Services make the repository
This paper provides an overview of the collaboration between the Perseus Project and the Digital Collection and Archives (DCA) at Tufts University in moving the collections of the Perseus Project into the DCA's Fedora based repository as well as a listing of potential services necessary to support a successful institutional repository.
Digital repositories, Fedora, digital library services
Many assume, for better or for worse, that libraries and archives will be able to maintain bits over time, migrating them from one system to another, converting one image format into another, and executing the suite of complex processes necessary to maintain digital objects in well-established formats. While such an assumption does not do justice to the challenges of the task, the preservation of digital objects is a necessary, but by no means sufficient condition for successful curation. At the same time, as institutional repositories (IRs) have multiplied at academic institutions around the world, those engaged in building and managing the IRs have lamented the poor levels of contributor participation that many repositories have seen.
While the literature about institutional repositories and digital repositories in particular is quite extensive, it has often focused on the use of repositories as a place to manage faculty preprints rather than diverse research materials or object types. Beyond simple dissemination, there has been lesss discussion of what types of behaviors will become essential components of these repositories if they are to become truly successful. Research has focused on how to build a repository (Gibbons 2004), how to get users to contribute content (Fried-Foster and Gibbons 2005), how to validate and trust content (Gladney 2004), and how to preserve content (Carpenter 2005, Stanescu 2005).
A recent large scale review of IRs explored the various services they provided, and ultimately reported that:
"It is clear that while many experiments are being launched (most commonly federated searching, print-on-demand, or automatic replication of metadata and/or content from one system to another), these services are still highly experimental, and it is too early to tell which ones will become successful and widely adopted. It seems clear, however, that various forms of cross-repository searching (either through federated search or harvested union catalogs) will be needed, and there are many such services, at least in prototype.” (Westrienen and Lynch 2005)
While services such as federated searching and print-on demand are quite important, we believe that the discussion of repository services need to expand beyond these more basic ideas to encompass services such as advanced text analysis.
Services for repositories can be considered in two tiers:
1. high-level, or infrastructure, repository services that support sharing of data between repositories, ingesting of data into repositories, and harvesting of content from repositories. These services also include methods for exposing content as a set of standardized behaviors in order to facilitate the use of repository content by additional services in tier 2.
2. low-level, or content based, services such as natural language processing tools and analysis services that support users in their interaction with materials that are already within repositories such as data sets, images, texts, and other primary source materials.
This paper will discuss both service layers with a special focus on the importance of the second tier of services for IRs, and how the collaboration between the Perseus Project and the Digital Collections and Archives at Tufts is seeking to meet this need.
Launching an IR is not a small endeavor. Addressing the issues of trustworthiness, authenticity, and preservation take a tremendous investment of resources both in terms of money and staff. How can an institution justify such an investment in something that is little used? A successful IR is one that is relevant. It must meet the needs of the community it serves and be flexible enough to change as the community's needs change while preserving the digital assets deposited in it. Preservation and trustworthiness are undoubtedly the foundation of any repository, but a successful repository must do more to make itself, its collections, and its services relevant. 1 Tufts adopted Fedora in 2001 as the application with which to build its institutional repository because Fedora was designed from the start not only to manage data but also to facilitate the creation of behaviors that can act over digital objects in its collections and facilitate the interfacing of external services, by which users experience the data, with the repository. The Perseus Digital Library 2 and the Tufts Digital Repository Program began working with Fedora primarily because of these capabilities. The Tufts Digital Repository 3 thus represented from the start a partnership between the library infrastructure and academic projects such as Perseus.
