Digital Object Identifiers and TheirUse in Libraries

Jue Wang

Available online 12 July 2007

Wang is AssistanCoordinator, WesKY 42101-1067, U

0098-7913/$–see frodoi:10.1016/j.serrev.

Digital Object Identifiers (DOIs) are persistent identifications that are assigned toany entity in order to manage intellectual content on digital networks. The DOIsystem has been managed by an open membership consortium, the InternationalDOI Foundation (IDF), founded in 1998. Usage of the DOI system is cross-industrybeyond academic publishing. This article provides an overview of what a DOI is; thestructure of DOI; DOI registration agencies (RA); metadata associated with DOIs,and advantages for libraries. Serials Review 2007; 33:161–164.Published by Elsevier Inc.

The Digital Object Identifier (DOI) initiative waslaunched in October 1997 at the Frankfurt Book Fairin order to develop a common mechanism to enableintellectual content management to be integrated withInternet technology. According to the International DOIFoundation (IDF), an organization established todevelop and manage the DOI System,

the Digital object identifier (DOI) is a system foridentifying content objects in the digital environment.DOIs are names assigned to any entity for use ondigital networks. They are used to provide currentinformation, including where they (or informationabout them) can be found on the Internet. Informationabout a digital object may change over time, includingwhere to find it, but its DOI will not change.1

Like the Universal Product Code (UPS) bar code usedon virtually every physical product, digital objects areidentified by DOIs. Since its introduction, the DOI hasprovided a stable, persistent link between content and adirectory on the Internet to which the content ownerwishes the DOI to point instead of a Web address orURL.The DOI is also used as an important emerging

international standard for identification of publishedmaterial online. Today many scientific journal pub-lishers are using the DOI to enable readers to moveseamlessly across the Internet from one location to

t Professor and Electronic Informationtern Kentucky University, Bowling GreenSA; e-mail: jue.wang@wku.edu.

nt matter. Published by Elsevier Inc.2007.05.006

161

another. Using DOI makes managing digital objects ina networked environment much easier and more con-venient for both publishers and their customers. Theusages of the DOIs are also very broad. Even thoughapplications of the DOI were first applied withinthe publishing industry, they extend well beyond thepublishing industry, especially for electronic commerceapplications.

The Structure of DOI

A DOI is an alphanumeric string or name that identifiesdigital content, such as an e-book, a journal article, or apiece of music. The DOI contains two components: theprefix and the suffix separated by a forward slash (“/”).The prefix, also known as the “Publisher ID,” isassigned by a DOI registration agency (the DirectoryManager) to the publisher. Organizations may choose tohave multiple prefixes for each of their imprints orproduct lines. All prefixes begin with “10” (thisdistinguishes a DOI from any other implementation ofthe Handle System®, see http://www.handle.net), fol-lowed by a period (“.”) and then a number designatingthe organization or publisher.

Example: Oxford University Press is 10.1093; Elsevieris 10.1016; and IEEE is 10.1109.

The suffix, also known as the “Item ID,” is assigned bythe publisher and can be made up of any alphanumericsequence of characters as long as each object can beuniquely identified. The length of the number can reach128 characters. An existing standard identification

Jue Wang Serials Review

system number, such as ISBN or ISSN, may beincorporated in a DOI by using it as the suffix.

Example incorporating ISBN into DOI with the ISBNis 0071381317 for the electronic version of the book21 Leaders for the 21st Century—How InnovativeLeaders Manage in the Digital Age by Fons Trompe-naars and Charles Hampden-Turner published byMcGraw-Hill. DOI is 10.1036/0071381317 (10.1036represents McGraw-Hill).

The DOI can also be applied at any level of “gra-nularity” or for any file types (such as text, image, oraudio-video). For a publisher, it can be applied to a wholebook and also to every chapter, every illustration,photograph, or table. In the case of music, it can identifya CD collection, as well as every individual track. As anexample, the DOI for Mona Lisa (painting) by Leonardoda Vinci is 10.1219/10223954 (10.1219: DOI for Corbis[a corporate site of creative images]). The suffix directorydatabase (the DOI repository) is maintained by thepublisher. Once a DOI is registered and assigned, it willnot be changed during its lifetime even if the ownershipand location of an object change. This makes DOIspermanent and persistent identifiers. When a digitalobject has been requested, a query will be sent to the DOIserver (a piece of software also called a link resolver). TheDOI server will find the record of theDOI and the addressof its associated object and send the location back to theuser’s browser and then show it on the user’s screen.

