My inner geek wants my training to be linear. For example, I want a step-by-step training program that I flawless execute with predictable sends as a result. My linear seeking behavior extends to my training facilities. I want a constant slab wall adjacent to a constant vertical wall followed by a constant off vertical, ... etc. The training area would stop short of a constant roof because that is just stupid. There would be none of those tricky transitions. I could perfectly train my movements and tactics in isolation. Later, I would bring those separate elements together.
I haven't (yet) built the full scale "Climbing Lab." I'm forced to train in the current real-world which includes transitions. Those nonlinear elements are either a crux or a rest, despite the best routesetting. My inner geek recoils at the "non-optimality" of the situation. My inner realist realizes there is an opportunity to train the real-world climbing elements of pacing, recovery, and jessery (Yes, jessery is a real-world climbing skill that should be trained). In the end, those elements will contribute more to sending than the perfect training facility.