Host Jon Stewart in the studio of The Daily Show
Image via Wikipedia

A few months ago, Jon Stewart on “The Daily Show” gave his take on Twitter. In particular, he went after the many members of Congress who spent a State of the Union Address sending insipid tweets out to their followers. During the report, Stewart brings up a question that should be asked as more and more people turn to social media tools — as being heard becomes more important than listening:

“Why aren’t you paying attention? There’s a reason they don’t allow cell phones in seventh grade classrooms.”

I’ve blogged about this before, but wanted to revisit the subject in light of recently announced research showing that tagging articles stops users from actually reading them. Raluca Budiu, a user-experience specialist for the Nielsen Norman Group, discusses her work revealing that users recall more when simply reading an article than when reading an article and adding tags.

On the surface, it seems like tags should be helpful, Budiu said, since they increase a user’s engagement with an article. In addition to reading, the user considers what tags to give it and enters them. It sounds similar to highlighting key passages of a textbook, or making notes in the margin. So why should they reduce recall?

Budiu found that adding tags cut into the time that each user spent actually reading an article in the first place. In other words, paying attention to tags came at the cost of paying attention to the text.

The inelegance of most social media tools is evident here — the act of using the tool becomes paramount over the activity that the tool was supposed to enhance.

Science blogging is often more about discussing science blogging than it is about science.  Publishers are asking researchers to spend precious time commenting on papers or to take on the work of journalists.  Though valuable, these things are peripheral to being a scientist, where doing the actual research is the raison d’etre.

What we need are tools to save time and and make activities more efficient.  As publishers, we need to develop ways to enhance the reading experience rather than piling tasks on top of it.  Budiu points out a system called SparTag.Us that lets readers tag specific words and passages by clicking on them as they read.  Her research found that this system enhanced recall.  While I’ve gone on at length that I think tagging articles is a horrible organizational system in light of full text searching, if you’re going to promote tagging then at least think about approaches that focus and benefit from concentration and natural workflow.

Social media tools should be about making jobs easier, not about creating new jobs altogether.

Reblog this post [with Zemanta]
David Crotty

David Crotty

David Crotty is the Editorial Director, Journals Policy for Oxford University Press. He serves on the Board of Directors for the STM Association, the Society for Scholarly Publishing and CHOR, Inc. David received his PhD in Genetics from Columbia University and did developmental neuroscience research at Caltech before moving from the bench to publishing.

View All Posts by David Crotty


9 Thoughts on "When Tools Dominate Tasks"

Unless I read that wrong, the tagging decreases recall for the tagger. But what about for the users searching on those tags?

If that decreases recall too, then I’m going to swear off my new addiction to Games With a Purpose (, where CMU computer scientist have turned tagging into an oddly compelling cooperative series of games. I’ve been justifying the time I spend there by saying I’m helping to improve the internet. But after reading this, maybe it isn’t true!

You read it correctly, it’s about losing recall because you’re thinking about the tags you’re going to put on the article, rather than focusing on the article itself.

Tagging can certainly help other readers find articles, but it’s difficult to convince people that this is a worthy activity. Telling a scientist who already feels overbooked that they need to spend time and effort (and apparently lose reading retention) in order to help unknown others with no benefit to themselves is a pretty hard sell. I guess the idea is that if everyone does it, you’ll benefit too. But everyone is not doing it, so sites that rely on it are pretty much useless. Furthermore, if everyone else does start doing it, you can still benefit without putting forth the effort yourself.

The one place where I can see value in tagging articles is tagging the articles you have authored yourself. It might make them easier for others to find and read, and you can see the benefit in having your own articles read and cited as often as possible.

David wrote:
The inelegance of most social media tools is evident here — the act of using the tool becomes paramount over the activity that the tool was supposed to enhance.

Perceptive point! Social media seems to be a technology desperate to find a market; or more pointedly, desperate to create a market.

You make the point repeatedly that scientists don’t have time for social media. I think what you really mean is that scientists reap few (if any) rewards from this media, which makes tasks like blogging, tagging, and commenting rank low on their priority lists.

Change the traditional reward system and you’ll get scientists to jump through hoops. The question, naturally, is whether there is a compelling reason to ask scientists to blog, tag, and comment. If you find yourself searching hard for an answer then it is likely that when the hype dies down, little will change.

Phil, I agree that most of these new tools are trying to create a niche to fill, rather than filling a need that already exists. But I don’t think the point is that scientists don’t have time for social media. It’s more complicated than that. Scientists don’t have a lot of time period. And, the further you advance in your career, the less time you have.

So that means any activity you take on, social or not, needs to have direct and useful results. So far, most of the new tools don’t provide those results. In particular, any tool that requires mass uptake to be useful is dead from the get-go. Because of the second point above, those who you’d most want to participate in your social venture are the least likely to do so. Which hurts any activity as it is most likely to be filled with early graduate students (who can spare the time) who don’t have the depth of knowledge their more experienced peers have.

Here’s my thought. If tagging articles is like using a highlighter (which seems a reasonable assumption) does this research suggest that using a highlighter reduces recall and retention, too? Anecdotal evidence with undergrads certainly suggests that it might.

One issue might be an excessive focus on the detailed content without due attention to the overall argument.

Personally, I always had a hard time concentrating when trying to highlight, and quickly gave it up as a practice. Although what’s interesting here is that the Nielsen Group is saying that a tool where you click on terms or select passages for tagging works well, which to me is the electronic equivalent of highlighting.

Either way, I was reminded of this blog entry.

I think you’ve really nailed it. It’s a simple cost/benefit balance. Its just that the productivity cost of using many social tools outweighs the benefits – even in aggregate. In science research this cost/benefit balance is particularly stark, making social tools a tougher sell for these tasks.


Leave a Comment