My favorite part of that talk is where he highlights the value of creating to solve a class of problems vs. a unique instance. Build in a way that others can bu

My favorite part of that talk is where he highlights the value of creating to solve a class of problems vs. a unique instance. Build in a way that others can build off of your work. Recently, I built an automated Skinner box that has the ability to run most of the protocols of classical behavioral research. It allows the mouse to be head fixed, enabling brain readings while the experiment is running^. We could have built just what we needed, but instead, we are going to open source this general solution in a methods paper soon. I am unsure if this would have happened if I had not read Hamming's talk. ^The hope is that this will reduce the amount of this type of research done with monkeys, enable reductions of costs/time, and enhancing the ability to learn more due to the huge genetic catalogs for mice.