In 2000 I attended a two week course on Geophysical & Environmental Fluid Dynamics (GEFD) organised by DAMPT in Cambridge. I was inspired by it, because we were taught fluid dynamics using the mathematics, but then given some experimental work to test it, eg: dropping blobs of ink into spinning tanks..etc. We exercised outside to balance all the academic work, went punting on the Cam and were invited to play our musical instruments in an evening concert in Isaac Newton's rooms. I have not forgotten this lesson in balance, which is not to say I've completely lived up to it since! Anyway, I volunteered to play my flute in the concert. The fellow before me stood up and played part of a piano concerto from memory. Then I stood up and played a simple folk song on the flute, and felt inadequate by comparison!
This is one case where I can safely point to myself and accuse myself of the error of judgement that I think modern physics often makes. Complexity requires a prodigious memory but does not necessarily make something better. Sometimes when the difference in music or theories can't be easily measured or understood, people rely on something more easily measurable: complexity. They assume that what they can't understand is impressive, whereas in proven science it has been found delightfully that it is usually the opposite: the simple ideas are often true (Ockham's Razor).
I have been told that MiHsC is 'too' simple, but I do not agree. Nature's laws often are simple, because only simple balances last (see "underlying randomness"). So I think that the criteria for a good theory in order of decreasing importance are: 1) it predicts the observations well, 2) it is simple and 3) it is self-consistent. In modern times this order has been reversed. String theory is self-consistent, apparently, but it is not simple and it is not predictive.
I have read a lot about the work of the scientific greats, and have worked rather on the applied edge of physics and avoided the pressure to conform, so I have designed MiHsC using the old-style criteria 1 and 2. Its funny how the same old "look at messy nature" and "nullius in verba" empirical method keeps cropping up in the productive parts of science, only to be neglected when people decide they can progress by thought alone.
Many a person fails to become a thinker,
because his memory is too good.