continuations difficult? (was Re: CPS in Parrot)

["Perry E. Metzger" <address@hidden>]

Geoffrey Knauth <geoff@knauth.org> writes:
> We need good pictures that show how continuations work, and good
> stories.

As a person of only modest intelligence, when I read the explanations
of call/cc in the scheme RnRS documents, it was far from obvious to me
that a continuation didn't include the complete set of bindings in
effect when it was captured -- as a result, I didn't see how it could
be implemented without copying a full snapshot of memory. The
explanations I've seen in texts have been equally confusing and
ambiguous. Mostly, they can only explain how the idea works to people
who already understand it, which is not very useful. (I mention scheme
because that was the first language I used with first class
continuations.)

Even now, I think there is a lack of good texts explaining simply and
clearly what the semantics of call/cc etc. are for beginners (or at
least that these texts are not well known if they do exist.) That
in and of itself makes it difficult to judge if the concept is
inherently difficult for people to understand, or if it is simply hard
for people to find ways to learn about it.


-- 
Perry E. Metzger		perry@piermont.com