Here’s how the scribblings of John Adams might help build a better search engine

John Adams, a Founding Father and the second president of the United States, scrawled comments in the margins of several passages of Julie, ou La nouvelle Héloïse.

John Adams once owned a copy of Jean-Jacques Rousseau’s 1761 philosophical novel Julie, ou La nouvelle Héloïse that now belongs to the Boston Public Library. We know this because Adams, a Founding Father and the second president of the United States, scrawled comments in the margins of several passages.

When Rousseau writes that aristocrats and plutocrats should be punished, Adams notes that it is society itself that institutes aristocracy or plutocracy: “Peoples, Nations, not Individuals, are guilty of this,” he scribbles. “Riches and fame are Chimeras too.”

Northeastern professor David Smith isn’t concerned with what authors such as  Rousseau have to say; he’d rather know how readers interact with their texts.

He’s analyzing the underlined and highlighted passages, the drawings, and the handwritten scribbles in the margins in order to get a glimpse into the mind of the reader. Thus, the more marked-up a text, he suggested, the better.

His latest project builds on his exploration of how news stories, fiction, and poetry went viral in the 19th century.

David Smith’s latest project builds on his exploration of how news stories, fiction, and poetry went viral in the 19th century. Courtesy photo.

“This project is applying those results to the question of, OK, can people do more with texts than just reprint them?” said Smith, an associate professor in the Khoury College of Computer and Information Sciences and a founding member of the NULab for Texts, Maps, and Networks. “Because we’ve seeded the world with identical copies of books it’s interesting to ask how people have customized them.”

Smith has received an $82,000 grant from the National Endowment for the Humanities that will enable him to use technology he has developed with his students to scan several copies of the same book simultaneously and transcribe the annotations in order to analyze reader engagement.

Smith said his findings could lead to the creation of better and more efficient search engine algorithms.

“I hope this will provide us with technology to build better search engines so that somebody who is interested in searching all the records of the past can more quickly focus on the more relevant passages,” he said.

Smith also said that his findings will enable researchers to examine the kinds of information that various groups of people, including former presidents and plebeians alike, have found important at particular periods of time. That’s something that could benefit students, he said.

“A lot of libraries contain books that students have studied and students write in these books,” Smith said. “So as a way of finding out what they find most relevant, we can work backwards to help students find the more relevant passages and find what they are struggling with.”

For media inquiries, please contact media@northeastern.edu.