Opening ShotsFEATURESAbout NYRM


“You’re testing people’s patience with 1,000 words,” Slate’s editor, Jacob Weisberg, says. Slate, the most successful web-only magazine, draws up to a million unique users daily, 10 to 11 million monthly—and the typical maximum length for its articles is 1,000 words. Slate accepts the assumptions of web-usability experts like Jakob Nielsen, who wrote, “People tend to be ruthless in abandoning long-winded sites; they mainly want to skim highlights.”

Weisberg chose 1,000 words arbitrarily, but he and Nielsen are echoing the prevailing wisdom that, on the web, shorter is always better. If you don’t keep it short, you’ll lose the reader’s attention—and once a reader is bored, all it takes is one mouse click to kill an article.

Are these assumptions justified? Based on my personal experience, an informal survey of the literature in the field and conversations with various social psychologists and social scientists, I believe they are not.

I’ll start with some anecdotal evidence. For all sorts of reasons, I’d rather read articles—of any length—on the web, rather than printed on paper, even those longer than 1,000 words. For example, I spent a great deal of time on the web doing research for this article.

Since embarking on a journalism career, I’ve become increasingly adept at and dependent on using the internet to find the majority of my background information. First, I’d read articles onscreen because I could access them anytime for free. Then, I’d do it to avoid wasting paper. Soon, I felt I was able to read faster on the web, and I even retained the information more easily. So when I need to read anything, from a two-page article to an 80-page study, I often choose the internet.

And I’m not the only one. Studies from Germany1 to Florida2 that track readers’ responses to reading news online and in print form debunk theories that people won’t read onscreen for long periods of time. These studies suggest that the online readers read more and retained more. The Poynter Institute in Florida found that online readers directed more visual attention to the story and were more likely to read the entire story, short or long, than its print counterpart.

Does that mean readers have longer attention spans on the web?

Attention span, it turns out, is a hard concept to pin down. In the literature on the subject, I found estimates of the average person’s span ranging from seven to 45 minutes. Even though a 1997 paper in Arizona State University’s Research magazine stated that a person can maintain only an eight-minute-long attention span before the mind begins to wander, I don’t recall ever reading something and shutting down at the eight-minute mark.

Did that figure apply to everyone? Not really. In fact, “There is no such thing as ‘an average attention span,’” Sam Goldstein says. He’s a research professor of psychology at George Mason University and editor in chief of the Journal of Attention Disorders. It varies from person to person, he asserts. The criterion is simple: the more interest you have in an activity, the more attention you give it, regardless of whether it’s long or short, online or in print.

Attention span is not a new field of study. Goldstein cites a series of papers written 300 years ago by a Scottish physician, Alexander Crichton. In An Inquiry into the Nature and Origin of Mental Derangement, Crichton writes, “The power of attention is different, not only in different individuals, but also in himself at different times,”3 and it can be influenced by what people eat, how tired they feel or whether they’re sick.

There are few definitive studies of article length online versus on paper. A recent one, EyeTrack07,4 was conducted by the Poynter Institute’s Sara Quinn and Pegie Stark Adam and the University of Florida’s Communications Department. After outfitting 582 people with eyetracking glasses to monitor eye movements and reading patterns for news websites, broadsheet newspapers and tabloids, the team reported that online readers consumed nearly 15 percent more than readers of broadsheets and 20 percent more than readers of tabloids. The data also revealed that article lengths had not influenced the readers as much as the researchers expected.

“The length of the articles did not make a difference,” says Mary Ann Ferguson, the University of Florida professor who contributed to the daily analysis. Online readers read more whether they read short stories (one to four inches long) or much longer ones (19 inches or more). And online readers were more inclined to read the entire article than were print readers.

In a recent interview, Quinn wondered whether online readers digested more because they had more control over the information presented onscreen. With 30 lines of text on a screen, “you know what’s on your plate at any given time,” she says. “With broadsheets, you see 100 inches of text, which can be a little daunting.”

So if length is not the issue, and online is less expensive and more convenient, does this mean that, as many have predicted, print is on its way out?

“I don’t think print will ever become obsolete,” says Pamela Rutledge, the director of the Media Psychology Research Center in Boston. Going online may give readers such perks as customizing the material and format to their tastes, she says, but it cannot replace the act of reading a book or magazine. On rainy days, people prefer to sit and read a book, not their computer or PDA. At the beach, they’re more inclined to flip through a magazine than try to use a computer. For Rutledge, this is “reading as part of an experience.”

Each medium—print and online—has something it does best, Rutledge says. With online magazines, “your curiosity is continually engaged.” Readers can take printed magazines anywhere, and she believes they serve a “more social purpose.” Given the distinctions, Rutledge doesn’t see the two media as competitors. Rather, “people will be more purposeful in what they choose to use.”

The editors of online sites have learned to adapt to the needs of readers. When Slate launched in 1996, it used page numbers and was updated once a week. “We thought people would print it out,” Weisberg says. Now, the page numbers are long gone and Slate offers a healthy selection of hyperlinks and multimedia tools that don’t translate to the printed page. The hyperlinks often lead readers to related Slate articles. It’s an effective tool, sort of like going down a rabbit hole. With enough tantalizing hyperlinks, a reader could sit in front of the glowing screen for hours on end.

