Reference Extract Project Gets $350K MacArthur Grant for Next Phase

Reference Extract, a search engine project being developed by researchers from OCLC, the information schools of Syracuse University, NY, and the University of Washington, Seattle, and software consultancy Zepheira, recently received an additional $350,000 grant from the John D. and Catherine T. MacArthur Foundation, allowing the project to move ahead from its initial planning phase.

As envisioned, the Reference Extract search engine project would draw on a collection of URLs and other citation data supplied by librarians in response to reference questions. Data would be collected from librarians during reference interactions and added to the database on an ongoing basis; that data would then be used to provide answers to Reference Extract search queries.

Reference Extract search functionality could also be integrated into other search engines and web tools, and used by librarians and patrons alike. Ideally, if enough librarians participate, it could create a searchable treasure trove of librarian expertise—in effect, a database continually curated by reference librarians.

As LJ reported, the project was first announced in November 2008 after receiving a $100,000 MacArthur planning grant. The resulting planning pulled in expertise from inside and outside of the library world.

Phase two
Jeff Penka, portfolio director for OCLC end user services, told LJ that researchers have started to build the technical infrastructure for Reference Extract in this next phase of the project, and are creating proof-of-concept utilities to show how it will work. Those will soon be made available.

The immediate plan is to reengage with the librarians to show how the project is moving along. "It's nothing without the librarian community participating," Penka said.

To that end, he said that OCLC representatives recently talked with several libraries about potentially participating and contributing to the Reference Extract research process. A Reference Extract-centered blog is planned to boost community engagement and supply information about the project.

Asked whether OCLC's own QuestionPoint reference management system might be used for data collection for Reference Extract, Penka said that it could potentially be a "feeder," but it would not be the only source. "There's no intent of exclusivity with QuestionPoint," he said.

R. David Lankes, an associate professor at the Syracuse iSchool and the director of its library science program, wrote about the Reference Extract project for LJ's netConnect supplement in April 2009. Lankes has called the Reference Extract concept a "credibility engine."