Academia reaches out to the common programmer and student all the time; they write books. Academic research papers are aimed at other researchers, i.e. people

Academia reaches out to the common programmer and student all the time; they write books. Academic research papers are aimed at other researchers, i.e. people who are up to speed with the concepts and terminology, if you or I don't understand the latest research on machine vision, then though luck, we just lack the necessary background, and that's discounting the fact that we are different audiences and seek for different content and form from each other. How often are we programmers interested in detailed proofs? Or even sketch proofs? If you decide to helpfully include code samples and they are full of boost::this or ApacheFooXMLFactoryThat, would that be of help to an academic who only implements his algorithms in isolation and doesn't know of massive APIs, or to researchers 20 years from now?