"When you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind."
Lord Kelvin (Sir William Thomson)
British Physicist and Engineer
So what does a nineteenth century British chemist have to do with productivity? Everything! Lord Kelvin succinctly put into the words the very essence of understanding productivity: measurement. Consider how often we are bombarded with pitches about how our lives will be better if we use this soap, or how we can change the world by reading this book, or how much more productive our work life will be if we buy this software? But you need to ask yourself, "Well, how does this guy know what he is talking about?" Good question.
In EDA, productivity is often synonymous with the speed of the tool. Certainly, if the tool does its job faster, we will eke out a few more minutes in the day. What happens though when the speed of the tool pales in comparison to that of the human element, as it does when we are looking at custom and analog engineering tasks? Custom design is not blessed with some of the high-speed tools available to our digital brethren (i.e., synthesis, hardware acceleration, event-driven simulation, behavioral modeling, etc.). Most custom tasks are manual in nature, with the analog engineer relying on "assistance" in developing and verifying the design. Understanding this human element is the most difficult of all because getting the data is time consuming and sometimes impossible without a constant monitoring procedure. So what to do? Well, there are three tactics to consider.
1. A scribe sitting next to the engineer
Although time consuming, this can be the most accurate way to measure the engineer's exact tasks. If you are part of a usability test, most likely you will be given a task and then monitored while you complete it. This observation is often done via video or a two-way mirror. An interesting technique often employed is to videotape the user's face. Our faces reveal much about what we are thinking, even if we are unable to verbalize our thoughts. While this technique is excellent in small settings, it is relatively intensive and can be expensive in terms of time, personnel required, and equipment setup. As such, the observation technique is best used to discover the "big" things that "everyone" might run into. Use it to make sure that the basic design flow and tool representations will have the greatest mass appeal. We aren't fine tuning, just making coarse adjustments to lead us down the right path when presented with a variety of good options. The drawback is that some of the individual "personalization" of analog design is lost by concentrating only on the big rocks.