Something that might have been interesting (information content of life forms vs. programs) has been wasted by equating base pairs with lines of code. This is t

Something that might have been interesting (information content of life forms vs. programs) has been wasted by equating base pairs with lines of code. This is totally arbitrary. Unfortunately I don't know of a specific better measure, but it might consider the used and potential information content in the structures encoded. Anybody know of such a metric for source code complexity?