Jonathan Rochkind has just completed some compelling research directly comparing the Application Programming Interfaces (APIs) of various discovery services. The study clearly illustrates the value, performance and ease-of-use of the Summon API, and it provides an excellent analysis of some of the key things to evaluate if one is considering integrating results from a discovery service such as the Summon service with a locally developed, custom interface.
However, the scope of his research only partially addresses the fact that library discovery services must do more than simply provide a single search box. They must provide a user experience that truly meets users’ expectations—developed from using open Web search engines—for relevance, speed, ease of navigation as well as the ability to support short and broad query terms. Discovery services must also adapt quickly to meet changing expectations or libraries will continue to be dis-intermediated from the research process. In short, user experience matters.
Recent studies into research behaviors of library users have all resulted in roughly the same conclusions. Users are often confused by the complexities of navigating library websites and resources and therefore bypass the library in favor of the convenience of starting their research with Google and other open Web search engines. (Law, 2009 [pdf]) As a result, the importance of the role of the library as a gateway for locating information has diminished over time and the library is increasingly dis-intermediated from the actual research process. (Housewright and Shonfeld, 2008) And, so to improve their chances of meeting users needs, libraries need to provide library systems that look and function more in line with Google, Amazon and other open Web search engines. (OCLC/JISC, 2010 [pdf]
With the above context as a driving force, the Summon service was launched in January 2009 to serve as a compelling digital front door for library resources with a mission to return researchers to the library. It remains the only discovery service that is built to purpose, expressly designed to meet the needs and expectations of today’s users. While Summon users start with a single search box, the idea of a single search box for searching across multiple library resources is nothing new. In fact, libraries have tried single search box approaches before—single-vendor aggregator platforms, federated search engines and next-generation catalogs. But the failings of these systems to stem the flow of researchers to Google and the open Web are well documented.
So what’s different with the Summon service? At its core is a single, unified index architecture which allows users to effectively search and navigate across almost all of the library’s resources in a manner that meets users’ expectations for what a modern search should be.
Unlike other discovery services, the Summon service is not a re-purposed database platform or hybrid next-generation catalog. The Summon service is built around a unified index that now contains over 1 billion items normalized and mapped to a common schema which allows results to be returned as a single, unified results set. This approach truly moves users beyond searching siloes of content and ensures that all content has an equal chance of being discovered without bias toward publisher, vendor or format.
The combination of a unified index and a unified result set impacts just about every factor that’s important to a discovery service. Features such as speed, unbiased relevance ranking, dynamic user display options, granularity of facets, ease of navigation across result sets, ability to pre-scope searches, discipline-scoped searching, recommendations and suggested searches, data-driven design features and important contextual research assistance are all possible in the Summon service because of its unique system architecture.
All of these features add up to allow the Summon service to provide a user experience that is optimized for the vast amounts of content being searched. System architecture also greatly impacts how customizable a discovery service is, which is where Rochkind’s analysis of APIs and the strengths of the Summon API only begins to scratch the surface. Libraries who wish to use discovery service APIs to integrate results into custom interfaces shouldn’t lose sight of the overall problem they’re trying to solve and the importance of user experience in solving it.
In order to keep pace with rapidly evolving user expectations, the Summon service was introduced to market as the most flexible and customizable discovery service available with a powerful API that ensures libraries can customize the Summon service and integrate Summon results in any Web page or custom application imaginable.
Many libraries don’t realize that the “out-of-the-box” Summon user interface is built using the Summon API. Designed from day one so that all searches, whether from a custom developed interface or the “out-of-the-box” interface, result in a search call made against the Summon unified index which then returns a unified set of results that can be displayed in any interface or application. In other words, the Summon API is inherent to the service.
Developers using the Summon API to develop their own applications have access to just about every feature available to users in the “out-of-the-box” Summon user interface and enjoy the same unmatched speeds, reliability and currency of the service. Serials Solutions’ Software as a Service and agile development model allows API users to take immediate advantage of enhancements, innovative features and functionality that are released every three weeks.
Today, dozens of innovative libraries such as the University of Michigan, the University of Toronto, North Carolina State University, Villanova University, Columbia University, Aarhus University and the California State University Library System are using the Summon API to customize the discovery experience for their users. But hundreds of libraries who use the “out-of-the-box” Summon interface also benefit from the rich and flexible capabilities of the Summon API including numerous opportunities to customize the user interface without programming skills or knowledge.
In fact, the vast majority of customers who do use the Summon API do so only to integrate Summon results with special projects, custom applications and Web pages. And, in many cases, they still also link their users to the “out-of-the-box” Summon interface which these libraries recognize provides groundbreaking features and an optimized discovery experience to meet the needs and expectations of today’s users.