In the current phase of library technology, it is common for libraries to implement discovery systems in a manner loosely coupled with core automation systems. We've evolved away from a time where the online catalog module of the integrated library system could stand as the primary search tool, at least for those with large and varied collections. An increasing number of libraries have implemented discovery interfaces more loosely tied to the core ILS. The discovery interface may or may not be produced by the same company as the ILS, and an interesting set of dynamics has emerged among the organizations offering these products, with the library often carrying the burden of making sure that everything works together smoothly.
From a technology perspective, discovery interfaces interact with the ILS through a fairly well-established set of protocols. Data from the ILS are exported and re-indexed in the discovery system. Behindthe-scenes API's and other mechanisms allow the discovery interface to interrogate the ILS regarding current status and availability of any given item as needed during a user session. In most cases, it's also possible for patrons to place holds or make other requests through the discovery system that that are then reflected in their account on the ILS. In 2008, a workgroup of the Digital Library Federation developed a set of protocols and practices called the Integrated Library System – Discovery Interface, or ILS-DI, to help regularize the interactions between these two types of products.
The scenario of mixing and matching different ILS products and discovery interfaces has become quite common. Libraries can generally expect that any of the major discovery products will work with their current ILS. I've developed a resource that displays a matrix that shows which ILS products have been implemented with each of the major open source and commercial discovery interfaces (See: http:// www.librarytechnology.org/discovery.pl). This resource reveals some interesting patterns. In the cases where the producer of an ILS also offers a discovery interface, libraries tend to stay within the fold. Encore, for example, sees use mostly by libraries using Millennium, even though it has been designed to operate independently and has been used by a few libraries operating other ILS products. In some cases, the discovery product has not been promoted outside the company's own customer base: examples include LS2 PAC used only with The Library Corporation's own Carl.X and Library.Solution automation systems and Enterprise. To this point, these have been marketed only to those running SirsiDynix's Symphony or Horizon. Ex Libris' Primo has been adopted by many of its own Aleph and Voyager sites, but the company has also seen considerable success in placing Primo in the libraries of its competitors.
Discovery Systems implemented in conjunction with the most diverse set of ILS products are those produced as open source projects, such as VuFind and Blacklight, or those from companies that do not offer an automation system, such as BiblioCommons, or those that came into the library automation space after the discovery product was well established. The latter category would include AquaBrowser and Summon, successful as discovery products long before Serials Solutions announced its Web-scale Management Solution last June, or WorldCat Local, which was on the scene for many years before OCLC came out with its Web-scale Management Services.
As the next phase of library automation plays out, I expect to see a shift away from discovery systems decoupled from back-end automation systems back toward more tightly integrated product suites. Organizations offering both next-generation automation products and discovery services will obviously design those two types of products to work together optimally in a way that will be exceedingly difficult to accomplish with third-party products.
We can expect most libraries that implement Ex Libris' new generation Alma platform, for example, to use it in conjunction with the company's Primo and Primo Central discovery products. While Ex Libris emphasizes that they will offer all the API's needed to operate other discovery products with Alma, it will be an integration that libraries would have to accomplish on their own. But by using Primo and Alma together, the interoperability will be built-in and fully supported.
Likewise, OCLC's Web-scale Management Services will be designed specifically to use WorldCat Local as its end-user discovery interface. Libraries won't be locked into using both together, but the affinity will be quite strong. The likely result will be that only those with compelling reasons to use other discovery products would use something other than WorldCat Local with WMS.
As Serials Solutions launches its yet un-named “Web-scale management solution” it seems clear that the product will be designed to operate primarily with Summon as an end-user discovery interface and in conjunction with the company's 360 suite of products. Other combinations will be possible, but they might be going against the grain.
In this next business cycle, we see several different strategies in play, but with the common goal of prodding libraries toward bundled suites of products offered by a single provider. Ex Libris is working hard to place Primo in libraries beyond its own customer base of Voyager and Aleph sites. Such a placement is a success in itself, given that these discovery products represent significant sources of revenue. But it's also an opportunity to gain additional business in more strategic products in the future as those libraries will eventually need to move to one of the next-generation systems. Having a foot through the doorway can be a competitive advantage. Serials Solutions likewise seems positioned to leverage its success with Summon and its other products as it enters the nextgeneration library automation market. If it meets its goal of allowing libraries to unplug their legacy ILS as they implement its Web-scale management solution, they will then become that library's core technology provider with a tightly integrated bundle of products.
If these patterns do indeed prevail, over the course of the next decade we may well see at least a partial return to a scenario where a library receives most of its automation support from a single company in the form of a matched set or a bundled suite of products. Such an outcome will have both positive and negative aspects. Some might be concerned that libraries would be vulnerable to a single vendor or organization. On the other hand, dealing with a simplified number of technology suppliers simplifies the overhead involved with maintaining business relationships with multiple companies. When the same organization provides both the discovery and the automation environment, it bears more of the responsibility for integration and interoperability. I also anticipate that this more tightly bundled arrangement may result in better functionality. Even in the best of circumstances, connecting disparate components through the ILS-DI can be somewhat of an inefficient approach. Having automation systems and discovery environments specifically designed to work together will hopefully result in more refined functionality, especially in the area of patron services.
I do expect that some libraries will continue the mix-andmatch approach. Libraries have a strong affinity toward open source products, and at least some will continue to use interfaces such as VuFind and Blacklight with their current and even next-generation library automation systems. In the public library arena, companies such as BiblioCommons that offer compelling discovery products seem to steady on a course of working with many different existing automation systems, but not necessarily moving into the territory of offering one of their own.
This current era of decoupled discovery has been enormously important in the evolution of patron-facing interfaces and services. We've seen the clunky online catalogs replaced by modern Web-based interfaces that compare favorably to what patrons see elsewhere on the Web, offering more powerful search and retrieval capabilities, broader reach into the full scope of library collections, and more aesthetically appealing interface designs. The expansion of search in recent years through large consolidated indexes of scholarly content at the article level represents a major leap forward, especially for academic and research libraries. With these advancements well underway, it's now time for the back-end automation systems to catch up, moving forward from an orientation focused on physical materials to library services platforms that manage all the different formats of materials that comprise library collections. The legacy ILS products existed (at best) in a loosely coupled relationship with the new generation of discovery services. Once the new round of library services platforms becomes established, the circle can become completed, allowing libraries to benefit from a newly re-integrated suite of products, hopefully with the potential to more efficiently support their internal operations and Web-delivered services.