The first 2 days here have given me a renewed sense that, in a funny way, the most important people aren't here: the people FOR WHOM we're designing sites. Listening to George Olsen's excellent presentation on personas, it struck me that IA is following a perilous path, similar to that travelled by librarians at the turn of the 20th century. And that is the path of trying to understand the "typical user" of our systems.
The peril does not lie in the task, but in the way one approaches it. Charles Cutter, back at the end of the 19th century, argued that controlled vocabularies should fit "the public's habitual way of looking at things." People can argue with that idea: what about outliers? What about marginalized groups? etc. But when we let our hair down and permit ourselves some self-honesty, we all aim to do the best we can for the greatest number of people, and identifying a persona, or a "typical user" is a necessary task.
The peril comes in the way you establish that typical user. In the early days of controlled vocabularies, we tended to assume that we somehow "could guess" who these users were, what they wanted, what they liked, what they disliked. Our ideals (admirable ones) were linked to poorly-executed constructions of these users. It so often seems like "common sense": but the last 2 days have been full of accounts that surprise us: log files that reveal things we never expected, accounts and research and tests that showed us just how off base our common sense might be.
IA, it seems to me, is very properly reiterating the need for research into what users want, need, like, dislike, etc. The practitioners and the researchers may not always agree on the desirable level and detail of that research. Nonetheless, there seems to be a sound and effective realization that without listening to the users SOMEHOW, our constructions of the "typical people" will not serve our real people.
Grant, you are right. Constructions of "real people" will not serve the real people out there. I try to keep in touch with "real people" by volunteering at my local public library, teaching older folks how to get onto the internet and why web sites are all so different. It forces me back into that beginner space.
But how real is that? It's not an average population either, mostly older folks who have suddenly inherited a computer and don't know how to use it. Or folks who have never been on the net. That population is shrinking, and is statistically only a small portion of "real people." But their needs are extreme -- bread crumbs must be explained, menus, clicking, all are frightening. And there's no distinguishing the "parts" of the design -- every piece is important and they are confused. Unlike us, who fly through a design because we are so used to using the components.
Posted by: Jan Wright at March 3, 2004 11:38 PM