The Tufts Digital Repository
The Tufts Fedora based IR is administered by the Digital Collections and Archives (DCA) and was developed in partnership with Tufts University Information Technology’s Academic Technology group. We adopted Fedora as our core digital depository because it is designed to support a service-oriented architecture, that is, the Fedora digital repository does not stand as a monolithic repository (Lagoze 2006). In our system architecture, Fedora provides the core repository services and infrastructure for collections of services (both consumer services and provider services) that communicate with each other to achieve any number of coordinated functions or activities. In a service-oriented architecture, a service is simply a function that is well-defined, self-contained, and does not depend on the context or state of other services in the overall architecture. In other words, in the Tufts University repository system, Fedora provides repository services and infrastructure to other services as well as a method of communication for interacting or coordinating with other services, repositories, or digital libraries both internal and external to the Tufts environment.
As we envisioned our IR, we began from the premise that a repository would need to be open, flexible, and able to deal with the wide variety of digital assets created at Tufts, be they electronic records, primary source research materials, faculty publications, or health science datasets. Openness and flexibility was an absolute must because Tufts already had a robust set of active-use digital libraries. We wanted our repository to complement these projects by providing a solid foundation for the preservation of digital assets over time and an expanding set of delivery tools to promote access. The partnership between DCA and the Perseus Digital Library is a realization of this vision.
The Perseus Digital Library
The Perseus Digital Library Project has curated a growing set of digital objects over twenty years and through multiple systems. For us, a successful repository would not only maintain well designed digital objects but also the behaviors that make these objects accessible and useful. Archives and libraries can define their role in the coming years and decades by assuming responsibility for increasingly powerful services built on top of their collections.
Since 1993, Tufts has been the home of the Perseus Digital Library (PDL), a growing collection with particular strengths in Greco-Roman antiquity and other cultural heritage areas. The PDL currently serves approximately 15,000,000 pages per month, each page representing, in effect, a dynamically generated report in response to a general query. For example, when we call up line 44 of book 23 in Homer's Iliad, we automatically aggregate and customize a view of all the information that we have about this canonical chunk, including multiple translations, editions, commentaries, etc. While we need to preserve the individual digital objects of the collections, dynamic behaviors that automatically tie digital objects together are what best characterize this digital library. Preserving Perseus thus implies not only preserving digital objects but also the services that go with them.
In the spring of 2006, the PDL began to move core digital objects, including original photography of museum objects and a library of TEI-compliant, interoperable, and open access primary texts into the Tufts digital repository. 4 Simply migrating these objects into Fedora is a major step forward for our digital library. The ultimate goal is to have all 13 million words of carefully transcribed and marked up Greek and Latin source texts as permanent components of the Tufts Digital Repository. Unrestricted by third party rights agreements and standard in form, these texts will be able to circulate freely and to provide the basis for new editions in the future. Substantial support is at hand to augment this collection and we expect to have all major Latin texts from the classical period available by summer 2007.
Building Standard Services for IRs
Although there is a current gap between the behaviors in the Fedora Digital Repository System as currently instantiated at Tufts University and those now standard in the PDL, the DCA and the PDL are actively collaborating to ensure that the necessary services became core components of the Tufts Digital Repository. For Perseus and for many of its users, services such as citation linking, morphological analysis, and named entity recognition are foundational. Digital repositories that do not incorporate such services may successfully provide long-term, secondary archives but will never, we believe, become front-line service providers. Put another way, institutional repositories that do not provide the more complex services on which users depend, but only provide routine data migration services may succeed in their defined goals but have relatively little impact.
There has been a recent flurry of activity in defining high level standardized services for repositories in order to assure interoperability and greater access. A concurrent trend has been the growing interest in Web 2.0 and now Library 2.0, and the argument that the services that users expect from the Web also hold potential for the future of libraries as well as IRs. 5 This need for interoperability and standardized services between digital repositories was the subject of a recent meeting sponsored by Microsoft, the Mellon Foundation, the Coalition for Networked Information (CNI), the Digital Library Federation (DLF) and the Joint Information Systems Committee (JISC). This meeting included participants from a variety of interested parties in the digital repository arena, including representatives from major universities such as Cornell, MIT and Stanford, commercial publishers such as Nature Publishing Group, and a variety of commercial interests. The participants agreed on a working definition of a repository as a “networked system that provides services pertaining to a collection of digital objects” (Bekaert 2006). This meeting concluded that a “well-behaved digital repository” must at the least, support three basic service interfaces: an obtain interface, a harvest interface, and a put interface. At the same time, the group decided that there is also a great need for “richer cross-repository services” that will allow for “the transfer of digital objects from digital repositories to parties that provide value-added services over the digital objects” to support users in their desires to be able to repurpose and make more advanced use of digital content.
