I've found the recent thread about sorting one's observing list to be pretty interesting. I've tangled with this a little bit myself.
Bob and Mark have pointed out ways of approaching this that work fairly well for them. Sorting by RA, Dec, or constellation are straightforward algorithms, and they've had good luck with these methods.
Matthew pointed out that there are 'traveling salesman' algorithms that might be applied, to do things like minimize hop distances. Turns out that astro planning software makes use of this. I use NGCView, which has its own proprietary "Observing Order" algorithm. I believe that SkyTools has something like this, too; Rashad might have some experience with this.
I haven't tested the usefulness of NGCView's 'Observing Order' yet, but it mostly seems to make sense. It does create a few weird, long hops from time to time, though. One other nice NGCView feature is "binning". You can create an observing list, sort it by 'Observing Order', and then bin it by SA2000 page or Uranometria page. This is pretty handy. I did this at Shingletown last year, and it was pretty helpful.
What I would really like to see is an "Azimuth Order" algorithm. Why? Because I look at DSOs with a dob. The big pain the neck for me is making changes in azimuth. Moving my Denver chair and accessory cart is a pain. Altitude changes are less of a big deal when I'm using my dob, especially if I exclude stuff above about 50 degrees altitude. I'd love to be able to sort the list so that I only make azimuth changes in one direction as I work through the list. Of course, other factors come in to play, such as the time taken on each object, the time of night, and whether or not I take breaks while working the list. Mark's approach of sorting an RA window by declination is probably a pretty close approximation to the solution I'm looking for.
I swear, I think I like this part of observing better than the observing itself - the "ergonomics" of it. I always want to know how I can make the process as easy and efficient as possible, so as to maximize time and energy spent looking at the object, and/or searching and destroying. I would call my approach to observing "Stop and smell the roses while engaged in your search and destroy mission." (Maybe just call it 'search and destroy the roses'? Hmm, that sounds a little brutal to me...)
Marek