On Tue, Feb 12, 2008 at 05:15:14PM -0800, Andrew Lentvorski wrote:
> Yes, but the difference is that the background info for mathematics *is*
> precise.  There is a directly derivable chain from very few, very simple
> first principles the whole way through partial derivatives to the
> simplification that derives the chain rule you talk about.  This is
> simply *not true* for programming in general.

We use carefully crafted physics/math software to efficiently communicate
your "directly derivable chain" as little or as far back as anyone wants to
go.   That is how software comes into this.  Programming is just a
communication tool.

> Even if I start with lambda calculus, it is very hard to get to a useful
> description of something as simple as Scheme.  Go look at Chapter 5 in
> Lisp in Small Pieces.  Look at how many "shortcuts" need to be taken to
> produce something even remotely tractable, and then start adding things
> like functions of variable arity, global and local environments, choice
> operators, etc.

I've said before that I'm not arguing against need for layers of abstraction.

cs

-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg

Reply via email to