DOI can be part of an OpenURL to solve the problemof moving users from citations to the full-text electronicjournals that may available to them via dynamicallycreated links. By using this new technology, patrons willbe able to access the full-text journal articles or otherresources available in the library with just a few clicks.Seamless access to a library’s e-journal collectionsthrough this linking service has tremendously improvedlibrary service and the use of library resources.

The International DOI Foundation (IDF)

The authority administering the DOI system is the not-for-profit International DOI Foundation (IDF). TheFoundation was set up by the Association of AmericanPublishers, which joined forces with the InternationalPublishers Association and the International Associationof Scientific Technical and Medical Publishers. TheInternational DOI Foundation (IDF) supports the needsof the intellectual property community in the digitalenvironment by developing and promoting the DOIsystem as a common infrastructure for content manage-ment. The foundation will govern the use of DOIs andthe IDF is governed by its members through an electedboard. Members of the board represent a wide crosssection of organizations interested in the management ofintellectual property in the network environment. Theboard is responsible for all aspects of management of theDOI System, particularly policy formulation and stan-dards maintenance. Board members include representa-tives from Elsevier, John Wiley & Sons, Springer SBM,etc. For updated information on IDF and DOI system

162

management, please visit the DOI Web site at http://www.doi.org/. IDF administrative and metadata systemsare being put into place under the guidance of thefoundation as part of an ongoing development.2

The Central DOI Directory—Maintained byDOI Registration Agencies (RA)

The DOI directory serves as the middleman between userand the information content holder/publisher. Sincedigital content may change location or ownership overtime, a central DOI depository directory has been usedfor the DOI system and is maintained by DOI registra-tion agencies (RA). When the DOI system was firstlaunched, the IDF was the only registration agency. Dueto the expansion and development of DOI names, moreRAs have been franchised. RAs should be able to im-plement mechanisms for quality control of DOIs andmetadata registration and have the abilities to supportand promote multiple resolutions. RAs will assignprefixes to new registrants in accordance with IDF stan-dard terms. RAs will also make sure that DOIs underthis prefix are loaded with corresponding URLs intoa globally available resolution system nominated bythe IDF. RAs ensure appropriate minimal supportingmetadata for each DOI and usually charge the clients(registrants) based on prefix allocation, numbers of DOIsallocated, and/or numbers of DOIs resolved.Each RA has a geographical basis and each RAwill be

required to become a member of the IDF under the“registration agency” category of membership. The IDFBoard will be responsible for considering all applicationssubmitted by candidate registration agencies. RA mem-bership allows participation in the Registration AgencyWorking Group and access to all IDF materials andworking with IDF members in supporting and develop-ing the system. Current registration agencies includeCrossRef, Copyright Agency Limited, mEDRA, NielsenBookData, OPOCE (Office of Publications of theEuropean Union), R. R. Bowker, and TIB (GermanNational Library of Science and Technology). Moreinformation on application profile and area of coverageof the agencies can be found at http://www.doi.org/registration_agencies.html.

The Content Database—Maintained by thePublishers

The publisher maintains a database that contains theactual content of the DOI, which was designed tocontrol the distribution of its information. Theinformation maintained on the publisher side mayactually be distributed among many databases, but thepublisher must be able to present metadata about theactual content to the DOI resolution system in order tocommunicate the location of the content at thepublisher’s site. The publishers will take the responsi-bility of maintaining the content directory over timein that publishers restructure the files, move the files

Digital Object Identifiers and Their Use in LibrariesVolume 33, Number 3, 2007

to a new server, or even change the ownership of thecontent.When the DOI information is transferred to the

publisher, it will look up the DOIs from bibliographicdata, citation linking, and find the contents. Forsubscribed users, the full-text article or downloadabledocument will appear on the screen; otherwise, informa-tion on how to obtain the content or other related datamay appear.

