Jonathan Arnowitz, Elizabeth Dykstra-Erickson
"This design isn't simple."
"This design isn't consistent."
How many times have we heard these design complaints? Engineers and marketing managers have heard us squawking about heuristics and they've finally found their dictionaries and are squawking back!
Resist being bound to oversimplified rules! They are not just parroting our beloved heuristics. They are appreciating the tactic we've been using to force our point of view: Follow "good" heuristics and you will achieve a desirable result.
Design heuristics abound. Our favorite set is the heuristics described in Dix, Finlay, Abowd, and Beale's classic text Human-Computer Interaction as being the best articulated and fully realized "thinking tools." But apparently just about anyone feels that they can successfully use Nielsen's set of ten from the '90s without benefit of shared interpretation.
All too often designers, consultants and pseudo-designers aim to improve the user interface by employing heuristics. Fifteen years ago this was seen as a way of demonstrating the value of HCI by exchanging arcane terminology for everyday wording like "speak the user's language" (does that mean French? Or biomedical?). The time has come for us to realize this over-reliance on simplistic heuristics has had an effect: to ruin our credibility. Although heuristics are useful, an over-reliance on them is risky and does not replace understanding the user population(s). Here are two particularly dangerous heuristics:
A popular interpretation of simplicity holds that just a few items should be on the screen. By virtue of not having much to decipher, it's simple, right? Whoa. Simplicity has to do with the cognitive load on the user. It's a (no pun intended) complex formula of the number of concepts a user must understand, not just a simple item count of words, options, or features. Visual complexity certainly affects the time it takes to understand what an interface presents, but stringing ten concepts into ten separate "wizard" windows doesn't at all reduce the complexity of the conceptit simply reduces visual clutter. "Reducing complexity" by shoving everything onto one screen may reduce the number of input events required to complete a complex operation since there is no window-hopping, but this kind of "one-stop-shopping" offers no reduction in complexity, either. If your marketing manager or engineer claims that the interface is too complicated, perhaps you should ask them if the complexity of the interface matches the complexity of what the user is trying to accomplish.
Does consistency refer to page layout or is it the trickier consistency in interaction and design patterns and meeting user expectations? Anyone can make screens look similar from one page to another. But just because every page is in nine-point Arial, with gray backgrounds and has an "ok" and "cancel" button in the lower right-hand corner, six pixels from each other (and the right margin), does not mean that every page will be usable. Consistency is at least as deep a concept as simplicity: It's not a hard and fast rule, and it should be applied when it makes sense. Disobey the guideline when you need tobeing a slave to consistency will at some point paint you into a very uncomfortable corner.
Heuristics are being oversold and we are overselling them. We advocate being more careful in what we present to our non-design colleagues as a design rationale or motivation for change. Heuristics, by their very nature, are mutable, contextual, flexible. Use them wisely, or they will be used against you!<eic>
©2005 ACM 1072-5220/05/1100 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2005 ACM, Inc.