> > I like the idea of simplifying these common constructs.  But making 
> > symbols act like blocks is a little problematic, IMHO.  The  
> duality will
> > break down eventually.  For instance, what if someone tries to fixTemps 
> > a symbol?
> 
> I don't see the problem at all.  The fact that blocks and symbols  
> share some common protocol does not imply either Symbol isa Block or  
> vice versa.  If fixTemps is necessary for invoking blocks, then  
> that's a different problem.


Well, it still says that Symbol isa EvaluableThing.  But you're right
that this is only a small maintenance problem, not a show stopper.

My dislike really comes from a different reason: in my mind,
symbols really *aren't* evaluable things.  A symbol is a
sequence of characters that can be used to give names to things.
An evaluable thing is... well, something else.

So why not keep the concepts clean for as long as we can?  I proposed
a different method for getting rid of the full block construct, which
seems to me almost as convenient but more conceptually clear.



Lex