When editing a multi-camera shoot for insertion into a larger project, my method in PP1 DV days was to import the clips as AVIs, cut it together as a standalone sequence, and then export a completed AVI which I simply dropped into the final project.
With HDV footage in CS3, I imported the footage as 1440X1080 mpegs by default, cut it, and then used AME to export it as a 1440X1080 mpg2, to be dropped into the master project. The footage was captured 'Lower Field First', so would I be best to export the mpg2 as 'Lower Field First' or 'Deinterlaced' for maximum quality...or does it not matter? I assume exporting it as 'Upper Field First' wouldn't be a bright idea. I'll be authoring the final product to DVD to start with, but I want to keep all the source footage as 1440X1080 for a Blu-Ray release in a few months.