2016-02-22

Get wrong to get right

You need to get it wrong to get right. For example, instead of creating 3 design "options" in 4 days and labouring over the details, how can you make 100 design iterations in 20 minute and flush out all the bad ideas and deadends early on? Don't ask "how do I get the design right?" Instead, ask "How can I make 100 mistakes in 20 minutes and clear the way for good things to happen?? And the best way to do this is to be very clear on what kind of mistakes should happen when. Too many mistakes at the same time will make it hard to judge what the right move should be. The hypothesis is that you should be very intentional in what you're trying to get wrong. Using techniques from matte painting, here's a hypothesis for the PAL project: The first mistake to make is in organization. Make every space too small and too big and dead and wasteful so that they can eventually be right: the right proportions, the right flow, the right light. The dimensions to evaluate are 2 dimensional. The next mistake to make is in massing. How is the organization going to be shaped? Only think about form and value relationships. Don't waste time thinking about materials yet. Draw fast and let it be ugly, clunky and awkward. Diagrammatically work the plans/sections or types/assemblies with the drawing so you know that it can still work. Eventually, a big grid of doodles will emerge and you can start to sort and distill into a single image or two that work. The doodle is mixing plan and section. 3.5D of information: X/Y/Z and values. IF you get stuck, just be systematic about the permutations and combinations you can make. This is a mixture of a line and a tonal thumbnail. If you've made it this far, congrats. You are now ready to make more mistakes! Its time to get materials and details wrong. Focus on making the materials as ugly and clash as possible. Try pinks, blues, reflectives and coloured lights. Don't worry. Try textures and smears and overlays. At this point, you should be having so much fun that there should be two massing perspectives to mess up! This is a colour study. Finally, focus your efforts on getting lighting as bad as possible. After this only are you ready to develop it in 3D.