Building Advanced Services for IRS and the Asset Actions Model
More advanced, second tier, services will require methods for accessing digital repository content or particular views of digital library content in order to process or manipulate that content. To this end, the Tufts Digital Repository Program continues to work with the University of Virginia and Northwestern University on developing Asset Action sets for digital repository content. Asset Actions provided layered access to a given digital object (Chavez, et al. 2006). They provide:
1. a very basic set of behaviors that can be addressed in exactly the same way, regardless of the type of object
2. a set of behaviors that are true for all members of their particular class, for example, basic image behaviors for image objects, basic text behaviors for text objects.
3. other behaviors that are intended to support more advanced content-based services, including behaviors that are needed for management functions
Asset Actions are a core part of our digital repository infrastructure primarily because they provide a means for disseminating a functional view of a unit of content that could be shared with other services and they also provide a way to generalize behaviors across repositories. This level of functionality enables manipulation of digital content in a wide range of additional services both internal and external to the digital repository supplying the content. The DLF-Aquifer Asset Actions Experiment provides a good illustration of how image objects from four different digital repositories (University of Virginia, Indiana University, Northwestern University, Tufts University) could be utilized by a search tool, an annotation tool, a digital object collector tool, and a ‘book bagging’ tool all by means of common Asset Action behaviors.
It is our hope that Asset Actions and similar methods for “selectively exposing content as a set of standardized behaviors or views” (Chavez, et al. 2006) will facilitate “deep sharing”, i.e. the creation and use of new content-based services demanded or expected by a given digital repository’s customer base, such as those discussed below.
Indeed, the rising expectations of users will become an important determinant in the ultimate success of any IR. A recent article in D-Lib Magazine explored the importance of a service framework for libraries that follows a SOA (service oriented architecture). They define a service as “a discrete piece of functionality, manifested in the form of a technical implementation, and deployed for use, usually on a network (e.g., as a Web service)” (Lavoie, et. al 2006). They believe that, in the future, more and more library processes will take the form of automated systems that will be built through a combination of services, which will mean that libraries and their repositories will need to make their services available where users need them, whether it is on their personal desktop, course management system, or the general Web.
Lorcan Dempsey has also recently explored how library services will need to change to remain relevant in today’s world. He believes that libraries, including repositories, will need to move toward “providing services which allow users and the library to create, analyse, organise, select, acquire, store, and preserve resources” (Dempsey 2006). Such services, he argues, will fit into different user workflows, including the support of learning, remixing of data, curating large data sets, metadata creation and repository services. For an example of repository services, he proposes a service which collects and exposes metadata from institutional repositories, thus likely increasing faculty desire to contribute.
For example, the University of Georgia has developed an IR that provides value added services in order to encourage faculty contributions, and reuse of the content in new environments (Walters 2006). Recent work by the OCKHAM project in developing digital library services such as the OCKHAM alerting service, MyLibrary@OCKHAM and a digital library service registry also serve as important examples of what can be done with basic open source protocols and tools (Frumkin 2006, Xiang and Morgan 2006). Similarly the JISC in the United Kingdom has recognized the need for a service registry so that different machine processes can access the different repository or library services that they might need (Apps 2006). In other recent work, a number of potential web services for digital libraries and repositories have been proposed (Fox 2006) , such as for name authority services (Hickey 2005), access to controlled vocabularies (Vizine Goetz 2006), personal name identification (Xia 2005), and information retrieval services (Fu and Mostafa 2004). These are just a few examples of how the exploration of different services for digital libraries and repositories has grown in the last few years.
