Thursday, March 27, 2008

The Art of the Prototype

"Plan to throw one away; you will, anyhow.'' (Fred Brooks, The Mythical Man-Month, Chapter 11) This highlights a point that's often overlooked in software: it's impossible to fully understand a problem until you attempted to solve it; therefore, your first solution will always be lacking. In software we tend to assume that we can do a few case studies, put together some user stories, and then roll out a piece of software that will meet the needs of our users. This has led to some major disasters in software; for some shining examples, just look at the market of enterprise software.

These problems could be lessened if more companies would invest in a serious prototyping stage. I'm not talking about a PowerPoint presentation showing the imagined functionality of your software. I mean pulling out the rapid prototyping tools, creating something that meets the core needs of your users to solve the basic problem, and rolling it out to 20-30% of your users to watch while they use it for 6 months. When we do this, we are following Brooks' council in planning to throw away the prototype, but saving time and energy by not wasting a lot of time creating a tool that has a good probability of begin fundamentally flawed.

No comments: