Libraries face incredible challenges to fulfill their missions of building and providing access to digital and print collections as well as an ever growing array of services to meet their clientele's information needs. So many libraries lack adequate financial resources to accomplish these goals and must continually make difficult decisions on how to allocate their funds and efforts. Technology forms part of the essential infrastructure upon which libraries rely on to carry out every aspect of their work. In the context of these challenges, the technology systems that support libraries must offer much more than traditional functions. They must also be populated with content components and be intrinsically designed to connect and communicate with a complex array of persons and organizations.
The traditional integrated library system might be seen as a mere starting point of what libraries ultimately need for technology infrastructure. An automation system itself, though offering practical functionality, seems but an empty shell until populated with content resources and connected into the fabric of the library's business infrastructure and its patron's personal and social networks. The bibliographic records providing a basic representation of its collection are a good starting point. Today's libraries benefit from a much richer approach to content and connections, which can be thought of in several categories.
E-Resource Knowledge Bases
Especially for academic libraries, e-resource knowledge bases are an essential content component needed for the management of collections increasingly dominated by electronic materials. The universe of scholarly and professional resources from which libraries select materials for their patrons via subscriptions or open access is vast and dynamic. Rather than requiring each library to track the titles and coverage available within each resource package offered, knowledge bases have been assembled in support of activities such as electronic resource management, OpenURL-based link resolvers, alphabetical lists of journal titles, and other finding aids. These knowledge bases enable portfolio-based resource management, where selecting a content package propagates the details of the titles and coverage of the materials it contains. To be effective, a knowledge base must accurately represent the totality of content packages offered so that libraries can activate the packages to which they subscribe. Naturally such comprehensiveness is but an ideal. Libraries and knowledge base providers must cooperate to fill in gaps and repair errors.
Assembling a knowledge base requires substantial resources, including the technical platform upon which it resides, the technical processes to populate it with data provided by publishers, personnel to manage its quality control, and business processes to negotiate with publishers to contribute title and holdings data. It is not surprising that only a handful of comprehensive e-resource knowledge bases have been created and that their producers are large and well-resourced organizations: ProQuest, Ex Libris, TDNet, EBSCO, and OCLC. Open access knowledge bases have been envisioned since the early days of electronic resource management, though none have approached the completeness of the commercial offerings. The Global Open Knowledgebase (GOKb) launched in recent years shows promise. New standards such as Knowledge Base And Related Tools (KBART) have significantly lowered the difficulty in compiling knowledge bases in more automated ways from title and coverage lists provided from publishers.
Library users bring along their expectations gained from Google and other search engines when they use the tools and interfaces provided by the library. The genre of web-scale or index-based discovery services aims to fulfill this expectation through a central index populated with the individual content items represented in the library's physical and electronic collections. While the traditional online catalog would list the titles of the periodical to which the library subscribes, these index-based discovery services take a giant step forward by addressing the individual articles or chapters. In many cases, these indexes are populated with the full text of the materials in addition to the structured citation terms.
It is important to understand the distinction between e-resource knowledge bases and discovery indexes. Knowledge bases address journal titles and the dates of coverage and may include a few million resource records. Discovery indexes are populated with articles and book chapters, and therefore the number of content items may surpass multiple billions. Discovery indexes ideally represent the totality of scholarly and professional literature. As libraries implement an index-based discovery service they may use their e-resource knowledge base to configure their implementation to address only the materials available via their subscriptions and open access selections.
Not only must these index-based discovery services address an incredibly vast scope of content, but they must also offer a user experience that meets the expectation of increasingly web-savvy users and provide convenient access to content items. Library patrons naturally prefer immediate access for reading or downloading, though patrons may need to visit the library for physical materials.
Even more than e-resource knowledge bases, index-based discovery services involve massive resources to create and maintain. The acquisition and maintenance of the content addressed in their central index requires highly scalable technology infrastructure and state-of-the-art search and retrieval capabilities and user experience design. Again, only a small number of organizations have the capacity to produce these index-based discovery services: ProQuest, EBSCO, and OCLC. No comprehensive open access discovery index seems in sight.