If digital repositories do not address the need for digital services, then they, and university libraries with them, may find some of their functions progressively shifting to a collaboration of large scale entities (e.g., the Open Content Alliance/Google Library) and various discipline-based organizations. As these different mass digitization projects pick up steam, there has been a growing discussion of what will need to be done to make the massive amounts of content more useful (Crane 2006). Many of the tools and services suggested to make these materials more useful could be of great use as value added services within a dynamic digital repository. A recent symposium at the University of Michigan discussed the future role of libraries in the face of mass digitization. One observer of the symposium noted that “copious digital textual content can provide the raw materials for an array of Web services and technological innovations limited only by the imagination” (Jones 2006). Services mentioned included linkages, citations, digital indexing, collaborative annotation, and mash-ups with services such as Google Maps.
Nonetheless, another observer at the University of Michigan symposium feared that libraries were perhaps missing the point when it came to services. Eric Lease Morgan, Head of the Digital Access and Information Architecture Department, University Libraries of Notre Dame, believes that the Google library project offers an excellent opportunity for libraries to rethink the types of services they provide and to go beyond just offering content. He believes that once users can carry around entire collections on digital devices, the role of libraries and librarians will not be about content but about services against that collection, through the creation of tools for researchers, students and casual readers. Morgan suggests a variety of potential services including word lookups in reference tools, comparing and contrasting definitions, annotating work with marginalia, looking at others marginalia, identifying major themes, and finding themes in other works As Morgan so aptly puts it “collections without services are useless, and services without collections are empty.” (Morgan 2006).
In a recent article, Bearman and Trant, consider the need to do more than simply store scanned pages of books. They believe that comprehensive digitization will create the raw materials needed for a “host of value-added services” as long as the semantic knowledge of these sources is exploited in order to create a universal knowledge base and as long as tools are available in order to explore it. They offer an intriguing view of the possibilities:
"Post-processing against the whole knowledge-base could reveal the historical meanings of words and ideas in older literature to contemporary readers, in foreign literatures to their non-native readers, and in specialized literatures to multi-disciplinary readers. Buildings, cities, people, boats and other subjects with proper names could be linked to data and images that depict them at the appropriate time. Allusions that authors make to earlier texts, and references made to them by later texts, can be linked, as is now done only in the most complete of scholarly editions" (Bearman and Trant 2005).
Bearman and Trant argue that there will be a growing need for services such as word look ups, named entity recognition, and citation analysis. These will become essential tools for repositories to offer to their users.
Further calls for the development of tools and services can be found in a recent report from the Commission on Cyberinfrastructure for the Humanities and Social Sciences. The report calls for the development of extensive and reusable digital collections with participation from scholars, archivists and librarians and argues that high priority must be given to building both tools and collections:
"Scholars in the humanities and social sciences should work with librarians, curators, publishers, and technologists to develop tools for producing, searching, analyzing, vetting, and representing knowledge, as well as standards for documenting data of all kinds. For hundreds of years, the most important tools of humanists and social scientists were pen or brush and paper. Now, scholars require a range of digital tools for research, teaching, and writing, including tools for finding, filtering and reviewing, processing and organizing, annotating, analyzing, and visualizing digital information." (ACLS 2006).
As this report indicates, both collections and tools will be essential to building a cyberinfrastructure for the humanities.
A recent report from the JISC in the UK has also stressed the importance of developing terminology services in the developing educational information environment, and explores their potential for providing better access to the content of repositories. The phrase "terminology services" is used to encompass a wide range of services from automated indexing and classification, named entity services, and text mining, to name only a few. The authors argue that repositories must not become “isolated silos of content”, but should instead offer services that make them an essential part of the technology infrastructure of scholarship (Tudhope 2006).
