The advancement of technology enables ever larger groups of libraries to participate in shared systems that manage their collections and enable access to their communities. Year-by-year hardware components become faster, smaller, with ever more computing power or storage capacities. Modern services-oriented architecture and clustering technologies enable the creation and global deployment of massive platforms able to serve millions of users. At least at the infrastructure level, computer applications that scale almost infinitely can be built. This characteristic of large-scale computing brings important implications into the realm of libraries. What is the right level of cooperation or consolidation relative to how libraries implement their automation systems? Does it benefit libraries to share systems at the largest possible scale, or do individual or modest-scale implementations hold advantages?
The model of an individual library deploying its own automation system falls at one end of a resource sharing spectrum. Such a system provides the library complete control and autonomy. It can configure and brand the system according to its own interests and preferences. Many libraries that have very large and complex collections appreciate the flexibility and control that comes with managing their own system. Some of the downsides of an independent automation system include higher costs and the increased need for technical personnel. From the perspective of library users, this approach emphasizes the resources owned directly by the library, though other mechanisms may be provided for discovering and requesting materials from other institutions.
Trade-offs Between Independent and Shared Systems
Most of the barriers to libraries cooperating to share automation infrastructure lie more in strategy, politics, and personalities than in technology. It is not a given that it is to the benefit of any given library to participate in a shared automation system. For many libraries, the control and functionality needed to meet its operational or strategic requirements may be best fulfilled through an independent, self-sufficient system. Value propositions vary. In some cases the benefits and cost savings weigh in favor of joining a shared system, while in other scenarios, the benefits of local control outweigh any cost savings or operational efficiencies.
Participation in a shared system requires confidence in its governance, operational stability, and many other factors. Libraries must be sure that a shared system will be managed and developed in a manner consistent with its needs and that each participant will have a voice in decision making. For a library to place its essential operations in the hands of an external organization requires an extremely high level of confidence in its integrity and responsiveness.
Shared automation systems must naturally offer the functionality that meets the needs of the participating libraries. Ideally, participating in a shared system should give the library access to a higher quality system than it would be able to afford on its own. The functionality also needs to be oriented to the types of libraries involved. Joining in a consortium should not mean having to settle for a least-common-denominator level of functionality with low-end requirements, but rather should afford a more sophisticated system with more powerful capabilities.
Shared systems comprising different types of organizations, such as a mixture of public libraries and academic libraries, may find more tension regarding functional issues than those with a more homogeneous composition. While examples of successful multi-type consortia abound, they involve more compromises than those able to deploy or customize systems oriented to a specific type of library.
Libraries that participate in some level of shared automation environment lose a certain amount of control, but gain other benefits. Shared systems allow libraries to distribute the costs as well as the technical and administrative overhead of operating a system. The aggregate collections of each of the participating members expand the pool of resources available to patrons. While there may be some costs involved in routing materials among the members to satisfy patron requests, these arrangements provide a universe of resources available to patrons far beyond what any given library might be able to acquire on its own. Shared systems also provide opportunities for libraries to collaborate or consolidate selected operational activities. Participants in these collective systems often choose to perform various areas of processing centrally, such as cataloging or acquisitions. Though shared systems enable centralized operations, these activities can still be distributed among the individual libraries or any of a variety of hybrid arrangements.
Another challenge that arises when libraries participate in shared technology infrastructure lies in how they assert their identity or brand to their users. It is important to present visual reminders to library patrons that the resources are made possible through their local library. Even in with shared system, it is usually possible to configure it so to present a patron's local library's branding, logos, and style sheets as well as to favor local content over that which might need to be requested from another library. When possible, the domain name of the URL should also carry the identity of the library rather than that of the vendor that provides or hosts the system.
Large-scale Shared Systems: National or State-wide
Another end of the spectrum lies in shared technology platforms that support the libraries in expansive groups, such as an entire country or state. Projects at this level may not necessarily press the limits of technology scale, but they present enormous challenges in all of the organizational and political issues mentioned earlier.
So far, we have seen automation projects at the national level in a limited number of cases. This model seems to be an option for countries of a relatively small population and with a political climate that favors centralized public sector projects. Within larger countries, automation at the state or province level has so far been the zenith of cooperation.
These comprehensive projects to automate all of the libraries within a given country or state have some important differences, especially compared to consortia or regional systems with opt-in participation. Providing a shared system to all of the libraries of a country or a state can lead to a much more even level of service. When each library must implement its own system or buy into a consortium based on its local funding and political climate, wide variations in the quality of systems, the size of collections, or services may emerge. When automation is provided at the national or state level, small or remote communities can enjoy a library service equal to what is seen in the large cities or urban areas.
National-level automation systems have an inherent tendency to embrace cataloging standards and practices. In a decentralized environment, individual libraries may follow their own preferences in describing and processing materials. While such practice may meet some local need, it results in inconsistencies that can complicate resource sharing. Libraries that operate separate systems within a geographic area can agree to common standards and practices, but working in a shared environment usually comes with a mandate for consistency.
In countries as large as the United States, a nationally shared library automation infrastructure would not be feasible, politically or technically. While hardware platforms might scale to that level, none of the software environments available today would likely be able to manage that level of complexity. Within the United States, very few truly state-wide projects have been implemented for public libraries. Hawaii stands as the one state that has implemented a comprehensive shared system for public libraries. Provinces in Canada that have implemented shared systems include Saskatchewan, Prince Edward Island, and New Brunswick.
While the numbers of countries or states opting for comprehensively shared library infrastructure isn't growing dramatically, new examples arise every year or two. This issue of Smart Libraries Newsletter features the recent strategy launched by Ireland to implement a national system to support all of its public libraries. It seems that this project has the potential to provide a more powerful library service to the residents of the country while achieving savings in technology costs. It will be interesting to note whether additional large-scale projects move forward in the coming years and whether this model will gain wider acceptance.