Something David N. Smith of IBM said recently on
comp.lang.smalltalk got me thinking.  During a discussion
of scientific programming in Smalltalk he said:

After all, FORTRAN is slowly creeping towards
objects (which is downright frightening).

This prompted me to reflect on the idea that the concept
of "objects" is something of a generalization of the
structured programming concepts of "function" and "data".
This further prompted me to wonder what the "next step"
is, i.e. what comes "after" objects?  What is currently
separate from the concept of "object" which can be
joined to it, thus simplifying things?  I expect that this
ties in somehow with the things Alan was talking about
earlier, with messaging, etc, but I don't see it, at least
not clearly enough to get a good grip on.  Anybody got
any thoughts on this?

Bob Jarvis
The Timken Company