In addition to Fedora, other digital repository and library systems are exploring how to best meet the needs of both digital preservation of content and providing sophisticated services for access to that content. The Cheshire digital library framework being developed by the University of California Berkeley and the University of Liverpool seeks to meets this need (Watry and Larsen 2006). This group is in the process of developing a high-level framework that will both support large scale preservation and support for semantic services and natural language processing tools such as text and data mining, ontological searching, and named entity recognition. Their research focuses on the important need for “integrating digital library and persistent archive systems into an existing framework.”
Potential Services for Institutional Repositories
In order for an IR to successfully meet the needs of its various user communities, it should provide not only preservation and access to digital objects but also a range of services that make these objects useful. Table One provides a list of potential services that repositories could implement. Successful repositories will often need to feature advanced browsing and searching, such as browsing of named entities, supporting searching for different types of entities (e.g. personal names, place names, gene names, chemical compounds); and visualization services, such as automatic generation of maps and timelines. For example, repositories could provide a gazetteer lookup service. Such a service could allow repository users to determine if a given document contains readily identifiable terms that lend themselves to rapid gazetteer lookup, such as technical terms, multi-word organizational names or place names.
Table 1. Potential Services for Repositories
|Automatic Document Alignment|
|Automatic Map and Timeline Generation|
|Dynamic/Ad-hoc Collection Building|
|Named Entity Identification|
|Personal Publication and Aggregation|
|Social Tagging and Bookmarking|
|Text Chunking and Alignment|
As repositories grow in size, it will also be important for automated systems to help users determine what glossaries, encyclopedias or other resources in the collection will best serve a user reading a particular document. Successful repositories should be able to associate new documents with the most useful resources. For example, a document on local politics in Boston in 1847 should be associated with contemporary directories of the city, biographies of prominent individuals alive at the time, and historical gazetteers from the mid-nineteenth century covering Massachusetts, the United States and the world.
Repositories should also support advanced text analysis features such as annotation services that track and identify passages that comment specifically upon arbitrary subsets of an object. Many digital objects that will eventually be placed in repositories can have complicated structures, so repositories should work towards supporting the display of sophisticated textual markup such as multiple text chunking schemes that capture overlapping hierarchies. Support can also be provided for scholars who wish to compare different editions of texts, with services that could include version analysis (e.g. if different editions are available, the system allows a user to analyze how much they differ or visualize differences between editions dynamically). Advanced indexing services can also be provided that can more effectively mine the content of repositories. This could include both manually generated indices and the results of automated processes such as named entity analysis. named entity analysis services could include automatic identification of named entities within documents (including confidence estimations in those identifications), and could also allow users to correct system errors, which could then be fed back into the system and be used to improve subsequent automatic analysis.
The Draft Audit Checklist for the Certification of Trusted Digital Repositories 6 published by the Research Libraries Group (RLG) and the National Archives and Records Administration (NARA) identifies many attributes necessary to establish trust, but trust is only one factor in determining a repository's success, albeit an important one. Focusing on infrastructure, policy, and procedure, the checklist does turn to usability in Section C in its discussion of designated communities and particularly C3, Use and Usability. The intention of these checklist criteria is to establish that one component of trust in a repository must be that the assets on deposit will be usable upon retrieval, though that usability is defined very broadly. C4 addresses understandability, or the repository's commitment to ensure that content information is understandable by designated communities even when a significant period of time has elapsed and the skills and tools of that community may have changed, through transformation and documentation.
All of these issues regarding trust and long-term usability are undoubtedly integral to the long term success of any digital repository and repository managers who ignore them do so at their peril. Meeting the Draft Audit Checklist criteria will be a significant challenge to most institutions, and it could seem sensible to focus on them exclusively, ignoring other areas of development that could be seen as distractions. However, at Tufts we have recognized that our viability as a repository program depends on forging strong partnerships with our stakeholders. Without their support we will not be able to continue to build our program to the point when we can, in fact, fulfill all of the checklist's requirements.
