A mopping up exercise

I gave a keynote talk this morning at JMLC2006 in Oxford, about Domain Specific Languages.  To be controversial, I said that the most important development in programming language design ever was the shift from procedure/data languages to class/instance languages, and that a good proportion of programming language research today is a mopping-up exercise resulting from this change.  Rather to my surprise, nobody disagreed.

Comments (2)

  1. Anthony Lauder says:

    Blimey, I am sure I would have disagreed. There are plenty of other developments that are at least as great – such as: the distinction between source code and object code (no more flipping switches on the front of the box to program in hex); the idea that data can have structure; the whole idea of functions (some folks still think Lisp is the greatest ever language); and so on. The class/instance is currently prominent in most folks mind because OO is till see as new-ish but not so new that is it risky.

    My personal feeling on the greatest improvement ever was the idea of the stored program as something that runs in the very memory that holds data. That was a revelation.

  2. thecoach says:

    hmmmm…it is hard to say, obviously they all matter, but I would partially agree by saying that making explicit the abstraction from program domain to solution domain is the big one.

    The biggest one ever will be the seamless interaction and composition of multiple DSLs. 😉