I'm coming from FCP where I ingested AVCHD footage straight from my camera, and only brought in the footage that I needed using I and O points during the ingesting process. With CS6 I see lots of people copying their entire cards to their hard drives and from there ingesting it into CS6. This seems it could quickly eat up a lot of hard drive space with copying large clips when you only need small segments. Is there a reason why it seems to be generally recommended to do it this way rather than just ingesting straight from the camera the parts you need?
Ingesting from a camera is real slow, because of the slow speed of the cards used. Second, you need the complete directory structure for reliable operation. Third, you need reliable backup of your source material. Last, HDD's are cheap, so what is a couple of 100 GB's of disk space.
FCP transcodes the media upon import. PP doesn't. The larger file sizes of ProRes will likely eat up a lot more space than the original files.