Partnerships are the foundation of our model. The Tufts DCA brings expertise in preservation, policies, and user services. Perseus brings extensive experience in developing innovative tools and pushing the boundaries of digital library technology for humanities collections. Other digital libraries at Tufts provide services supporting teaching in art history, integrated course management and curricular tools for health sciences, and more. These partnerships entail allowances, such as asset-level APIs for management and retrieval of assets, open transport protocols, and unified content models and asset typing, within the digital repository to facilitate use, reuse and interoperability with external applications and systems. While we could each forge ahead alone, we can strengthen all of our efforts by working together.
From our perspective, creating a successful repository will mean building one that is more than trusted; it must be relevant, flexible, and able to meet our diverse stakeholders' needs. For the digital repository we are building at Tufts, we believe that this marriage of preservation and services, data and tools, permanence and flexibility will make our repository not only viable, but vibrant, long into the future.
- The issue of what constitutes success for a digital repository is receiving an increasing amount of attention, for example see (Westell 2006)
- All of the TEI-XML files for our texts can be downloaded under a Creative Commons license.
- For an excellent discussion of the potential of Web 2.0 and services for libraries, please see (Van Veen 2006).
American Council of Learned Societies (ACLS). (2006). "Our Cultural Commonwealth: The report of the American Council of Learned Societies' Commission on Cyberinfrastructure for Humanities and Social Sciences." http://www.acls.org/cyberinfrastructure/OurCulturalCommonwealth.pdf
Apps, A. (2005). “A Middleware Registry for the Discovery of Collections and Services.” In Proceedings of NCeSS2005: The First International Conference on e-Social Science, (Manchester, UK, 22-24 June 2005), http://epub.mimas.ac.uk/papers/ncess2005/apps-ncess2005.html
Bearman, D. and Trant, J. (2005). “Converting Scanned Images of the Print History of the World to Knowledge: A Reference Model and Research Strategy.” RDLP, Vol. 8, No. 5, http://www.elbib.ru/index.phtml?page=elbib/eng/journal/2005/part5/BT
Bekaert, J. and Van de Sompel, H. (2006). "Augmenting Interoperability Across Scholarly Repositories." http://eprints.rclis.org/archive/00006924/
Carpenter, L. (2005). “Supporting Digital Preservation and Asset Management in Institutions.” Ariadne, 43, April, http://www.ariadne.ac.uk/issue43/carpenter/.
Chavez, R., Cole, T. W., Dunn, J., Foulonneau, M., Habing, T. G., Parod, W., and Staples, T. (2006). "DLF-Aquifer Asset Actions Experiment: Demonstrating Value of Actionable URLs." D-Lib Magazine, Vol. 12, No. 10, http://www.dlib.org/dlib/october06/cole/10cole.html
Crane, G. (2006). “What Do You Do With a Million Books?” D-Lib Magazine, Vol. 12, No. 3, http://www.dlib.org/dlib/march06/crane/03crane.html
Dempsey, L. (2006). “The (Digital) Library Environment: Ten Years After.” Ariadne, Issue 46, February, http://www.ariadne.ac.uk/issue46/dempsey/.
Fox, R. (2006). “Lingua Franca of Digital Libraries.” OCLC Systems & Services, Vol. 22, No. 1, 26-33.
Fried-Foster N., Gibbons, S. (2005). “Understanding Faculty to Improve Content Recruitment for Institutional Repositories.” D-Lib Magazine, Vol 11, No. 1, http://dlib.org/dlib/january05/foster/01foster.html.
Frumkin, J. (2006). “The Need for a Digital Library Service Registry." OCLC Systems & Services, Vol. 22, No. 1, 23-25.
Fu, Y. and Mostafa, J. (2004). “Toward Information Retrieval Web Services for Digital Libraries.” In Proceedings of the 4th ACM/IEEE-CS Joint Conference on Digital Libraries (New York, NY: ACM Press), pp. 370-371. http://portal.acm.org/citation.cfm?id=996350.996436
Gibbons, S. (2004). “Establishing an Institutional Repository.” Library Technology Report, Vol. 40, No. 4, 11-14.
