April 10, 2002

Jared Spool in London

A very entertaining lecture (“Beyond Common Sense”) Monday night by New York’s foremost usability expert.

For me, the most important point he makes is that today’s design process does not have a feedback loop built in. We build, and then move on to the next job (“throw it over the fence”), never listening to the users or learning from the mistakes we made. Or for that matter, learning from what works.

This ties in with his vision for a design process that boasts the same degree of empiric knowledge that engineers draw on: Design patterns1. Documented designs that have been proved suitable for facilitating certain behaviours. In the Q&A Tim Ostler of Scient asked whether IP law wouldn’t get in the way, but Jared says clever design gets around that. (I hope that doesn’t mean such similar-but-different solutions like Adobe, Macromedia and Microsoft’s different takes on the “tear-off palette”, or a hundred different ways of saying “one-click”.) But Jared’s message is a valuable one in an era when “creativity” and “innovation” (often just self-expression by another name) have become such incontestible touchstones. “Until you’ve got a better answer, you copy”, I just happened to read today in Ogilvy on advertising (quite a Jakob himself).

Another very useful insight, regarding user testing, was “user-created tasks”. Find out by interview what are realistic on-line tasks for each respondant, and test those. Arbitrary, forced tasks will not reveal typical behaviour.

Despite the abundant (very funny) Jakob-bashing, it has to be said that Jared makes just as many controversial (or, frankly, dubious) blanket statements. Yes, he’ll quote the user-testing figures to prove his case, but often I’d draw different conclusions from them. For example Search, which he claims to never result in a successful transaction, since people only turn to them after failing to spot the appropriate ”trigger word”, usually a category. Firstly, this just stresses the necessity of keyword-mapping and misspelling-tolerant search engines. Furthermore, how does he think anyone uses Amazon or eBay if not by searching? And how can you compare that with shopping for clothes where there simply isn’t an exact term to describe a unique article, even if you wanted to shop for clothes that way? He did, effectively, retract his blanket ban on Search by acknowledging its use for “indexable” goods like books, but now I have to deal with people who just remembered “Search is bad.” In the same way, I’ve seen designers interpret his “Back button is the Button of Doom” mantra to mean that they should put Back buttons (by whatever name) in the middle of their pages.

I also found some of his analogies debatable. The “perfect sale situation” he sketches — where you have supply, demand and cash, which he likens to buying milk at 7-Eleven (and yet e-tailers still manage to lose the sale) — may be good for rolling your eyes at, but completely ignores the lifetime’s store of accumulated knowledge that allows people to shop at stores. Can anyone remember the first time, as a child, you were sent to buy the milk? You may well have failed in your mission (because of the scary shopkeeper, not finding it but being too embarrassed to ask, or whatever), and that’s notwithstanding the number of times you’ve seen already watched your mother shop. If you’ve never shopped on-line before, you are like a child.

By harping on faults I could find, I may have given the impression that the lecture wasn’t worthwhile. The only reason why I don’t state all the excellent points he made is that you can read them for yourself.

1. Further reading on the concept of Design Patterns: Christopher Alexander, Martijn van Welie’s Web Design Patterns, The Interaction Design Patterns Home Page, Hypermedia Design Pattern Repository, Common Ground.

Comments