We can see a definite trend of consolidation among the organizations that create technology products for libraries. Through a long series of mergers and acquisitions, the industry has gradually transformed from being fractured in a way that pits relatively small companies against each other with overlapping products and limited revenue opportunities to one that is highly consolidated with a smaller number of large companies dominating the scene. This consolidation has resulted in a fewer number of choices for libraries, but the economic realities seem to press toward gaining efficiencies gained through combining organizations. This pattern isn't seen only among library vendors, it's a global phenomenon.
The cooperative organizations that bring groups of libraries together for automation and other services show the same pattern. In this issue of Smart Libraries Newsletter, we discuss several examples of library organizations that have merged together and in some cases are also consolidating multiple library automation implementations into a single large-scale system. I expect to see this strategy repeated many times over in the coming years.
From the earliest times of computerization of library operations, sharing systems has been a routine strategy for lowering costs and improving services. Municipal and county-wide library organizations naturally share a system among their many separate facilities. Multiple libraries organizationally independent of each other within a geographic region often form consortia for the purpose of sharing a library automation system or other services. By banding together, libraries not only operate more efficiently by spreading out the costs of hardware, software, and technical support, but they also improve services to patrons by offering access to more materials as represented in these aggregate collections. Resource sharing opportunities drive library consolidation just as much as operational efficiency.
Previously, the initial configuration of consortia and shared automation systems was set, at least to some extent by the limitations of what a given ILS implementation might reasonably support. Few library consortia, for example, were assembled that included more than a couple of hundred library facilities.
The hardware and operating systems of those times had limitations in processing power, the number of transactions that could be processed per second, in storage capacity; and the bandwidth needed to connect to the libraries was finite. Software applications, such as integrated library systems, also may not have been designed to work with the complexities of many hundreds of libraries with different policies and practices.
In recent years, however, many factors have come into play that overcome these limitations and stand to reshape the landscape for library cooperation and resource sharing. From an organizational perspective, difficult economic times have caused some library organization to find strategies to operate more efficiently, which include amalgamation. Examples include the Illinois Heartland Library System and the Reaching Across Illinois Library System both formed from the consolidation of multiple regional library systems. Likewise in Massachusetts, all the previously established regional libraries have been discontinued in favor of the Massachusetts Library System. Mergers can also cross governmental jurisdictions. Almost a decade ago the Orbis Cascade Alliance, spanning Oregon, Washington, and Idaho, was formed from the merger of two previously separate consortia of academic libraries. As in the business world, rolling up many separate library organizations into one eliminates redundancies in administrative and operational areas.
Changes in technology have also enabled ever larger groups of libraries to share an automation system. New computer architectures that support software as a service and other cloud-based technologies enable implementations of an application of almost limitless scale. Through clustering and related technologies, very large numbers of servers can work together to create very large-scale systems with extremely high reliability. Major Web destinations such as Google and Amazon.com rely on many tens of thousands of servers distributed globally.
Considering new-generation library services platforms, I anticipate that we will see very large numbers of libraries come together to participate in offerings such as WorldShare Management Service from OCLC. Alma from Ex Libris, or Intota from Serials Solutions, all designed for a multi-tenant software as a service. These products have been created using many of the same technologies that power the popular Web sites sustaining incredibly high levels of use.
Even automation products based on older architectures can be expected to perform well in ever larger implementations. Fast-multi-core processors, abundant memory, and practically unlimited disk storage enable inefficient applications to achieve remarkable performance and scalability. OCLC takes the concept of libraries' sharing of technology infrastructure to its logical conclusion. A slide that I've seen at many OCLC presentations, using figures developed by Mike Teets and Tam Dalrymple, estimates that the 1.2 billion libraries worldwide would generate an aggregate transaction load for both public and staff functions of about 18.9 million per day or 5,265 per second.
Such a transaction rate would be easily supported on a well-architected technology platform. Whether OCLC's WorldShare Platform will attract enough libraries to put that calculus to the test remains to be seen. We can reasonably observe, however, that technology platforms can be assembled to support groupings of libraries well beyond even the largest regional or state-wide consortia.
This trend of libraries banding together in ever larger numbers for shared automation projects will have an impact on the industry. The market will increasingly favor the organizations able to develop and sustain technology products that support large-scale implementations. This dynamic applies to both open source and proprietary projects. Most library automation products enter the scene with relatively small installations and over time work their way incrementally into ever larger implementations. Few organizations seeking an automation system for a very large number of libraries would opt for a system lacking a well-established track record of large-scale projects. Such a trend favors the established vendors in the industry and presents difficulties for new entrants.
A pattern of consolidation among existing automation implementations likewise disrupts the industry a bit. We see many examples of libraries that are basically satisfied with their incumbent system migrating to new ones as they join a merged system. In this issue we see examples of libraries that would have likely have stayed with their current system for many years but for the efficiencies gained through consolidating. As I track the losses of library customers among the organizations in the industry, an increasing number of these can be attributed to organizational consolidation. I expect this pattern to increase over the next few years.
A shift in the library automation industry toward a smaller number of larger installations also has implications for the potential revenues of the companies involved. Larger implementations means opportunities to win very large contracts, but the revenue per library represented may decline a bit. Consortia and other organizations involved in procuring these platforms naturally expect to see lower unit costs as they negotiate for larger groups of libraries. At the same time, I expect overall revenues in the industry to continue to see incremental increases as factors such as software as a service transfer costs previously borne locally by the library into new revenues for the vendors providing the hosting services.
Overall, I think that the consolidation of libraries into larger automation and discovery environments is a positive development. It's a reasonable strategy in response to weakened budgets, allowing libraries to shift more resources away from low-level tasks such as hardware and software maintenance toward highervalue activities related to patron services. Larger systems also mean more resources available to patrons and lower costs for fulfilling requests when desired items are not in the local library.