>>2166Perhaps I ought to have made this clear earlier, but I am a complete and utter hack, so most of my stories are a few decent scenes cobbled together to form a semi-coherent narrative.
Kubrick talked about 'non-submersible units': these are fundamental story-pieces, the irreducible core of a narrative when all the non essential 'padding' has been stripped away. Brian Adliss: "[Stanley] had a contempt for narrative, I was hooked on narrative. But he said to me: forget it, all you need for a movie is 6 or 8 non-submersible units."
I've no idea if this method works outside of film, but that is my approach to writing.
I am even guilty of J. J. Abrams' 'mystery box' approach: I come up with a mystery for the characters to look into/investigate, and then make it up as I write. As long as the reader can follow the logic in where the story and investigation goes, and I ground the story enough, it could work. Hopefully. I include a lot of details surrounding the case, details that help sell the idea that this is a real case, and the investigators know what they are doing and follow the leads, but in actuality I have no idea where it is going, or if there is anything of importance to a particular clue.
Some clues may end up becoming redherrings, while other may end up being important. If I, as the writer, have no real idea where it is going, it forces me to be creative, makes me think on how I can connect these clues in a logical and interesting way.
In my 'dramatis personæ' list I make sure to include their relationships to the other characters if that is important - relative, love interest, acquaintances, enemies, etc.
If you are lucky the bio you come up with for characters can prove to be important down the line, how characters interact and think about each other.
I'm betting people will overlook most of my stories' faults as long as there are a few scenes that really work, and the characters are engaging.