First algorithms, then interfaces
c2.com
c2.com
Developing a useful, general framework for expressing the relations among different types of entities (what philosophers call ``ontology'') seems intractably difficult. The main difference between the confusion that existed ten years ago and the confusion that exists now is that now a variety of inadequate ontological theories have been embodied in a plethora of correspondingly inadequate programming languages. For example, much of the complexity of object-oriented programming languages -- and the subtle and confusing differences among contemporary object-oriented languages -- centers on the treatment of generic operations on interrelated types.
Ouch. This was written in 1996, but I don't think that languages like Java and Ruby have changed the state of the art that is being discussed. See http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-18.html... for the full context of this quote. (I was quoting from footnote 52.)
In an odd unrelated note, I've noted that there is some correlation between people's opinions on object oriented programming and how they eat corn. See http://bentilly.blogspot.com/2010/08/analysis-vs-algebra-pre... for more on this.
And this obsession is also a great weakness of course, because it makes central something that is not central. I have great trouble with some parsing issues (which I think is very algebraic - certainly regular expressions are), because it is just that little bit too complex for me to intuitively grasp the efficiency issues. Instead, I should use (inefficient) algebraic abstraction. But... it's just not quite natural for me, and I just can't quite grasp it.
I would say Java has probably made it worse. Ruby is, however, (at least loosely) based on Smalltalk which IMO has the nicest OO library for the things it addresses (e.g. collections).