I use the Ultimate-S script pretty heavily.

I prep my footage one tape at a time.

I load the footage from a single tape onto the timeline, I trim the garbage out of the starts and stops of all the footage with the ripple edit feature turned on so as not to create gaps between clips.

I select the correct audio track on all the clips. On B-Roll, usually I am set up so that the shotgun mic is going to both channels, one a few DB down from the other. That way I have a lower level recording for safety when the levels are peaked. So usually I select all the clips for instance and then just choose the left channel. Any clips where the audio is peaked I switch over to the lower level right channel audio. If there are clips where there is a wireless mic audio track, I select that.

I then select any clips where there is dialog I want to emphasize and run the function in Ultimate_S that normalizes all the selected clips. I don't normalize audio when it is just ambience. Only on the clips where the talking is something I want to emphasize.

Then I insert text events in before important clips which describe the footage which follows. I might insert a title that says "wedding vows", "Aunt May sings" or "Best Man's toast", etc. If it is an interview, I will have edited the footage down to just the answers and I will insert text events which state the questions being answered.

I run the Ultimate-S script which adds markers to all the events. Then I delete any unnecessary markers. For instance, after running the script there will be a marker before and after the text event which describes a section of footage. I will delete the second marker. Sometimes there will be several edits in the answer to a question. I don't need markers at all these events.

At this point I render the timeline to a clip with the tape name plus something descriptive. It might be something like "Tape_3_backpack_distribution_in_Otavola" or something like that. The render is a smart-render, so all the footage is still the same video-wise and the audio is ready to go for the most part.

When the render is done, I drag the rendered clip onto the timeline so that it draws out the audio waveform, then undo the add.

When I am done I delete all the orignal clips to save disc space.

What I end up with is a subdirectory of clips relating to a given project. These will be large clips with names like "Tape_1_arrival_in_Ecuador"..."Tape_13_interview_with_Maria", etc.

When you load any of these clips into the trimmer, you see a bunch of markers. When you click on the markers, often you will see a title which describes the footage which follows. If it is an interview, each marker that you click on will show a text frame which states the question being answered by the following footage. Some of the markers are named with notes. It's like having notes built into your footage.

Once the footage is prepped like this it is incredibly easy to find what you need in the trimmer. Keep in mind that with documentary style work, you need to keep a huge amount of footage around and how quickly you can find what you need is pretty important.

This workflow solves a couple of Vegas problems.

One is that Vegas doesn't seem to play well with too large a number of long GOP clips. This approach keeps the number of clips quite low. Ten tapes is ten clips. These low numbers of clips keep Vegas quick and responsive.

Two is that Vegas seems to preview video most smoothly when the audio is playing back in stereo. I don't know why, but when you select just a single audio channel, Vegas's preview seems to become much less efficient. During the smart-render, all the mono audio gets doubled for regular stereo playback and it seems to work much better that way, at least on my Core2Duo.

Three is that as many here have noted, Vegas doesn't have things like "bins" or clips that are references to other clips. Having markers and regions embedded into the clips makes the Vegas way of working much more organized.

Four is that for quick renders without color correction, I can go all the way to Bluray or m2t master on my PS3 with incredibly quick smart-renders. I smart-render to a little 8B thumbdrive and pop it in one of the USB ports on my PS3. It is almost immediate gratification to see an edit on the HDTV in my living room so quickly. These smart-renders look like original footage because, well, they are original footage. Final renders with color correction, Mercali and Neat Video noise reduction take longer, but I use those sparingly.

Is this workflow for everyone? Probably not. But it works well for me and the type of project I work on.