Tufte's Signal to Noise and the Additive Paradox

Signal to Noise is an engineering term used to describe static mathematically. It is the ratio of intended broadcast information (sound or video) against the unintended (static/garbage) information actually going out on the signal.

This is a term Tufte uses to analyse graphics with too much chrome; if the element in a design is not adding to the design it is by definition taking away from its informational value. This includes stuff like 3D bar charts, font games, superfluous use of color, background patterns, etc. In short, ornament.

Ornament is intended to make a design pretty and interesting. It is like the Chihuahua (sp?) in the Taco Bell ads. After Taco Bell ran their campaign for a while they did impact analysis and found that while they had not increased their market share considerably, the vast sums they invested did give a real shot in the arm to the sale of Chihuahuas.

The presumption behind ornament is that the content you are presenting is boring. Or, in short, that your audience is made of of children that need to be entertained. If that is true, then you probably are understimating either your audience or your subject matter, or both.

The Additive Paradox

Even useful information (signal) fights with other useful information when placed in the same context. This is the additive paradox. The more information you put in a single view, the more difficult it is to focus on any one piece of information. The antidote too the additive paradox is to pass your information in manageable chunks. How many chunks needs to be adjusted to the audience and the subject matter.

It always boils down to: "Make every word Count." And if a picture is worth a thousand words, make sure they are a thousand helpful words.