Guiding Principles of a Good Design


Posted by Ken Ramsley , Nov 14,2002,14:05     Forum

Among the various alternatives to be considered in any project, it is rare to see the better choice. And for this reason almost every design process becomes a nettlesome quagmire to be unraveled with considerable pain by the organizing mind of the designer.
But there is hope. There are guiding principles, even if they are not precise.

Good designs are simple, scalable, practical and affordable.

A simple design requires fewer parts, streamlined procedures, and straightforward elements which are reused in as many places as possible. The design contains no unnecessary complexity or unneeded technology.

A good design can be expanded or contracted to a reasonable extent without having to start over. It could be bigger or smaller, faster or slower, louder or softer, and so forth, without a fundamental need to rearrange the basic design concept. Designs which can not be scaled are houses of cards ready to collapse when the slightest thing needs to change.

A good design is practical. There are no superfluous features. Everything works as expected by the end-user. A good design does what it is supposed to do, and nothing more. There are no features included that will not be used.

A good design costs less to develop, less to produce, less for customer service, and less for technical support. Every dime spent to build and test and ship the product is well spent. Every dime spent by the customer pays only for the intended value.

Unfortunately, it usually takes a disaster to learn good design practice.

In the early 1970's, General Motors introduced the Chevy Vega, a four cylinder economy car intended to compete with the Ford Pinto and the Japanese imports. But instead of building a simple car using standard technology, GM decided to design a brand-new aluminum engine -- in theory to save weight and increase gas mileage. The engine program, however, was delayed by a myriad of technical problems because nobody at GM had ever built a production aluminum powerplant before. When the Vega finally did roll out -- under intense pressure from senior management -- they were a nightmare of leaks, blown head gaskets, poor gas mileage, and endless costly recalls. Was an aluminum engine required? No. Did the company meet its goal of building a low-cost vehicle? No, not in terms of the total cost.

The human ego is a formidable impediment to clear thinking, and sometimes more than one disaster is needed to learn the lessens of good design practice. But learn we do, or else to the scrap heap of history do we trudge.