Perfect the first time:

As I write and edit my fictions, I submit chapters to a website for comment. A few fine people then rip my work apart, showing me all the places I wrote ambiguously, left something out, forgot to state the obvious, made my characters act irrationally, misused words and syntax, etc. (This is a very friendly and constructive process, we do it to each other.) Over and over I've submitted a piece, feeling that it's really good, only to be shocked by the length, depth and perception of the critiques I receive.

This week, something special happened, I think. I submitted a piece, and almost all the critiques were short. That tells me that either my critiquers are falling asleep, or that – hooray – there wasn't much to criticize. All of which reminds me of a much more interesting experience.

I started writing computer programs in 1961. Throughout my career, I have often been asked to write a very small program to do something useful. It occurred to me early on, that I ought to be able to write a perfect first draft of a small program, something that would compile and run perfectly, in my first try; no edits or corrections required.

Each time the need for one of these small programs arose, I tried hard to polish it off at one shot. Usually, I failed to get past the compiler for some stupid typo or other, and then there might be a little debugging to do. I've faced this challenge in assembly programming, and in many different software languages. I've succeeded a few times, but the sobering fact is that it took me more than twenty years to conquer this little challenge for the first time.