|
Even though I object to being called an acolyte for Dick, I'll address your question from my personal point of view.
I love knowledge, I suppose most people here love it too. When I was younger I thought science was marvelous, it answered so many questions and it promised so much more. But I have reached a point in my life where science started to appear as a disappointment, the promises I was expecting to be fulfilled never materialized. I was quite surprised that by age 30 or something there was little of any value in science for me to learn. I am not disgusted with science, I realized there are too many things that cannot be known at this point and it's OK to speculate and make mistakes.
The thing that bothers me has nothing to do with science in particular, it has to do with people's attutide regarding knowledge in general. You see, I work as a software analyst, I spend forty hours a week with logic and, believe me, those one and zeroes have more to teach you than you can possibly imagine. I used to have a friend who kept a voodoo doll by his computer, because he said the machine seemed to be possessed by the devil sometimes. He couldn't make any sense of it despite the certainty that whatever the machine was doing was perfectly logical.
What is the most important lesson I learned about logic? Two things actually. First, I learned that the habit of making careless assumptions are one of the greatest evils on this planet. Every sloppy assumption you make when writing software is bound to give you a headache sooner or later. Every single one of them, it's pretty amazing. I developed a terror about assumptions that is almost pathological, but at the same time I seldom get complaints about bugs in my code. So that's logics lesson one.
Lesson two is about definitions. An outsider would think a software professional has the most trouble implementing what the users ask him to do, right? That's what one would naturally expect, and that's what I expected when I started on this business. To my amazement, it didn't take long for me to learn that the most difficult job in the computer business is to get from the users what on earth is it that they want. Incredible as it seems, nobody knows what they want.
I usually spend half my time trying to extract from users what they expect any new application to do, about twenty percent actually writing code, and the other thirty percent fixing the software because the users realize, after the fact, that what they asked was not exactly what they wanted. Why do I spend so little time writing code? Simply because when you have the problem properly defined, down to its smallest details, the software almost writes itself out.
What do I get from Dick's ideas? Two things essentially: avoid as many assumptions as you can, and be very careful about those you can't. Define your problem correctly, don't miss any details. Then watch the solution drop out by itself.
I don't think I need to say more, do I? |
