In the last few month I’ve been dealing with a field called “Evidence Based Medicine“. The idea about this concept is that clinical decisions should be made based on evidences for examples clinical trials, clinical guide lines and protocols.
This approach claims that patients will get a better treatment when physicians trust less on their own subjective history (which usually biased) , hunches and intuitions and more on facts.
Maybe the best definition is “an effort which helps medicine become a science” (so we will get less Dr. House-like decision, but this is a topic to another post 🙂 )
It is true that “if all you have is a hammer, everything looks like a nail”, and in my case I start thinking of development. Too much code was written using non scientific (or more accurate engineering) methods.
TDD programming is closest paradigm I know that can be tagged as “Evidence Based Programming“.
The red-green-refactor steps are your evidences – they can help you claim that code will produces the correct output.
BTW, When you start comparing physicians to developers you get interesting results for example: you find that both have large ego, both trust massively their own experience, both must keep updated with the latest technology. On the other hand a developer that does not follow the best practices will (mostly) not kill you while a physician will 🙂