Lo-fi prototyping

Comments on the paper Prototyping for Tiny Fingers Marc Rettig, Communications of the ACM, Vol 37, Nr 4, April 1994, pp 21--27.

Good points

  1. The lo-fi approach is refreshingly low-tech. In an age of technology-driven solutions, it is great to come across a down-to-earth technique that focuses on the problem (evaluating user interaction) rather than the technology.
  2. The author's reminder that this technique, like any other, is effective only with proper preparation, performance, and analysis. Usually, the problem is the other way around -- the high-tech solution forgets the fundamentals -- but it is nonetheless important that they be emphasized with any unfamiliar technique.
  3. It sounds like fun!

Bad points

  1. The title. At first, I thought the paper was something to do with user interfaces for children. Having now read the paper through twice, I still don't understand what the title means. Perhaps it's cultural -- at any rate, this is a good example of a usability expert in one area blithely ignoring usability issues in other areas. A meaningful title is surely one of the first rules of communicating technical information. (A similar comment applies to the typography in the Interactions journal.)
  2. Lack of verification. All evidence that the technique described in the paper actually works is firstly, anecdotal, and secondly, based on an assumed correspondance between the paper prototype and the finished product. (I guess this is really two points.) The only benefit actually reported in this paper is that the people who did the lo-fi prototyping felt good about it. There is no evidence that this translates into a better end product. The author states: "In six hours we had learned the technique, designed an interface, and measurably improved the original design." Hang on a minute -- improved what, exactly? The lo-fi prototype!
  3. The assumption that hi-fi is expensive. The author makes much of the difference in cost between lo-fi and hi-fi prototyping. I don't see why this is necessarily the case -- perhaps his hi-fi tools are just mediocre. Given the differences between what the user sees in lo-fi and hi-fi prototypes (and the lack of any empirical support in favour of the lo-fi technique), it is hard to accept, untested, any assertion that lo-fi ultimately provides a better interface at lower cost.

John Reekie, February 10th, 1998.