Keeping Found Things Found: Web Tools Don’t Always Mesh With How People Work

Of all the personal computers to be unwrapped during the holiday season, more than 80 percent will be used to go online and search the Web’s more than 92 million gigabytes of data (comparable to a 2 billion-volume encyclopedia). Getting online is the easy part, finding a useful Web page is a bit harder-keeping track of a useful Web page is another issue altogether.

People have devised many tricks-such as sending e- mails to themselves or jotting on sticky notes-for keeping track of Web pages, but William Jones and Harry Bruce at the University of Washington’s Information School and Susan Dumais of Microsoft Research have found that often people don’t use any of them when it comes time to revisit a Web page. Instead, they rely on their ability to find the Web page all over again.

“People are terribly challenged by this, and the problem is becoming worse,” Bruce said. “People should have fast, easy access to the right information, at the right time, in the right place, in the right quantity to complete the task at hand.”

Maybe you’ve tried some or all of these ways to keep tabs on a found Web page: Send e-mail to yourself with the link. Send email to others and find the link later in your sent mail. Print out the Web page and file it-or pile it-in your office. Save the Web page as a file with your other documents. Bookmark the page as one of your favorites. Add the link to a personal Web site. Write down the URL and comments on a sticky note or piece of paper. Paste the URL into another document. Create a note in a task manager or calendar system.

“Our results so far tell us the tools for keeping track of Web pages don’t mesh well with how people work with the Web,” Jones said. The researchers led a panel discussion and tutorial on the topic at the recent American Society for Information Science and Technology 2003 meeting. “People make up with their own creativity for what’s lacking in their tools.”

The researchers are studying how people “keep found things found” when looking for information on the Web. Their goal is to understand why people do what they do and to develop more useful tools for managing this information. The work is supported by the National Science Foundation (NSF), the independent federal agency that supports fundamental research and education across all fields of science and engineering.

The research suggests that “keeping” techniques arise from the different ways people intend to use the information. Sometimes a person desires the portability of paper, the anywhere access of email, the permanence of a saved file or the speedy access from the browser’s bookmarks. Aside from speed, though, bookmarks-the primary “keeping” tool provided by most Web browsers-rank low on many characteristics that users want, the researchers found.

In addition, regardless of your “keeping” technique, Jones, Bruce and Dumais have found that, when you want to revisit a Web site, there’s a good chance you first try three other options: directly entering the URL in your Web browser (often with help from the browser’s autocompletion feature); searching with a search engine; or accessing it via another Web site or portal.

However, the researchers note the limits of relying on memory, a search engine or the browser’s help. A memory lapse, a change in the search engine site rankings or a move to another computer can eliminate a person’s path to the Web page.

As an experiment to address some of the issues, Jones and Bruce with their students created an alternative “Add to Favorites” dialog for a Web browser. From a single dialog, the person can add comments about a link, send it via e-mail or save the page to their hard drive. While the dialog offers some incremental enhancements, they found that most testers had lost the habit of using bookmarks and didn’t adopt the new Add to Favorites option.

In a follow-on project, the researchers are developing a conceptual framework for how people decide to keep information available for use later on. Bruce calls this the “personal anticipation of information need,” or PAIN. The acronym suggests two motivations for users struggling to keep track of information, according to Bruce. “People are motivated by PAIN,” he said.

The researchers are also looking at the broader problem of “information fragmentation.” People often have the information required to complete a task scattered across e-mail, files, contact information, Web references and other types of information-each with its own distinct organization. The team is pursuing a patent for techniques and tools to bring together the various types of information into a “My Life” personal taxonomy that helps, rather than hinders, efforts to complete the task at hand.

“We live in the Information Age, and the effective use of information is the key to prosperity,” said Maria Zemankova, NSF program officer for the Information and Data Management program. “This project provides the theoretical underpinnings for a humancentric information environment that will enable individuals, families, organizations and societies to continually build upon their experience and gain control over the seemingly endless volumes of information confronting them.”