You can learn an amazing amount about how to write science fiction from reading Robert A. Heinlein. But his most often quoted bit of writing advice won't help you at all. If anything, it'll mess you up.
I've lost count of how many times I've seen people quoting Heinlein's rules for writing:
1.) You must write.
2.) You must finish what you write.
3.) You must refrain from rewriting, except to editorial order.
4.) You must put the work on the market.
5.) You must keep the work on the market until it is sold.
I don't actually think anybody could disagree with any of these rules, except for #3, (and maybe #5, but that's another topic.) In particular, I think that rule #3 was designed for an earlier era, but people also tend to misinterpret it pretty often.
What's wrong with rule #3? Patricia C. Wrede has a pretty good explanation:
Rule #3 doesn't fit with the other four rules. "Sit down and write," "Finish it," "Send it out," and "Keep sending it out" don't prescribe any part of how one goes about writing and submitting; they only say that you must do it. The writer is free to find or develop whatever process works for their particular personality. These four rules are about procedures, and business procedures at that (which means they don't vary much from writer to writer).
"Don't edit unless an editor asks you to," on the other hand, is about process. Process varies wildly from writer to writer; what works for one, won't work for someone else. This rule, in particular, will work fine for those writers who, like Heinlein, can produce an almost-perfect first draft (and/or those few who still have professional editors they can rely on to ask for in-dept revisions when needed). It will work not at all for those writers whose first draft is over- or under- written, or which is otherwise deeply flawed.
Late in his career, Heinlein himself admitted that he did, in fact, revise/rewrite his work before sending it out, but he never, to the best of my knowledge, explained why he had laid down this particular rule.
I have a couple of theories about that.
The first possibility is that Heinlein was of the school of thought that felt that "good enough" was all that was necessary, ever. Since he began making a living from writing in the days when you could support yourself selling short stories to the magazines if you were prolific enough, and since "prolific enough" often involved not having time to polish and revise much (if at all), this attitude would be understandable in him. It still begs the question of how one gets to "good enough" without revising, though, especially if one tends to flawed first drafts.
My other thought, and the one I think is more likely, is that Mr. Heinlein had run into a disproportionate number of extreme revisers (those writers who polish and polish and polish, ten or twenty or fifty drafts' worth, and still won't send the story out because "It's not finished; I have to go over it one more time"). Since he himself did not tend to excessive revision, he drew the conclusion that this was a common beginner mistake (it is, but it isn't exclusively a beginner problem) which needed to be addressed. Hence the rule.
I would go a bit further than Wrede does here, and say that the writer who can produce an almost-perfect first draft is one in a million — and pretty much the way you get to be that sort of writer is by rewriting your work, over and over, until you see what you've been doing wrong and learn not to do it wrong any more.
Writing an endless series of first drafts will never make you the kind of writer whose first drafts are perfect. Only third and fourth, and sometimes tenth and eleventh, drafts will help you break out of some very bad habits that you probably don't even know you're stuck with. Because if you only ever produce first drafts, you'll never see the problems that you keep re-creating — only rewriting will show you those.
There's another saying that people toss around, too — "all good writing is re-writing," which is as much an exaggeration or distortion as "you must refrain from rewriting." Still, if I had to choose between "refrain from rewriting" and "all good writing is rewriting," I'd live by the latter rather than the former. It's often true that the stuff that makes a story really work as a story, rather than just as a collection of events or as a plot with occasional splashes of color, happens in rewriting.
But it's definitely true that editors are a lot less patient, and a lot less willing to take a flawed mess and refine it into greatness, than they were back in the late 1940s, when Heinlein wrote that essay. I remember my jaw dropped when I was reading Again, Dangerous Visions, the classic 1972 anthology edited by Harlan Ellison, and found the introduction to Piers Anthony's story, in which Ellison quotes Anthony as saying that he hadn't really gotten as deep into the story or the characters in the draft that he sent to Ellison, because he wanted to wait and see whether Ellison wanted the story before fully committing to "open a vein." (I can't find my copy of ADV as I write this, but I'm pretty sure that's what he says.) No editor nowadays would take a weak draft with the assurance that the writer would "open a vein" and make it really good if the editor accepts it.
But of course, it's also extremely likely that Heinlein is overstating his case, and he doesn't really mean that you should send out your first drafts into the world. Rather, he means that if you have a third or fourth draft that is in really good shape, there's no point in fussing over it endlessly to decide whether a comma should really be a semicolon. Which is advice that pretty much everybody can agree with.
Because you have to remember that just as Heinlein's advice to keep persevering is woven into the unofficial religion of science fiction writers, so too is the idea that your writing is going to suck at first. Here's Isaac Asimov, from an essay reprinted in his book Asimov on Science Fiction:
Since writing is itself a schooling, you can't very well expect to sell the first story you write. (Yes, I know Bob Heinlein did it, but he was Bob Heinlein. You are only you.) But then, why should that discourage you? After you finished the first grade at school, you weren't through, were you? You went on to the second grade, then the third, then the fourth, and so on. If each story you write is one more step in your literary education, a rejection shouldn't matter. [Editors don't reject writers; they reject pieces of paper that have been typed on. Ed.] The next story will be better, and the next one after that still better, and eventually But then why bother to submit the stories? If you don't, how can you possibly know when you graduate? After all, you don't know which story you'll sell. You might even sell the first. You almost certainly won't, but you just might. Of course, even after you sell a story, you may fail to place the next dozen, but having done it once, it is quite likely that you will eventually do it again, if you persevere. But what if you write and write and write and you don't seem to be getting any better and all you collect are printed rejection slips? Once again, it may be that you are not a writer and will have to settle for a lesser post such as that of chief justice of the Supreme Court.
And actually, even Heinlein had half a dozen rejections in a row after that first story, according to Robert James' afterword of his book For Us, The Living.
So if your writing is bound to suck at first — unless you're a super-genius who just starts writing perfect prose as soon as you master language — then the question becomes, how do you get better? And most people nowadays would include revising your work, in some cases over and over until you see what you're doing wrong, as part of the process. Even if editors were still willing to provide detailed feedback on the stories they reject, at least part of what you'll learn from studying their responses is how to please one editor. You can also join a writers' group or workshop — but they'll also encourage you to rewrite your work, until you've gotten the small stuff right, as well as the huge glaring plot and character problems that you may be unaware of.
The other benefit of rewriting, of course, is that you can have a lot more freedom in your drafts if you know that you're going to fix them later. Sometimes you can make some intuitive leaps and then figure them out afterwards, or you can push the story forwards and then fill in the little character moments afterwards.
Want a second opinion? Here's writer extraordinaire Gene Wolfe, talking about revision in a fantastic interview with Clarkesworld:
What do you enjoy about writing?
Primarily revision. I like to polish and tidy up, like trying to make good stuff great. First drafts are work. Are fighting, really. I know where I'm going, but I know too that I mustn't get there too fast.
What happens on the way? How to make the reader see, smell, hear, and feel it? ...
Does there ever come a point when you have to accept that a particular piece is no longer worth fighting with or for?
There comes a point at which I'm no longer sure that what I'm doing is improving the piece. That's when I stop working on it and send it in. Usually – not always, but usually – I get there after four drafts. A fifth draft may find me reverting to the second or the third, and that's a bad sign.
So read Robert Heinlein's novels and stories for their brilliant story-telling, and learn from his example. But if people go around quoting rule #3 at you, tell them that rewriting is how you're going to make your work stand out from the pack. Because in this day and age, you have to assume that the "editorial order" to rewrite your work is already in effect before you even start putting it out there.