The OP laid out some differences between two systems, one of which is science as done today, and suggested that the other system was measurably more efficient at converting resources into useful beliefs.

I would also say that, while the other system is indeed "more efficient at converting resources into useful beliefs", it's not so very different from the original system, both in terms of structure and in terms of performance. Thus, unlike (I think) Luke, I see no particular burning need to drop everything we're doing and begin the conversion process.