As I was wandering along a new-to-me trail today, I got thinking about some of the things I do while collecting data for NW Trails.
One of the biggest 'problems' I've seen is trying to get all the trails in an area - limited time vs. unlimited trails (it seems in some areas...). What I started doing is to walk 20-50 feet down the side trail and then back to the 'main' trail. This leaves a 'stub' on the track, Jon has been nice enough to leave those in the DB so it's easy to see where new data needs to be collected. When I'm working an area I've also made waypoints in MapSource of these stubs so they are easy to see on the small screen.
The other advantage of this, is others can see these and if they are in the area can map those trails. I tend to believe others try to do complete coverage also, but I've learned that's not always the case (and even I've missed a trail or two

).
Another thing that I do, is to mark waypoints in MapSource to leave notes to Jon about something special there - trails that don't touch, but the tracks do (darn EPE!) or 'missing' trails (shown but not in the field).
In the field, I turn the track on and off to 'record' changes in trails (such as, leaving paved trail for dirt). That way when I'm editing the tracks (to specify the type of trail/road, etc.) it's easy for me to remember where such changes happened. I also make notes on my PDA of trail names (I also have my little book of cache hide sites, but those I
don't share with Jon!) or use a waypoint.