New Vision for the Monograph

The idea of a product that, at a glance, can point scholars to the exact point in a book that is relevant to their research may sound like science fiction, but that’s how the scholarly database JSTOR is pitching a recently released research tool.

The tool, called Topicgraph, is part of a JSTOR project to take the digital scholarly monograph from a PDF to something more useful for researchers. The organization on Monday published the final version of a white paper outlining 13 ideas on how to do so, ranging from the convenient -- such as giving readers more navigational tools within digital books -- to the complicated, like removing restrictions on how readers use and reuse books.

“Books just have not made the same digital transition in the way that journals have,” Laura Brown, managing director of JSTOR, said in an interview. “We’ve been on a mission to see if we can help unlock that value.”

JSTOR has over the past few years shown -- and responded to -- a growing interest in scholarly monographs. The database in 2013 launched the Books at JSTOR program and has in less than five years added tens of thousands of titles and more than 1,000 library customers.

The database's own metrics show JSTOR's users are eager book readers, Brown said. Yet few would argue that long PDF files make for ideal digital reading. As more and more researchers use general search engines and library websites as a starting point for research, making digital books more easily discoverable could help showcase the monograph -- still seen as the gold standard of scholarly work by many -- to a new generation of scholars, she said.

"We are just at the beginning of that journey," Brown said about JSTOR's work with scholarly monographs. "The usage of books on the JSTOR platform is just exploding."

Alex Humphreys, director of JSTOR Labs, said the results from the Books at JSTOR program suggest the database has tapped into an appetite among researchers to access scholarly monographs digitally. But the work to digitize scholarly monographs has come at a symbolic cost. Breaking up books into chapters has made longer manuscripts more accessible to readers, who tend to only read five to 10 pages of a digital title before determining whether it is relevant to their research, but it has to some extent “journalized” books, he said.

“The value of a long-form piece of scholarship, a continuous argument, a real exploration of a single topic -- some of that gets lost when it’s split up into chapters,” Humphreys said. “We were hoping to find new tools and ways that would bring that value back.”

JSTOR Labs is working on building those tools, Humphreys said. The group can perhaps best be described as a Skunk Works within the scholarly database, which in turn is part of the higher education nonprofit Ithaka. The group works on projects that could bring new functionality to the database and also help scholars in general.

While the group can’t dramatically transform the monograph publishing market on its own, it is focusing on tools it can build to benefit researchers. Examples include works in progress with names such as the “Book-as-Portal-to-Other-Scholarship,” the “Scholarly Reader,” the “Scholarly Influence Graph,” the “Topic Explorer” and the “Way-Better Table of Contents.”

Working with faculty members and graduate students at Columbia University, JSTOR Labs chose to focus first on the “Topic Explorer” idea. Topicgraph, the completed prototype, uses text analysis to find key terms in a manuscript and group them into relevant topics. The tool then displays the top 15 to 25 topics and a graph showing the frequency at which they appear throughout the book next to the manuscript itself. Clicking at a point on the graph navigates to the corresponding page of the manuscript, where the relevant terms are highlighted.

The prototype features about 60 titles from a handful of university presses to show how the tool works, and JSTOR Labs is inviting testers to submit their own manuscripts to be “topicgraphed.” At 52 pages, JSTOR Labs’ own white paper is shorter than the manuscripts that the tool is designed to work with, but feeding the paper to the tool still identifies “digital publications” as the No. 1 topic.

Should researchers respond positively to Topicgraph, the tool could become a standard part of how JSTOR displays scholarly books and journals, Humphreys said. A different JSTOR Labs tool called the Text Analyzer, which scans the contents of a document and recommends similar titles, launched in beta form in March (to praise from Inside Higher Ed blogger Barbara Fister).

As JSTOR Labs continues to refine Topicgraph, it will also build new prototypes, Humphreys said. He added that he is particularly interested in working on tools that turn digital monographs into portals to other research and help researchers with citation and reference mining.

For other projects -- for example, building a dedicated book reader for scholarly works -- JSTOR Labs hopes to partner with other groups involved in scholarly communication, Brown said.

“The reimagined monograph -- whatever that ultimately means -- will not be built in a single step, or by a single organization,” the white paper reads. “Libraries, publishers, scholars, scholarly societies and others will all have a role to play -- in promoting standards, in convening thinkers, in carrying out technology development and so on -- and in doing so, they will be drawing on the wonderful history of collaboration in the scholarly communications community.”