[Axiom-developer] Re: Continuations in SPAD

[Stephen Wilson]

Hi Ralf,

Ralf Hemmecke <address@hidden> writes:

> > Ralf Hemmecke <address@hidden> writes:
> >> Could you demonstrate some SPAD-code-to-be which uses continuations
> >>  and solves a problem that cannot be solved easily by an Aldor
> >> equivalent?
> > I am not to sure what you want me to give. Writing a lot of pseudo
> > code is not terribly useful in all cases.
> 
> No not pseudo code. Define a syntax for continuations in
> SPAD-to-be. And then write real SPAD-to-be code.

Well, for that you might need to wait a while :) I understand the
motivation for asking, but it is not really practical at this point.

[...]
> I am looking for something that would cause me headaches to implement
> in (current) Aldor and which would be easy with continuations. Is must
> be related to implementing mathematics. I don't yet see how useful
> continuations are in a mathematical library context.
> 
> You mentioned parallel execution, but there is also
> http://www.inf.ethz.ch/personal/mannhart/piit/
> http://www.ph.ed.ac.uk/~bj/paraldor/WWW/

Perhaps the following paper would be worth looking at:
   
    www.csd.uwo.ca/~watt/pub/reprints/1995-issac-dynev.pdf

Which gives some of the basic ideas behind continuations, how they can
be implemented and incorporated into Aldor, and gives examples of the
kinds of (non-trivial) algorithms for which continuations are a
natural fit.

I am not really advocating anything new by adopting the notion of
continuations in SPAD.


Take care,
Steve