Deconstructing Programming Languages for Reconstruction
I had a good discussion with Prof. Mark Guzdial from Georgia Tech today. I have been interested in the idea of having programming languages that people can use differently but at the same time have a social context to the programming by allowing differnt people to collaborate.
However, this raises an important question and problem. Differnt people prefer different language. How do we get them to collaborate on the same code base ?
For example, visual programming languages have been used for introductory programming, however I don't necessarily feel like checking someone's work in a visual language. I'd like to work off a text based language.
This led into my discussion with Mark. He hinted at some work done in the 60's that proved that all programs could be deconstructed using only four constructs if , then , else and while.
This doesn't seem to be our real problem. The real problem seems to be the reconstruction. After deconstructing a programming into just these constructs what would be the best way to reconstruct them in a sane manner. That is an interesting question and one for which I do not currently have an answer.