[gambit-list] Efficient support for delimited continuations?

Newcombe, Chris cnewcom at amazon.com
Wed Dec 21 22:05:22 EST 2005


Or perhaps Hieb & Dybvig's 'spawn' operator would be a better match for Gambit's threading system?

http://www.cs.indiana.edu/~dyb/papers/LaSC-7-1-pp83-110.pdf

  "In the presence of concurrency, the effect of a control operation must
be defined in terms of the branches of a process tree. By "process tree,"
we mean simply a tree-structured continuation record. Since traditional
continuation control operators are derived from the notion of representing
continuations as stacks, it is not surprising that such operators are inadequate
for controlling concurrency. The spawn operator, on the other hand,
is designed specifically for the control of tree-structured concurrency."
 

-----Original Message-----
From: gambit-list-bounces at iro.umontreal.ca [mailto:gambit-list-bounces at iro.umontreal.ca] On Behalf Of Newcombe, Chris
Sent: Wednesday, December 21, 2005 5:35 PM
To: gambit-list at iro.umontreal.ca
Subject: [gambit-list] Efficient support for delimited continuations?


Marc,

Does (or will) Gambit v4 have native support for shift/reset or splitter/abort/call/pc?
And if so, can the resulting partial continuations be serialized?

I read your paper on 'A Better API for First-Class Continuations" (http://repository.readscheme.org/ftp/papers/sw2001/feeley.pdf), and I see that Gambit v4 does have continuation-graft.
But that paper doesn't mention partial continuations.  

Rationale: I am eager to make use of Oleg Kiselyov's 'Zipper In Scheme': http://okmij.org/ftp/Scheme/zipper-in-scheme.txt

Interestingly, at the end of that article, Oleg says "It doesn't matter which particular delimited continuation operator (shift, control, shift0, splitter, cupto, etc) is supported -- all of them are equally expressible", and cites a paper by Chung-chieh Shan which I haven't read yet.  However, Christian Queinnec seems to make a good case that his own splitter operator is less dangerous/confusing than shift/reset when nested: http://www-spi.lip6.fr/~queinnec/PDF/contlib.pdf

There's a discussion thread here: http://lambda-the-ultimate.org/node/view/606

Thanks, 

Chris
_______________________________________________
Gambit-list mailing list
Gambit-list at iro.umontreal.ca
http://mailman.iro.umontreal.ca/mailman/listinfo/gambit-list



More information about the Gambit-list mailing list