Lying specs

I received an email from a reader on a rant about lying specs. How manufacturers simply lie about the performance of their products. In calming the person down by explaining manufacturers generally don't print falsehoods—at least no manufacturer I have ever known—I pointed out that while only a rare few are lying there are specs that qualify as misleading by careful selection of what's being measured or obfuscation by lots of technobabble. Part of the problem for manufacturers is trying to decide what to publish for specifications. Too much and we run the risk of confusion. Too little and we're not doing our job. On the one hand, there are the essentials to decide if a product is going to work: input impedance, sensitivity, gain, wattage, power consumption, etc. On the other hand, there are specs whose sole purpose is for comparison with competitors: distortion, frequency response, noise, etc. And lastly, there are specs and measurements like square wave and pulse response that mean a lot to an engineer but little to most prospective buyers. Published specs rarely lie though the story they tell can be anything from a revealing novel to a fairy tale.