[no subject]

[BAK@MIT-AI]

The current way of doing EVALUATE BUFFER is to do all the reading before any of the 
evaling-printing.  There is one screw in this: When there is an unbalanced paren
somewhere in the file there is no dribble of read and evald functions to give you an 
indication of where the misbalanced paren is.  You have to search through the
entire file looking for it.  What is the argument for doing it this way instead
of the old-fashioned: read one function - eval - print  loop?  If it is not compelling
I would suggest switching back.