Competitive and Comparative Intelligence
Librarians need lots of relevant data in order to make well-informed resource allocation decisions. Flat collection budgets and unrelenting subscription costs mean continually having to make difficult decisions regarding what items to cancel or renew and what new resources can be acquired. Traditionally such decision making has been informed via reports generated within the scope of the library's own automation system. It is naturally helpful to understand usage data, budget allocations, and other local data. Libraries benefit further through an expanded universe of data, such as the ability to compare their collection profiles with peer institutions, to have access to impact metrics of journals, and to have access to other relevant data sources.
It's likewise important to have strong analytical tools. As the volume of data increases, the ability to incorporate it in resource decisions becomes difficult without tools to visualize the data and make analytical comparisons. The provision of content and analytical tools for decision support is a component of technical infrastructure of increasing strategic importance.
Libraries have little time for redundant data entry. To the fullest extent possible, library systems must dynamically communicate with all relevant internal and external business partners. Patron records should be automatically populated and synchronized with student records and personnel systems of the parent institution. Financial transactions should be securely transferred to ERP or other financial systems. Orders, invoices, and other procurement transactions should be electronically exchanged with suppliers. Resource fulfillment requests should be disseminated to interlibrary loan or document delivery providers.
These are just some examples of the array of business transactions needed in a library's technical infrastructure so that it can function with the least manual intervention possible. In the current phase of library systems, many of these activities take place through batch transfers of data files. As the API ecosystem matures, we can anticipate that these will increasingly become more transactional and dynamic.
Connections with Content Providers
The ability to seamlessly incorporate access to content provided by external providers ranks as an essential capability. Public libraries offering e-book and audiobook lending services typically take advantage of external providers, such as Overdrive, Bibliotheca, or Baker & Taylor. These providers offer a catalog of titles licensed from a wide array of publishers and have created platforms to deliver digital content on behalf of a library.
In the early phase of e-book lending, libraries would essentially jettison their patrons to the platform of their content provider to search, select, and download titles of interest. In this current phase, libraries increasingly demand that their e-book lending services take place within their own virtual environment, maintaining their branding and the attention of their patrons.
This more library-centered paradigm of digital lending requires a sophisticated infrastructure. Instead of a simple hand-off, library catalogs and discovery interfaces need to engage in a complex set of interactions with the content provider platform to incorporate its titles in search results, present descriptive information, and ultimately check out and download a digital item. Supporting technologies include harvesting metadata into the library's discovery environment and a suite of APIs to manage the presentation and streaming or downloading of content.
Important progress has been made in the integration of e-book lending services into a library's virtual presence. Many of the developers of library catalogs have worked out partnerships with the major digital lending providers to implement mutual APIs to conduct these transactions, though some gaps remain in the overall experience. The process remains technically challenging with different APIs or other technical mechanisms needed for each combination of discovery interface and digital content provider. The complexity and other shortcomings of the current state of integration technologies drive the need for a more consistent and robust approach. The recently formed NISO workgroup on API framework for e-content integration, covered in this issue of Smart Libraries Newsletter, promises to make important progress in this area.
Connections with Patrons
Ultimately, library content and services must flow to the patron. It is important to present the virtual presence of the library in a way that meets expectation of user experience and to deliver content in forms that can be conveniently accessed, stored, and manipulated. Patrons, of course, don't come with predictable technical infrastructure as do the other entities with which libraries interact. The means to deliver content in more meaningful ways will always be highly variable. One of the key challenges for libraries lies in improving the ways that we deliver information resources to mobile devices and to social and professional networks.
It is easy to think of library systems in terms of checklists of functionality that they can perform. Such an approach was adequate in times where libraries were more isolated and worked mostly with collections they owned, which were housed in their own physical facilities. Times have changed dramatically. Library collections are increasingly digital and delivered through a long roster of external providers. Library collections can no longer be managed through the brute force of one-item-at-a-time procedures, but must rely on processes which work at scale, powered by knowledge bases and pre-populated indexes. Libraries today require not only appropriate features and functionality in their technology environment, but an ecosystem of content components and connections into the broad landscape of their partners and providers.