| > We apply the Occam's razor here. Occam's razor is ... flawed. "Lazy" if I'm not trying to be polite. Its a tool you use to keep from going mad, but it doesn't provide correct answers, just _sometimes_ statistically correct ones. Sure fine, you admit its not a very good metric, I do disagree here. > You are not accounting the entire philosophy of test driven development, where one specifies functionality as tests upfront, and later writes code to pass those tests. Some of that philosophy is flawed too. Your anecdotes are interesting, I would like to see a comparison of the difference between the programs and your synthesized versions. But back to the philosophy bit - if you specify your functionality in your tests, you are still writing the code in test form. It doesn't matter if you translate the test back to code manually or automatically, you're still employing some meta language to code your program. Saying "if this do that" in a test and then writing code with an if/else loop is not impressive, or useful. To me, the philosophy of the test is to verify functionality, not to completely specify it. Going backwards to code, and just making it very easy to do so, you've either written your program in test form already, or you've produced some swiss cheese code. |