A good science fiction story should be able to predict not the automobile but the traffic jam.
I perceived quite early that I was a reader, and most of the people I came into contact with were not. It made a barrier. What they wanted to talk about were things they had eaten, touched, or done. What I wanted to talk about was what I had read.
Stories where the author has known very little, but run a computer program that tells him how to construct a planet, and looked up specific things about rocketry and so on, really suck.
Stephen Hawking said he spent most of his first couple of years at Cambridge reading science fiction (and I believe that, because his grades weren't all that great).
If you don't care about science enough to be interested in it on its own, you shouldn't try to write hard science fiction. You can write like Ray Bradbury and Harlan Ellison as much as you want.
That's the method: restructure the world we live in in some way, then see what happens.
I don't think the scientific method and the science fictional method are really analogous. The thing about them is that neither is really practiced very much, at least not consciously. But the fact that they are methodical does relate them.
