When our theater chain started having the slide projector advertising, we wanted to automate it to make the presentation a little nicer. We wanted there to be a 2 minute delay after the show ended, then the screen wash lights dim, the slide projector comes on, then the curtian opens. When it's time to start the show, we wanted the curtian to close, lights go up, slide projector stops, delay one minute, then start the feature.
Our tech started fooling around with those little microcontroller PC boards and a programming language called, "Relay Ladder Logic." You write (with a text file or actually draw a schematic) instructions that are written with a PC, then burn it into the EPROM inside the microcontroller IC, which then is plugged in the microcontroller PC board. The board can accept inputs (so external events are "given" to the instructions) and has relays for outputs. You buy a board based on how many things you want to control, ie: some boards have 8 inputs and 4 relays; 8 inputs and 8 relays, etc.
If you need to make a lot of custom interfaces, this might be a good way to go. I sure you could make a whole projection automation controller fairly easily, and have it do exactly what you want. A down side would be if you wanted to change the way the show ran, you would have to erase and reburn the IC.
Price is always a good reference (to see if it's worth looking into.) If I remember it was about $90 for a "8 input 4 output board", then you needed to buy a EPROM burner for the microprocessor IC you are using, another $70. Also, I think it was about $200 for the "Relay Ladder Logic" program/compiler.
Some people may not make enough interfaces (or make only very simple ones) to make this investment worthwhile. Don't forget that there's also a learning curve to understand "Relay Ladder Logic."
But it worked out very well for us. We wanted these slide projector interfaces to work with several different kinds of (existing) regular projector automations. Being software configurable, we made one "hardware" interface and software-adjusted each one at the theater to the type of dimmer (electronic or motor-driven,) the time it takes for the curtian to close (so the lights don't come up until it's fully closed,) etc.....
PS: Sorry for the long posting, but I thought some people may never have heard of microcontrollers or "Relay Ladder Logic." So I wanted to give more than just a quick explanation..

Good luck with your designs....
Aaron
One of my "designs" expanded on the programmable nature of Rank Mark IV's, without the trouble of diode pins, and utilized a different matrixing scheme. A technician friend of mine said I was crazy and that it wouldn't work (it used diodes, but not in the pins). He said it'd be too much trouble to make, and nobody would buy it. Fast-forward 20 years to this tech, now head of a major theatre supply house, who is building his own elaborate, programmable automations.
Guess whose idea that wouldn't work, magically became "okay" for him to use?

Seriously... At M.H. College, our proj. has virtually no automation.
The "Start" button is a 4p/2t relay set up in "latching" configuartion. There's a
"Kill" switch that has to be on, too. When you press the start button, the coil of the relay is energized and the 1st contact holds power to the coil. If the failsafe drops, the power is cut out. The other contacts of the coil power the blower, proj. motor, lamp, etc. You used to have to turn the sound on by hand and dim the lights manually but I've fixed that.
The failsafe has proximity-type cue detectors. It took me a while to figure out how to wire them, but now I've got a 'relay box' hooked up to them. One N/O contact per cue. Two of them are now used to turn the sould on and off.
I want to hook up the lights to the 3rd cue but that's going to take some work. The house dimmers are Strand System-6. It's a 0-10v system with 'fixed' controls. I'd have to wire in to the control panel on the wall to make the proj. control the dimmer. I called Strand Lighting and all they wanted to do was sell me a $5,000 'retrofit'! All I wanted was the stupid wiring diag.! ARRRGH!
I WAS going to hook the C/O dowser into the same cues that turn the sound on and off but (A) it would cost $500+ to get the parts and (B) I'd rather not tempt people into pushing the start button and walking away... If they have to open and close the 'lamphouse' dowser by hand, they're forced to pay attention to the picture. (At least for 1 minute)
Moral: I've gotten used to having little or no automation. (Although I'd LIKE to have something more)
Sometimes I think I even like it better this way.
PS: As far as pogrammable automations go, has anybody ever thought of building one out of a "Basic Stamp"? (or similar)
Jeez! All proj. automation is, anyway, is a scaled-down industrial process controller! What about an off-the-shelf P.L.C.?
Yes, you could do that. They are progammable, so you would have to wire it, then program it.
As I said, it's not a 'giveawawy;' there's a learning curve. But I think it's the best way to go if you need to make many different kinds of controllers.
For a description, try: http://www.relayladder.com/index.html
http://www.manufacturing.net/magazine/ce/archives/1998/ctl0402.98/04ebas.htm
hardware/software: http://www.entertron.com/controllers.htm
1. Thread the projector
2. Take out one lens (by hand) and put in the other one.
3. Take out the aperature plate (also by hand) and put in the other one
4. Turn on the projector with a switch on the pedestal
5. Turn down the lights (by hand)
6. Turn down the sound using the master volume control on the processor (by hand)
7. Switch the proeccessor to stereo srnd.
8. Watch for the count down to end
9. Turn up the sound and open the dowser
10. Check frame and focus
At the end of the show:
1. Run up to the booth and turn up the house lights half way
2. At the end of the credits, turn all the lights back up and close the dowser.
It may be a little more work, but it never malfunctions. You also never have a problem with other people leaving thier foil cues on. 

I have run a few different kinds of automation and, in the long run, i have found them to be more trouble than they are worth. Levers, switches and buttons are easier to use, more reliable, and when they go wrong, its easy to diagnose and fix the problem.