I think there are a bunch of fields, like medicine, medical research (except doing some software simulations of stuff), finance (the process management, not the

I think there are a bunch of fields, like medicine, medical research (except doing some software simulations of stuff), finance (the process management, not the calculation part), politics and law etc. that are inherently hostile to revolution by software - most competent programmers run away screaming after contact with any such domain and only the dumb ones remain, churning dumb software that doesn't actually solve the problems. And the few good ones that stay become dumbed down and only work at ~40% of their iq and productivity because their minds basically choke on the sheer amount of disorganized domain specific knowledge they have to absorb. I think if we were to make an analogy of "information as alcohol", a programmer's mind is like an installation that only works well with very "distilled" information (say "vodka" and "whisky" :) ) and may work ok with "wine", but it completely sluggs down if you give it "beer" (completely undistilled information). In some domains (medicine is one that I know of), information is so undistilled, diluted and corrupted that it's basically the analogue of "beer with piss in it and rats swimming in the kegs" ("the rats" are my analogue for the "mud minded" people that "just don't get it and never will" and prevent smart geeks from actually doing their jobs)... that's why I wouldn't touch it with a 100 foot pol and I imagine lots of people feels the same about these areas. I imagine this part of finance is a similar "beer pool".