September 14, 2009
Software makes library searches fast, easy, comprehensive
Results will draw automatically from different disciplines, encouraging sharing and discovery
susan.hagen@rochester.edu
The new eXtensible Catalog will “revolutionize the researcher’s experience,” says Susan Gibbons, the dean of River Campus Libraries. “In a sense, we are doing for library resources what Google has done for the Web,” she says.
In the age of Google, quick, one-stop Web searches are de rigueur—except at the very institutions that have traditionally been the greatest repositories of knowledge. At research libraries, users must wade through hundreds of online databases, from AccessScience to Zoological Record Plus, to fully explore the collections.
Now, thanks to the eXtensible Catalog (XC) software developed at the University and provided free to other institutions, scholars will soon be able to mine these online research collections through a single, user-friendly search tool.
“This will revolutionize the researcher’s experience,” says Susan Gibbons, the Andrew H. and Janet Dayton Neilly Dean of River Campus Libraries. With the XC application, library patrons will be able to conduct the kinds of comprehensive searches of specialized library collections they have come to expect from other online applications.
“In a sense, we are doing for library resources what Google has done for the Web,” says Gibbons.
So why not just use Google? The problem is that most libraries’ digital collections are subscription-based databases of journals, newspapers, and digital images, such as Lexus/Nexus, JStor, or Proquest. For example, Rush Rhees Library spends between $4 million and $5 million annually on access to such online material, which is average for a research institution.
“Copyright holders of the material want to sell the rights to their collections, not give it away for free to search engines,” Gibbons points out. “So the best scholarly information is still outside of Google. By creating a tool to search these repositories, the eXtensible Catalog will shine a light on some of the highest quality information online today.”
To scale the walls of proprietary databases, the library has developed a two-pronged approach. First they developed software to index and standardize data from different catalogues, allowing the information to be searched quickly and effectively. Through standardization, the software can recognize, for example, when the same journal or book is duplicated across different platforms and return a single reference instead of multiple hits for the same work. Second, the library is seeking permission from copyright holders to harvest data from their systems on an ongoing basis.
The XC software also indexes and standardizes libraries’ in-house collections, such as catalogs of print holdings or digital copies of master’s and doctoral theses. The end result is a search system that will return results from multiple silos of information within a library. By its very structure, the eXtensible Catalog will introduce researchers to material they never knew existed, Gibbons says. Results will draw automatically from different disciplines, encouraging sharing and discovery across academic fields.
Funded by the Andrew W. Mellon Foundation, the University, partner libraries, consortia, and private companies, XC was built over the past two years by a University team of seven programmers and four principal investigators, assisted by research and development partners at 12 institutions. To ensure that the end product answered the needs of serious scholars, the design was guided by a series of anthropological studies of graduate students and faculty at Rochester, Yale, Cornell, and Ohio State.
“One of the things we discovered was that people really hate using microfiche and microfilm,” said Nancy Fried Foster, director of anthropological research for River Campus Libraries and a principal investigator on the project. That finding underlined the importance of building a search tool that would indicate the different formats in which a resource is available, allowing scholars to choose what works best for their research. Along with formats, the software also allows search results to be grouped by customizable categories like author or date, the way that Amazon and other commercial sites return results.
Helping libraries introduce such standard functionality to their Web sites is an important goal of the project, says Jennifer Bowen, assistant dean for information management services and also a principal investigator on the project. “Libraries cannot afford much technology. They have been stuck while the rest of the Internet has moved on.”
In recognition of those financial constraints, XC was designed as open-source software available for download at no cost. According to David Lindahl, software development director and another of the principal investigators, two of the planned five toolkits are already online at www.eXtensibleCatalog.org, and the final three toolkits are slated for release by January 2010. Libraries are encouraged to adopt and adapt the software to meet local needs, and a full-time programmer is available through the Web site for the next two years to answer users’ questions, work out software glitches, and incorporate community contributions to the open-source code.
“We would love for all kinds of institutions to build on it,” says Bowen. “We’ve started it, and we hope it will thrive.”
| Previous story | Next story |
