differs from commonly used Internet pointers to material such as the URL, because it identifies an object as a first-class entity, not simply the place where the object is located. A DOI name also differs from identifiers such as the ISBN, ISRC, etc. because it can be associated with defined services and is immediately actionable on a network.

The comparison of persistent identifier approaches is difficult because they are not all doing the same thing. Imprecisely referring to a set of schemes as "identifiers" doesn't mean that they can be compared easily. Similarly, when any two technologies (e.g., two web browsers) are compared, the criteria used for comparison must be defined.

The DOI System offers persistent, semantically interoperable resolution to related current data, and is best suited to material that will be used in services outside the direct control of the issuing assigner (e.g., public citation, or managing content of value). It uses a managed registry (providing social and technical infrastructure). It does not assume any specific business model for the provision of identifiers or services, and enables other existing services to link to it in defined ways.

Other "identifier systems" may be enabling technologies with low barriers to entry, providing an easy to use labelling mechanism where anyone can set up a new instance (examples include PURL, URLs, GUIDS, etc.) but which may lack some of the functionality of a registry-controlled scheme and usually lack accompanying metadata in a controlled scheme. The DOI System does not have this approach and should not be compared directly to such identifier schemes. Various applications using such enabling technologies with added features have been devised which meet some of the features offered by the DOI System (e.g. ARK) for specific sectors.

A DOI name is not dependent on the object's location and, in this way, is similar to a Uniform Resource Name (URN) or Persistent Uniform Resource Locator (PURL) but differs from an ordinary Uniform Resource Locator (URL). URLs are often used as substitute identifiers for documents on the Internet (better characterised as URIs) although the same document at two different locations has two URLs. Persistent identifiers such as DOI names identify objects as first class entities: two instances of the same object would have the same DOI name.

Structure of DOI name (identifier string)

A DOI name consists of a unique character string (case-insensitive, legal graphic characters of Unicode not in practice using certain characters such as pointed brackets "<>") divided into two parts: a prefix and a suffix.

An example of a complete DOI name is:

10.1000/182

where:

10.1000 is the prefix:

10 is the directory code. All DOI names start with "10.". This distinguishes a DOI name from any other implementation of the Handle System.

1000 is the registrant's code (colloquially publisher ID, although it may represent a publishers imprint, one journal, or a whole organization) identifying the registrant. In this DOI name, the number "1000" identifies the International DOI Foundation.

182 is the suffix, or item ID, identifying the single object. For this DOI name, the object corresponding to doi:10.1000/182 is the latest version of the DOI Handbook. (Typical suffixes are longer than this example, e.g., hdy.2009.9 or j.1365-313X.2008.03660.x).

The prefix is assigned by a DOI Registration Agency to a specific registrant. The suffix is assigned by the registrant and must be unique within a prefix. It can integrate existing standard identifiers such as an ISBN or ISSN, or SICI. An example of an application integrating the ISBN with DOI was launched in 2009 .

The DOI is considered an "opaque string": nothing can be inferred from the number with respect to its use in the DOI System.

Citations using DOI names should be printed as doi:10.1000/182. When the citation is a hypertext link, it is recommended to embed the link as a http proxy expression by appending http://dx.doi.org/ to the DOI name beginning 10. (e.g. the text doi:10.1000/182 is linked as http://dx.doi.org/10.1000/182).

Resolution

DOI name resolution is provided through the Handle System, developed by Corporation for National Research Initiatives, and is freely available to any user encountering a DOI name. Resolution redirects the user from a DOI name to one or more pieces of typed data: URLs representing instances of the object, services such as e-mail, or one or more items of metadata. To the Handle System, a DOI name is a handle, and so has a set of values assigned to it and may be thought of as a record that consists of a group of fields. Each handle value must have a data type specified in its "<type>" field, that defines the syntax and semantics of its data.

To resolve a DOI name, it may be input to a DOI resolver (e.g., at www.doi.org) or may be represented as a http string by preceding the DOI name by the string

http://dx.doi.org/

For example, to resolve the DOI name 10.1000/182, enter the address: "http://dx.doi.org/10.1000/182". Web pages or other hypertext documents can include hypertext links in this form. Some browsers allow the direct resolution of a DOI (or other handles) with an add-on, e.g., Mozilla Handle/DOI Protocol Handler.

Metadata

Each DOI name is associated with a series of metadata. The extent of this metadata may be defined by an application profile; a small kernel of common data for all DOI names can be optionally extended with other relevant data, which may be public or restricted. The metadata can be existing data from another scheme, which can be mapped to a DOI Application Profile using a data dictionary based on the indecs Content Model.

Registrants may update metadata about their contents any time they wish (when some publication data changes, when the primary URL the DOI name resolves to is modified, etc.).

DOI assignment fees

Unlike non-standardized URL indexing services, which are generally free, there is usually a charge to assign a new DOI name, to cover the costs of providing and operating services. These fees are set independently by each individual Registration Agency. Internally, an administrative fee is paid by the RA to the IDF, to support the costs of developing and maintaining the system. The DOI system overall, through the IDF, operates on a not-for-profit cost-recovery basis.

See also

• Handle System
• Indecs Content Model
• Uniform Resource Identifier (URI)
• Persistent Uniform Resource Locator (PURL)
• Life Science Identifiers
• OAI
• Object identifier
• PubMed
• Bibcode
• Extensible Resource Identifier (XRI)
• Universally Unique Identifier (UUID)

External links

• The DOI System
• Factsheet: DOI System and Internet Identifier Specifications
• The Handle System