Self-explaining computation

Behind every value lies a computation struggling to get out. That’s the idea behind what I call the explodable user interface. (Forget wearable. Explodable.) By “explodable” what I have in mind is the ability to pick any part of an application’s GUI that you happen to be interested in and interactively “unpack” it into a story that explains how it was computed. It should be as though code lurks behind everything you can see, all the way back to the data sources your application works with. In Lisp you can explode an atom into its constituent characters, but the relationship between a Lisp atom and its characters has no computational content to speak of. In a typical application, the relationship between a value and its parts is non-trivial. If it were otherwise, you wouldn’t call it an “application”: you’d call it “a bunch of data”. Whenever this non-trivial structure is present, you should be able to browse into it in order to understand it or change it.

http://dynamicaspects.org/blog/2012/06/10/self-explaining-computation–2/