Using StarNet to compute a filed Record of Survey

I used WinCMM to compute the as-built outline of a building once. It was a huge, oddly-shaped condo high rise with hundreds of little jogs, all measured to the nearest tenth. I used the corners the crew located with the instrument as control, assigning them error estimates based on the error ellipses calculated for them by the as-built survey control network, and entered all the field measurements as traverse data. It resulted in pretty big uncertainties in the line bearings and distances, as I recall, but there is no way I could have gotten it to close otherwise. I would probably do the exact same thing now with Star*Net.