The paper is based around observation Cook makes on the different factors that are involved in the creation of new digital music controllers.
Overall, his principles given toward the start of the paper gave me the impression that we were talking about very simplistic instruments; nothing unnecessarily fancy! (use of wires, hatred of programmability). I found the concept of human bandwidth interesting, that is, that some players have spare bandwidth while other don't, such as how a guitar player has his feet spare, whereas a drummer or pianist have nothing (unless they want to start bashing their head against something :P) He also had a valid point about how algorithms and controller design go hand in, each one inspiring the other.
Many digital instrument designs were presented, but my favourite had to be SPASM, a voice synthesizer. Capable of manipulating the voice, it could change the parameters and output in real-time, which I thought was really cool.