DOIs and Metadata Standards

Metadata are structured information that describes,explains, locates, or otherwise makes it easier to retrieve,use, or manage the characteristics of an informationresource. It is a component of data which describes thedata or “data about data.” The metadata describe the“who, what, when, where, why, and how” about a dataset. It provides the essential link between the informationcreator and the information user. Each DOI is associatedwith a series of metadata, a set of bibliographical andcommercial information concerning the content (title,author, publication date, copyright, price, etc.).Answers for a number of basic questions about the

identified resource (kernel metadata elements, a mini-mum level of publicly available structured metadata)should be known by the RA at the time the DOI name isissued and is becoming mandatory for all DOIs that areregistered. Some of these data include an identifierassociated with the entity from a legacy identificationscheme, such as an ISBN (identifier), a name by whichthe entity is known (title), the primary type of intellectualproperty (type), the sensory mode such as visual/audio(mode) and primary agent, etc. Therefore, when thepublishers register DOIs, they also register kernel meta-data that are associated with the DOIs. In order to re-gister these kernel metadata, publishers must locate itwithin its internal system and they are responsible formaintaining these DOI associated metadata. DigitalObject Identifiers and their associated metadata are keyingredients in building an infrastructure to support across-publisher database of journal articles and citationsbased on a distributed production model.3 In order forthe DOI to fulfill its wider potential of providing thebasis for a full range of services relating to intellectualproperty in the network environment, metadata becomean essential component of the DOI System as a whole.4

Use of DOIs in Libraries

Publishers today are investing in linking technologyprovided by the CrossRef (one of the DOI RegistrationAgents (http://www.crossref.org/) to benefit those whosearch the journal literature using the DOI as an openstandard to make citation linking efficient and reliable.Publishers like ScienceDirect, Wiley InterScience, Black-well Synergy, SpringerLink, and others link their docu-ments via CrossRef. A demonstration of these linkingarticles can be found at http://www.crossref.org/

163

03libraries/18gallery.html. As the reference-linking net-work for scholarly and professional publishers, CrossReflinking enables the user to navigate the literature, movingfrom one article or idea to another across journals andpublishers.CrossRef is a full-scale implementation of theDOIwith

coverage of mainly scholarly and professional researchcontent, journal articles, books, conference proceedings,etc. According to Amy Brand, CrossRef’s director ofbusiness development, CrossRef is “a publisher member-ship association for collaborative technologies, andofficial DOI registration agency, and a cross-publishercitation-linking network.”5 To date, over 2,300 publish-ers and societies have participated in CrossRef’s colla-borative service, enhancing the value of millions ofarticles from thousands of journals. As of this writing,more than twenty-four million DOIs had been registeredin the CrossRef system since its inception in early 2000with more than 15,000 journals. The best news for thelibraries is that CrossRef access is FREE for libraries! Yes,there is no end-user charge associated with their use.Other benefits for libraries include increasing usage of

acquired resources, expanded access to content notowned, and enhanced localized linking. Even thoughthe total number of participating libraries is 1,216 asMay 3, 2007, the agency has much work ahead toconvince libraries to participate. In the United Statesalone, there were 9,129 public libraries in the fifty statesand the District of Columbia in fiscal year 2001.6

According to the National Center for Education Statis-tics (2005), there were 4,140 colleges and universities inthe US in 2003, and about 90 percent of them had theirown academic library, or about 3,700 academic libraries.Compared to the total libraries that could be part of thisdevelopment, the percentage of participating libraries isstill small.On October 30, 2006, CrossRef announced the launch

of its freely available Simple-Text Query service tofacilitate DOI look-up for researchers and publishers.Libraries would do well to take the advantage of theCrossRef services and provide more efficient linkingstrategies for their patrons. For example, a library canretrieve DOIs and metadata to make persistent links tofull-text works online and use DOIs for online reservesby embedding in reserve URLs. Scott Warren hasdemonstrated the construction of deeplinked e-reservesand provided examples of how such linking could takeplace (DOI: 10.1300/J124v22n04_01).7

CrossRef publishers support libraries not only byallowing patrons to move seamlessly through therelevant body of research but also by expanding accessto content which they do not own, thereby increasingusage of acquired resources and enhancing end-usersatisfaction with new levels of service. Libraries that joinCrossRef can be part of these exciting developments andadd value to their resources. For more information onhow to join CrossRef and understand the benefits ofmembership, visit http://www.crossref.org/03libraries/28benefits.html.The benefits for libraries include increasing usage of

acquired resources; expanded access to content not

Jue Wang Serials Review

owned; enhanced localized linking; and DOI andmetadata retrieval privileges at no cost.