But Weisberg still hesitates to bombard readers with large expanses of text. So Slate breaks its text into digestible chunks. “It’s impossible to make a 5,000-word article work unless you break it up,” he says. It may appear as a series of five 1,000-word articles, or as an ongoing dialogue that continues for several days.

At the other end of the spectrum is The Virginia Quarterly Review, whose editor, Ted Genoways, says many people may think of the 285-page print edition—filled with long-form narratives set up in a single column of text—as more of a book than a magazine.

“It’s rare to run with something less than 6,000 words,” he says. That’s being conservative. The 83-year-old journal, which has won several National Magazine Awards in recent years, typically publishes 7,000-to-10,000-word articles. And, surprisingly, VQR posts the full articles on its website.

How can VQR put 10,000 words on the web and hope to keep its readers from abandoning the site? Like Slate, The Virginia Quarterly Review features links to past articles to transport readers down the rabbit hole. “We draw on archives more and more,” Genoways says.

He is also working with the Pulitzer Center in Washington to create interactive packages to accompany some of the journal’s articles. The first experiment came with the most recent issue. The online version of a 20-page article on HIV/AIDS in Jamaica included photos and audio that could be viewed on the Pulitzer Center’s website. Genoways hopes this will drive more readers to the quarterly’s website to read the original articles.

And the magazine is considering other ways of keeping readers coming to its web edition, from offering articles of varied lengths to including short daily news updates. “We’ve talked about the possibility of serializing things,” he says.

Although it loses that book feel on the web, VQR’s editors are encouraged by the reports on user activity on their site. Genoways says that some reader views exceed 30 minutes, which suggests that people are taking the time to read the articles.

Recently, I had to read a 70-page science article on the web. At one point, I realized that I’d been staring at the computer screen for several hours, and so I decided to print out the article. Since I use recycled paper, I wouldn’t be a total jerk to the planet.

My inkjet light started to blink, warning me of low levels. I crossed my fingers as the printer whined and churned out the first pages. To avoid suffering through the slow progress, I left the room to make a cup of tea and toast. When I returned, the printer was beeping. It had stopped after page 41. I’d run out of paper.

So I went back to reading onscreen, nibbling on toast while I scrolled down page after page of the article. As I brushed crumbs off the keyboard, I thought of Weisberg’s and Nielsen’s views about reading on the web. I had read the whole thing onscreen, some 27,000 words. While it might have been pleasant to read the printed article in my lounge chair, doing it onscreen had not been a chore. In fact, I would’ve found more distractions with the printed version in my hand—I might have stopped to put on socks, get a blanket to ward off some nonexistent chill, make more tea or reorganize my files. Reading on the computer turned out to be more efficient for me.

Maybe I’m quirky, but studies like the ones at the Poynter Institute suggest that I’m not alone.

1. Bucher, H. and Schumacher, P. (2006). “The relevance of attention for selecting news content. An eye-tracking study on attention patterns in the reception of print and online media.” Communications: The European Journal of Communications Research, 31, 347-368.
2. Quinn, S. and Adam, P. S. (2007). “Eyetracking the News: A study of print and online reading.” Poynter Institute. Available (consulted 16 April 2008) at www. eyetrack.poynter.org/index
3. Crichton, A. (1798). “An Inquiry Into the Nature and Origin of Mental Derangement: Comprehending a Concise System of the Physiology and Pathology of the Human Mind and a History of the Passions and Their Effects.” Medicine, Science and Technology, 1, 254-290.
4. Quinn, S. and Adam, P. S. (2007). “Eyetracking the News: A study of print and online reading.” Poynter Institute. Available (consulted 16 April 2008) at www. eyetrack.poynter.org/index

Save your pupils: Read a print magazine (for now)

Probably the biggest factor keeping magazines from chucking their print editions altogether is eyesight. Parents who pull children away from a television screen may use the excuse, “It’s bad for your eyes.” They’ve aimed the same accusation at computers.

Bill Hill, a Microsoft researcher, says of early computer screens: “Reading on the screen was so painful. We hadn’t evolved then.” His colleague, cognitive psychologist Kevin Larson, agrees that the clarity, or readability, of computer text has been the primary deterrent for reading online. In the technology magazine IEEE Spectrum, Larson writes, “It’s not sentimentality that makes most prefer reading books and magazines to squinting at their laptops. The quality of computer text is awful.”

But they have a solution: improve the equipment. “The reading experience shouldn’t be tied to specific hardware,” Hill says. He envisions a world in which document readers—such as those sold by companies with names like Zinio and Texterity—will provide digital versions of print editions that can be read on a handheld PDA as well as a computer monitor. These readers should not only simulate the print-on-paper reading experience, but improve it as well.

Hill looks at such details as typefaces, type size, leading (the space between lines) and line length, and bases much of his research on studies conducted at the University of Minnesota more than 80 years ago, long before home computers arrived. The researchers, Miles A. Tinker and Donald G. Paterson, concluded that people read most efficiently when the type size was 10 points and the line length roughly 80 millimeters—up to 12 words per line.

Increasingly sophisticated hardware utilizing the methodology of reading experts could make online-only magazines more appealing. For Hill, it’s only a matter of time before computer technology completely recreates the experience of reading a book or magazine. “The future of reading is definitely online,” he says.




About | Site Map | Archive | Masthead

Copyright 2008
Columbia University Graduate School of Journalism
George T. Delacorte Center for Magazine Journalism | 2950 Broadway, NY, NY 10027