ARTS & FARCES internet

Lou Rosenfeld absolutely nails the dark side of working in user experience, especially the information architecture piece of it, in his “Banned terms for information architects.” Rosenfeld spotlights a handful of terms that just make my skin crawl when they come up in meetings:

• redesign
• featured sites/resources/anything else
• related links and anything else
• advanced search
• building community
• social media and social anything else
• portal

I’ve been fortunate enough not to have heard “portal” from anyone other than the totally clueless in more than a year, so that’s a tiny step in the right direction.

But if you work in the web, you here these terms from your stakeholders just about every day. These terms are worse than useless; Rosenfeld artfully refers to them as “the lorem ipsums of information architecture.” They’re worse than useless because every individual projects his or her own low-resolution definition on the term. As Rosenfeld points out, they’re like placeholders, serving only to “enable us to defer dealing with tough issues.”

When your website sucks so absolutely horribly that you’re compelled to start over from absolute scratch, guess what? That’s not a redesign. That’s starting over from scratch.

Rosenfeld is my hero for the day because he came up with a creative solution, drawing on his experience as a parent:

“At a client meeting today, I did something that I’ve always wanted to do: I banned the term ‘redesign.” I tried to make it fun, charging US$1 if the clients said it, and US$5 if I did (the money was to go to the poor temp who got stuck with the job of taking notes). Kind of like what we do at home when we swear in front of the kids. Anyway, it seemed to work; kick ass discussion and, thankfully, it’s no longer a “redesign” project.”

Read the comments too.

The Information Architecture Summit 2010 took place this week in Phoenix. I rarely, if ever, attend big honking conferences like these any more because they’re too expensive, dialysis on the road is a bitch, and I don’t have anything to sell. They only make sense if you have something to sell.

Anyway, I was tempted to go to this one because Richard Saul Wurman delivered one of the keynotes. Wurman coined the term “information architecture” in—wait for it—1976 and the totality of his work has informed mine since stumbling across Aspen Visible in the 1970s.

It was really interesting—and disappointing—to follow the twitterstream (#ias10) from the conference. Wurman has to be one of the most abrasive human beings on the planet, but he’s unquestionably brilliant. The Twitter comments during Wurman’s keynote were blistering. Here’s a sampling: “To call this talk wide-ranging is like calling the universe big.” “RSWurman is so cranky he makes me seem perky and light-hearted by comparison.” “Richard Saul Wurman proving that hypertext is better delivered on a screen than in a speech.” “Learning that information architecture happens without ever reaching a point.” “I guess when people treat you like a legend for 20 years, self-importance is an occupational hazard.” “The puppies in the room are cutting RSW no slack, which is as it should be.” “I think RSW is a hoot, but I really wish he’d stop telling us all of the things we don’t know how to do.” “I can’t decide if the pressure is off for my closing keynote tomorrow, or if I’m doubly screwed after this.” “The good news? RSW hasn’t said ‘I shit genius every day!’”

Maybe it’s because Jim Klee, my major professor in both undergraduate and graduate school lectured exactly like Wurman does. As a vaudevillian—winding and looping multiple, seemingly disconnected, stories with many layers together to make a point sometime in the future. Sometimes the loops don’t gell until months into the future.

Or maybe it’s because my attention span is longer than that of the youngs (conferences like the IA Summit are populated almost exclusively by the youngs) who are constantly bathing in the stream that is the internet—a plethora of information bits demanding attention now. A Wurman presentation—like a Klee lecture—is more than anything like an auditory meditation; a modern oral tradition experience. It requires extremely close and careful attention, and the payback isn’t always immediate. And a pretty vast collection of frames of reference doesn’t hurt. I always appreciated the intellectual exercise, always got immense value from them, and will be forever grateful for the experience.

Here’s what I find puzzling: While they used the Twitter backchannel to rip Wurman during his keynote, the youngs gave him a standing ovation at the end of his address. That’s just sad.

A recent eye-tracking usability study indicates that users ignore real-time results in searches. How users use search is, well, weird science. We’ve known for some time, for example, that a typical user will click within 10 seconds (.pdf; 172KB) of the search results being displayed. Users make nearly instantaneous decisions in about one-tenth of a second (the time it takes to fixate).

What’s especially interesting about Oneupweb’s study is that the majority of the participants were active on at least one social media network, so would presumably be predisposed to real-time search results. Both groups of participants—Oneupweb recruited two sets of participants, “consumers” and “information foragers”—fixated on the top results in, on average, under 2 seconds. The participants took, again on average, more than 10 seconds to fixate on the real-time results (“consumers” took 9 seconds; “information foragers” took 14 seconds). More importantly, “consumers” clicked 10% fewer times on the real-time results than the “information foragers.”

Only 25% of the “consumer” participants and 47% of the “information forager” participants liked the real-time results; the majority were “indifferent” to the real-time results. From the Oneupweb study: “What the data showed, and the participants confirmed, was that the links that were likely to get clicks need to appear credible and relevant. And users do trust Google’s top rankings to be the most relevant to their search query.” The study confirms that content placement on a page is a primary predictor of attention. No surprises there. But Oneupweb’s assessment of the study results indicates a Heisenberg’s Uncertainty Principle search paradox: “... as social inputs increase, the relevancy of top ranked search results becomes less predictable.”

This is potentially really bad news for the search engines. Google and Bing have both cut deals—reported to be US$15 million and US$10 million respectively—with Twitter for real-time search results. Chump change for Google and Microsoft, but real money for Twitter.

Dave Winer pegs precisely why we ignore real-time results:

“It’s impossible to convey much information in 140 characters. So when a search hits a tweet you get at most a soundbite, telling you something you probably already knew. When you search you’re looking for information you don’t have but want.”

Usability improved by about six percent each year over the past decade according to Jakob Nielsen’s research. That works out to a seemingly-astounding 77 percent. That sounds like incredibly good news but it’s not: Usability is improving at a much slower rate than other areas of computing. Nielsen estimates that it’ll take “74 years to reach acceptable user experience quality.”

Ouch.

We’ve known for years that usability success rates are misleading. If 70 percent of users are successful, 30 percent fail. As usability experts, we tend to forget the 30 percent failure rate and focus on the 70 percent success rate.

It takes Nielsen most of a screen to get to the really useful data: “During the last decade, we’ve collected formal usability metrics for 262 websites. In 2000, the average failure rate was 39%; in 2010, the average failure rate is 22%.” Nielsen attributes a doubling of conversion rates (one percent to two percent) to the doubling of website usability.

The big elephant in the room question remains unanswered: What’s an acceptable usability failure rate?

ARTS & FARCES has specialized in user experience for many years, but I’ve been reluctant to write much about the topic. For a variety of reasons, that’s changing now and I’m adding a new “User experience” section to Hasten down the wire.

What better way to kick-off the new section than to publish the basic process the company has been using for more than 30 years to architect information, first in video and print and now most recently on the web. The process has evolved over the years and while I’ll begin with a brief skeleton, my intention is to expand it using Hasten down the wire‘s wiki and eventually publish it as a book-length work.

Important note: This is a work in progress, an attempt to make years of notes understandable to a broad audience. I’ll be adding to it regularly, as time allows, so bookmark it and come back often. In a break from my usual practice of being fully transparent in changes made after initial publication, I’m just going to integrate changes and additions into this article in order to keep it easier to read. And don’t forget to check the project wiki.