The best art all looks effortless, and the best artists all make it seem like play. The trick is to take everything that looks like work and make it not look like work.
The first time I had this codified for me was, I am sorry to say, when I saw Amadeus, a work (both the movie and the play) I no longer think as highly of as I once did. Historical inaccuracy has nothing to do with it; it's that it's a facile story that takes forever to make what few points it has (which mostly amount to "jealousy is bad"). And on top of everything else, it reinforced the wretched pop-culture idea of creative genius as being lightning that just strikes randomly instead of power that's drawn together and tamed. It looks nice, to be sure, but so did my wife's eyes after retinal surgery.
Anyway, what I did get from Amadeus that proved useful later was the insight that for some of us, this stuff is play. It's where we want to hang out and just try stuff to see what happens. There's no moralizing of failure; there's rarely any "failure" to begin with, because we're just ... seeing what happens! The Mozart of that movie enjoyed what he did; there was no place he would rather be. (Unfortunately the movie has no idea what to do with this one genuine insight. But enough about that for now.)
"We were put on this earth to fart around," Kurt Vonnegut once said, "and don't let anyone tell you different." And one of the corollaries of "farting around" is not taking ourselves so terribly seriously. Kids playing made-up games take the game seriously, but not themselves. They don't care about looking ridiculous. And neither should we when we're in the business of creating.
Paul Krugman has talked about the use of models as a way to simplify a complex domain -- to shrink it down to the size of a toy so you can play with it. The model does not have to be complete or accurate; it just needs to be something you can wrap your head around conceptually. The play's the thing.
All this I want to separate from the process of learning from one's own work. To each thing there is a season, and all that; a time to fart around and a time to analyze. When you're messing around, just mess around, but when you're trying to figure out what in your work did or didn't hit its mark, do that completely too. The more hats you can learn to wear in your own work, the better.
Other Lives Of The Mind