DOIs: Benefits and Questions

The best feature of DOI is persistence. The DOIinterlinking across publisher’s content will transportseamlessly, with the content itself, to a local site, enablingusers to experience the same richly and contextuallyinterlinked content as if they were on the publisher’ssite.8 Perhaps the most immediate and exciting benefit toa publisher of adopting the DOI, though, is that bymaking the DOI available to end-users, the publisher caneffortlessly turn any pre-existing identifier into anpersistent, actionable identifier with an efficient, scal-able, Internet-based resolution and routing systembehind it.9 This makes easier for the publishers tomanage their data/contents and keep tracking the itemsin their system. By using DOI, even if the content changeshands between publishers and/or vendors, publishersand/or vendors only need to make one change in oneplace to make sure that all links point to the correctlocation.

DOI is an open system and anyone can build systemswhich integrate DOI into local environments. Forexample, a library can integrate DOI to access its locallylicensed materials. As mentioned before, the DOI systemis fully managed via registration agencies and can beapplied to any form of intellectual property at any levelof granularity. The system interoperates with other datafrom other sources. For the end users, information can beaccessed seamlessly cross the Internet.

Some issues related to the DOIs include intellectualproperty management. Who owns the copyright of thematerial? How does the system deal with manifesta-tions (the same content in another form, e.g., Word orPDF formats) and version (e.g., edited or translated)?Should different DOIs be assigned for each manifesta-tion and for each version? What metadata standardshould be used when assigning DOIs? Even thoughthere are some questions regarding DOI that requirefurther discussion, the DOI has been an extraordinaryventure for the publishing industry and their customersand the end users have benefited tremendously from itsimplementation.

Summary

The Digital Object Identifier (DOI) system not onlyprovides a unique identification for digital content, but

164

also provides a way to link users to the right materialholders via an automated digital environment. Thepersistent, unique, and reliable identifier, along with theresolution system, provides a perfect solution to managethe intellectual content in a digital environment. In thetransition from paper to electronic publishing, the DOIhas brought publishers and technologies together andhas overcome the significant obstacles of managingelectronic content. Libraries should consider this oppor-tunity to benefit their patrons and add value to electronicservices.Some scholars state that in the near future the value of

a publication will be determined by the number of linksto it. The widespread adoption of the DOI will not onlybenefit the scholarly community but also will enhancethe e-commerce market for digital asset management.Even though some issues are still surround the DOI, suchas copyright protection for DOI and metadata standardsvia assigning DOIs and missing links, libraries andagencies should continue to educate and market the useof the DOI and encourage the development of newapplications that allow the DOI to fit into the emergingWeb infrastructure.

Notes

1. See http://www.doi.org/.

2. See http://www.doi.org/sun_pap2.html.

3. “Discussion Paper: Role of the Digital Object Identifier in theJournal Portal Initiative and Its Ramifications/Requirements in aDistributed Environment” (http://www.hil.unb.ca/Texts/burk/sshrc_grant/Web/paper1.htm).

4. David Sidman, “The Digital Object Identifier: The Keystone forDigital Rights Management” (http://www.contentdirections.com/materials/SIIA-DOIandDRM-DavidSidman.htm).

5. Marla Misek, “CrossRef: Citation Linking Gets a Backbone,”Econtent 27, no. 6 (June 2004): 44–45.

6. Adrienne Chute and P. Elaine Kroe, “Public Libraries in the UnitedStates: Fiscal Year 2001,” Education Statistics Quarterly 5, no. 2(2003): 147–149.

7. Scott A. Warren, “DOIs and Deeplinked E-Reserves: InnovativeLinks for the Future,” Technical Services Quarterly 22, no. 4(2005): 1–17.

8. Hal Espo, “Using the Digital Object Identifier (DOI) for IntranetLinking” (http://www.intranetstoday.com/Articles/Default2.aspx?ArticleID=5501&IssueID=182).

9. David Sidman and Tom Davidson, “A Practical Guide toAutomating the Digital Supply Chain with the Digital ObjectIdentifier (DOI)” (http://www.contentdirections.com/materials/PRQ-CDIPracticalGuide.htm).