In astronomy, there are two types of telescope observing. "Classical" observing is what most people think an astronomer does -- she drives up the mountain, stays up all night in the same building as the telescope, and then drives down with a car full of pictures and data. This is the only type of observing I have done.
The other type of observing is known as "queue" observing. In this case, the astronomer stays at home, and the staff of the observatory does all of the observing. They may take data for several different astronomers in one night. An advantage of this type of observing is that, as the weather changes, different observing programs can be attempted. Some observing requires crystal clear weather and rock-steady skies, while other programs can be done even if we are looking through thin wispy clouds on a windy night. Queue scheduling searches for the program that can best use the existing conditions.
Another advantage of queue observing is that valuable science can be done all night long. In classical observing, it often happens that the object the astronomer is interested in moves through the sky and sets before dawn, leaving the astronomer with an hour or two in which she has to find something to do. In queue scheduling, when one person's object sets, someone else's is rising.
Some telescopes have to work in queue mode, like the Hubble Space Telescope. I can't drive on up to space to use the telescope -- its schedule has to be planned months in advance, and then replanned when each glitch messes up some observing.
There are disadvantages to queue scheduling. Since the astronomer is not there, the astronomer has to rely on other people to take the appropriate data. Sometimes in an observing run you point at the wrong object, perhaps you have bad coordinates, perhaps your finder charts (pictures of the sky you will be pointing at to help you find your target) are bad. In classical observing you pull up the internet, do a few minutes of searching, and figure out what is wrong. In queue observing, at best the observer gives up and moves on, at worst he or she takes pictures of the wrong thing, and uses up your time on a worthless object.
Queues are also less flexible. Suppose I am going to the telescope in January, and in December somebody "scoops" me by publishing data on the object I was going to observe. In classical mode, I just switch to a different object, but I still get my telescope time. In queue mode I can cancel my observations, but I can't put a new one in.
McDonald Observatory is partners in the Hobby-Eberly Telescope (HET), a large telescope with a mirror about 30 feet across in West Texas. Because of its unique design, the HET must be run in queue mode. It is set at a fixed angle, and so can only look at things 55 degrees above the horizon, giving it a donut-shaped view of the sky. As objects go across that donut (it takes about an hour), they can be observed. So, with lots of astronomers having lots of objects spread around the sky, something is always crossing the donut. But if I were to go and use the telescope myself, chances are good that, for much of the night, nothing I want to look at would be visible!
My co-workers and I were given 13 hours (about one night) of total time on the telescope this past week. Our observations will be placed in a queue to be observed sometime between January 1 and April 30, probably just as one hour here and one hour there. But I have to turn in all of my materials that the observers will need (finding charts, lists of targets, how long I want to look at each target, what camera I want to look at each target, and so on) by the end of next week. So, there's a lot of work to do in a short time to prepare. And, since this is my first time doing queue observing, I am hoping that I don't mess up!