Search in Context
Search-as-application demand is also growing in industries that have fallen under increasing regulatory oversight, such as financial services, pharmaceuticals, energy, and healthcare. John Kreisa, director of industry solutions for Mark Logic, which provides information infrastructure software, says, "We're seeing new demand coming from enterprises with dynamic information environments paired with regulatory challenges." JetBlue is among the airlines using Mark Logic solutions to facilitate access and retrieval of information by crew members, partners, and regulators.
But even less stringently regulated industries are looking for search to do more. Endeca's Sonderegger observes, "There are more people inside organizations who have to make important decisions on a daily basis. We try to improve discovery even for people who have no specific technical expertise."
One of Endeca's customers is the Harris Corp., an international communications and information technology company serving government and commercial markets in more than 150 countries. Its engineers use search to help them select parts for products they are designing. "Harris has 3 million parts in use, but they also have to give engineers access to parts in the market that they're not using-and there are 70 million of those," Sonderegger explains.
Endeca's solution for Harris enables engineers to query both structured and unstructured data sets, and it returns summaries of how the parts, suppliers, and cost data interrelate. "There is a lot
of pressure on supply chains right now to improve decision making and shorten the time it takes," says Sonderegger. "This enables the company to point engineers toward preferred manufacturers."
Knovel, which offers a web-based application integrating technical information with analytical and search tools, also strives to keep context at the center of its search solutions. Chris Forbes, president of Knovel, says, "For engineers, search needs to be nuanced to be useful." He points out that, for an engineer, search is a nexus for a wide range of content types, from texts to tables to graphs to mathematical formulas. A typical search might be "find me a series of equations to measure the bending rates of a heart valve," says Forbes.
The Knovel application architecture is sensitive to information types, metadata, and attributes that are unique to its user constituency. Users can run simple keyword and phrase searches, but Knovel also supports unlimited Boolean searching with wild-card truncation, phrase matching, and bibliographic limit fields across data that may appear in a table, graph, or chart.
The University of Virginia Press' (UVaP) Rotunda project presents original digital scholarship including the American Founding Era collection, along with newly digitized critical and documentary editions in the humanities and social sciences. This valuable collection, however, isn't necessarily straightforward to navigate for the modern scholar.
Mark Saunders, assistant director to the UVaP, says, "The founding fathers' papers contain all sorts of strange abbreviations, for instance ‘Philadelphia' can be Philapa or Philada. And the term ‘slavery' is almost never used, even though the content of the documents may refer to it." Users can perform full-text searches using criteria such as author, recipient, and date, but there is behind-the-scenes curatorial expertise at work as well. The Mark Logic back end enables the UVaP team to curate and load word lists and thesauri to make perusing both the original documents and the scholarly annotations easier.