Gladney, H. M. (2004). “Trustworthy 100-Year Digital Objects: Evidence after Every Witness is Dead.” ACM Transactions on Information. Systems, Vol. 22, No. 3, July, 406-36, http://portal.acm.org/citation.cfm?doid=1010614.1010617
Hickey, T. B. (2005). “Authority Web Services.” Presentation at ALA LITA-ALCTS CCS ACIG, June, 26, 2005, http://www.ala.org/ala/lita/litamembership/litaigs/authorityalcts/WebService2005Annual.ppt
Jones, E. (2006). “Pushing the Digitization Dialogue Forward: "Scholarship and Libraries in Transition" at the University of Michigan.” Journal of Electronic Publishing, Vol. 9, No. 2, Summer, http://www.hti.umich.edu/cgi/t/text/text-idx?c=jep;cc=jep;view=text;rgn=main;idno=3336451.0009.204
Lagoze, C., Payette, S., Shin, E. and Wilper, C. (2006). “Fedora: An Architecture for Complex Objects and their Relationships.” International Journal of Digital Libraries, Vol. 6, No. 2, 124–138, http://arxiv.org/pdf/cs.DL/0501012
Lavoie, B., Henry, G., Dempsey, L. (2006). “A Service Framework for Libraries.” D-Lib Magazine, Vol. 12, No. 7/8, July/August, http://www.dlib.org/dlib/july06/lavoie/07lavoie.html.
Morgan, E. L. (2006). “Mass Digitization.” http://www.infomotions.com/musings/mass-digitization/, Last updated March 18, 2006.
Stanescu, A. (2005). “Assessing the Durability of Formats in a Digital Preservation Environment: The INFORM methodology.” OCLC Systems & Services: International Digital Library Perspectives, Vol. 21, No.1, 61-81.
Tudhope, D., Koch, T., and Heery, R. (2006). “Terminology Services and Technology: JISC State of the Art Review.” September, http://www.ukoln.ac.uk/terminology/JISC-review2006.html
van Veen, T. (2006). “Serving Services in Web 2.0”. Ariadne, Issue 47, April, http://www.ariadne.ac.uk/issue47/vanveen/
Vizine-Goetz, D., Houghton, A. and Childress, E. (2006). “Web Services for Controlled Vocabularies.” ASIST Bulletin, June/July, http://www.asis.org/Bulletin/Jun-06/vizine-goetz_houghton_childress.html
Walters, T. O. (2006). “Institutional Repositories and the Need for "Value-added" Services.” Presentation made at the Coalition of Networked Information (CNI) Spring 2006 Task Force meeting. http://hdl.handle.net/1853/9302
Watry, P. and Larson, R. (2005). “Cheshire 3 Framework White Paper: Implementing Support for Digital Repositories in a Data Grid Environment.” In Local to Global Data Interoperability - Challenges and Technologies, (Los Alamitos, CA, IEEE Computer Society Press), pp. 60- 64, http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1612466
Westell, M. (2006). “Institutional Repositories: Proposed Indicators of Success.” Library Hi-Tech, Vol. 24, No. 2, 211-226.
Westrienen, G., Lynch, C. (2005). “Academic Institutional Repositories: Deployment Status in 13 Nations as of Mid 2005.” D-Lib Magazine, Vol. 11, No. 9, September, http://www.dlib.org/dlib/september05/westrienen/09westrienen.html
Xia, J. (2006). “Personal name Identification in the Practice of Digital Repositories.” Program, Vol. 40, No. 3, 256-67.
Xiang, X. and Morgan, E. L. (2005). “Exploiting “Light-weight” Protocols and Open Source Tools to Implement Digital Library Collections and Services.” D-Lib Magazine, Vol. 11, No. 10, October, http://www.dlib.org/dlib/october05/morgan/10morgan.html