INSPIRE updates a classic

The INSPIRE website. (Click to go there.)

This article first appeared in the May 10 issue of SLAC Today.

INSPIRE, the next-generation high energy physics information system developed by a team from CERN, DESY, Fermilab and SLAC, is nearing production status but still very much open to feedback.

INSPIRE is the successor to SPIRES-HEP, the 40-year-old information system originally designed to provide access to preprints, or scientific papers that have not yet been published in peer-reviewed journals. Timely access to preprints can prevent duplicated effort, confirm or discount the direction of a proposed experiment or provide a vital clue to a struggling researcher, but the publication process can take weeks, months—even years. In addition to providing access to preprints, SPIRES now serves a role in providing access to all high energy physics literature, both pre- and post-publication, and is even used to ensure that the preprint version is unified with the final published version.

SPIRES-HEP has been responsible for a series of milestones in its day—including becoming the first database accessible through the World Wide Web in 1991. But SPIRES-HEP has been feeling its age, according to SLAC’s Manager of Scientific Information Systems Travis Brooks.

“We get about 50 emails per day” for SPIRES problems, Brooks said, including an increasing number of complaints that SPIRES is simply obsolete—too slow and lacking in features compatible with modern systems. INSPIRE is designed to address these increasingly common complaints, offering faster searches, a variety of search and display options, and more detailed record pages, to name only a few of the upgrades.

“One of the biggest improvements we’ve made is author disambiguation, or distinguishing between authors with the same name,” Brooks said. In fact, scholarly publications often identify authors by only last name and first initial, a convention he termed “a bear to deal with.”

To clear up the confusion with names the development team added a graduate student at CERN to their team who is writing specialized code so cutting-edge that his work on INSPIRE forms the backbone of his PhD thesis in Computer Science, according to Brooks.

“We’re using algorithms hot off the presses to solve these problems,” Brooks said. “We’re also reaching out to scientists to help.” INSPIRE will include tools for users to correct inaccuracies themselves. “The next addition will probably be a tool to help fix citations,” he added.

According to Brooks, late summer is the team’s target date for turning off SPIRES-HEP and moving solely to INSPIRE, and in the meantime, they welcome feedback. They’re making sure they’re easy to find—there’s a new INSPIRE blog and even a twitter feed.

The adoption of social media reflects the team’s philosophy. “One of the most important things for me is to reach out to the community” to both give and receive help and advice, Brooks said. “Now that we have a solid infrastructure there’s an opportunity to use this tool to facilitate HEP communication, establishing a dialogue with our users to ensure we meet their needs with regard to the tools we provide as well as the accuracy of our content.”

But, as Brooks puts it, “Reading new papers is almost the definition of being a scientist.” The core purpose of INSPIRE—to provide access to scientific information quickly and easily—remains unchanged.

Lori Ann White