Features

net.wars: Eavesdropping

A date. Just a simple date. That's not asking for much, is it? I just want to know if your verkochte press release is a month, a year, or a decade old. IS THAT SO GODDAMN MUCH TO ASK?

Actually, it turns out that missing dates are only number three on Jakob Nielsen's list of Top Ten Web Site Mistakes of 2003. Number one is: not saying what the company does.

Number two - I know you will be fascinated to hear - is moving archived content to a new URL. Actually, this was my biggest complaint in 2002. Here is the page my long-suffering assistant, Rachel, maintains of links to my articles around the Web. Once upon a time, they all worked. Last year there was some kind of reshuffle meme going around. The Daily Telegraph for example, entirely redesigned its site, moving every single article. About half were moved to URLs that could be straightforwardly predicted from the old ones. Some of the rest could be manually searched and changed. But there is a bunch that have simply vanished, perhaps forever, into the paper's mass of back articles. No wonder they call it the "morgue".

Nielsen was talking about all this at this week's "User Experience", five fun-filled days of documenting user abuse. The most interesting part of Wednesday's main event was the presentation by Kara Pernice Coyne, who has been developing methods of testing the usability of products. Evolutionary biologists would love this, because of the fabulously self-deceptive behaviour on display.

Coyne is looking at products on three levels: behavioural, reflective, and visceral. Kind of like Freud's ego, superego, and id. Coyne tested three versions each of vegetable peelers, digital cameras, alarm clocks, cappuccino makers, and Web sites, and displayed video snippets of parts of the testing.

These were, of course, the best part. Just as we all love to believe we have an above-average sense of humour, good taste, and a highly tuned ability to judge character, most of us believe we are deeply rational beings who care more about function than form. Or, in Coyne terms, "behavioural attributes [rather] than visceral and reflective ones." In other words, you are not a frivolous, shallow person who would buy a tea strainer because it was cute and cuddly. No, you are a rational, thinking person who is not distracted from the essential need to strain tea by cute and colourful. Personally, I just let the leaves settle to the bottom of the cup; you see what a slob I am. But a rational slob. I have a rule. I only buy things I've fallen in love with.

A number of Alessi products like that tea strainer figured in the presentations on Wednesday, in part because the company specialises in quirky, entertaining, yet functional designs. But its Web site commits the number eight Nielsen Web sin: organising products by designer instead of by attributes that are meaningful to the site visitor.

It was not, however, an Alessi product that scared one of the taped users. "It looks," the woman observed of an industrial strength corkscrew, "like something you'd find in a gynecologist's office." It does. On the other hand: a lot of her revulsion vanished after she tried it and discovered it worked extremely well.

Even more fun was the woman who instantly fell in love with a tiny Casio digital camera. She couldn't stop fondling it or talking about it, even after she'd reluctantly put it down and moved on to the other models, which she disparaged in comparison to the love object.

"I'm sure," she mused, turning it over in her hands yet again, "if I read the manual it would be easy to use." Yet this woman rated herself as highly behavioural motivated - and then wrote on the evaluation form to explain why she liked that camera, "easy to use".

This all, as Coyne says, shows the limitations of focus groups and market surveys. What motivates people is often not what they think. It's not that they're lying to you deliberately; it's that they don't know themselves how they behave. As a friend of mine described his own mind, "It's a big place, and it's dark in there."

Coyne's video clips serve another important purpose. They make companies and designers watch the effects of their products and sites on the innocent users upon whom they are foisted. When I was researching a recent article on Web usability I'd have heard a lot less from the Royal Mail and Odeon about "how much users must like our Web sites because millions of them use it" if those designers had ever watched user after user scream in frustration. Their "popular equals good" argument is what I call the Clapham Junction argument: the fact that millions of people have to change trains there every day doesn't mean they like it.

With this in mind, I invite Nielsen-Norman's UK PR firm to visit me any time so they can suffer along with me as I wait a month for every under-informing page on their Web site to load. Even if they do "deliver Western Europe". And if you want to hear what I think, I've recorded it ...