[r6rs-discuss] Implicitly Concurrent Scheme
alaric at snell-pym.org.uk
Thu Oct 29 07:56:20 EDT 2009
On 29 Oct 2009, at 10:52 am, Robert Ransom wrote:
> On Thu, Oct 29, 2009 at 03:02, Alaric Snell-Pym <alaric at snell-pym.org.uk
> > wrote:
> Haskell implementors may have some relevant experience with this sort
> of thing, in particular :-) I wonder what fun could be had
> generalising lazy evaluation to potentially involve parallel threads?
> I guess the same sorts of issues with continuations (and, therefore,
> exceptions) arise there.
> No. Haskell does not allow continuation capture (except as a side-
> effect within the continuation monad; look up Control.Monad.Cont for
Oh, yes, I wasn't talking about the semantic issues with combining
laziness (or concurrency) and general continuations (and all they
imply, such as continuable exceptions) - although I suspect those
needn't be a problem within the restricted scope of evaluating
arguments where the application doesn't proceed until all the thunks
have been thought - but that I think some work has been done on how to
optimise laziness to avoid having to actually thunk everything, which
I suspect would be applicable in this case too.
More information about the r6rs-discuss