Pieces of Advice
Working with light is quite a hard thing to do. Many of the problems we had were because of the difficult lighting conditions we worked in. So if you use light as an input and you have either light sensors or a camera tracking system you need to have a unique light condition; this will avoid resetting the threshold all the time, and, more important, it will let you understand the sensitiveness of the sensors you are using (each light sensor is different from the others in terms of accuracy in determinating the light change).
Wiring something like 18 photodetectors to the Arduino takes time, patience and precision. It is best not to leave it till the last minute. This means that you need to have a perfect light condition, and only then wire the whole thing up. Also, as Durrell Bishop suggested, it is best to first try 1 or 2 wired sensors and only then wire the whole thing up. Unfortunately we had to rush all this phase because of the lack of time.
The fake wall panels were quite easy to do, but it is a long procedure (it took us almost 2 days). The materials you need are: polypropylene panels, vinilic glue, acrilic colours, and fine wood shavings. Prepare all the polypropylene panels by grating them so as to have a coarse surface. Stir energically the wood shavings with water and glue and then add the basic colour for the panels. Cover the panels with the paste so as to have an irregular surface. Take some diluted paint and cover the panels with irregular layers of 2 or 3 different colours. Let the panels dry, sunlight is the best way.
Concerning the code. The light dots simulated with the projector have been controlled one by one within the code by simply copying and pasting a few lines of code. However, creating a class for the light dots and then making an array would have been the best solution. We went for the fast and dirty way just because we were running out of time.
Reflections
Surely, one thing we learned is that it is best to split tasks and work on parallel lines once that the brainstorming phase is over. On top of that, having some tests for whatever you do is not a waste of time but rather the contrary: more than once we had to fix something because we actually were gone too far. It is good to do things step by step in order to be able to come back to a previous stage where what you did was perfectly working.
Ideas are always useful, at whatever stage of the pre- and prototyping phases: never discard ideas as dead ends. We started with a location and some technological solutions which were quite far from what we eventually realised: nonetheless, it is only because of all the studies that those ideas brought that we could actually realise our final idea.
To prototype in an efficient way you must make sure of 3 things: having the right tools, having the right materials and last, but not least, having a versatile space, where you can easily remove obstacles or unnecessary items so as to rearrange it as rapidly as possible.