[plt-scheme] Serializable continuations?

From: Matthias Felleisen (matthias at ccs.neu.edu)
Date: Fri Feb 10 10:31:27 EST 2006

On Feb 10, 2006, at 12:08 AM, Jim Blandy wrote:

> On 2/9/06, Jay McCarthy <jay.mccarthy at gmail.com> wrote:
>> 1.  
>> http://cs.brown.edu/~sk/Publications/Papers/Published/pcmkf-cont- 
>> from-gen-stack-insp/
>
> Fun!
>
> (I like to read papers Matthias has contributed to just for the
> notation. One taught me about contexts,

Thanks but contexts go back to Barendregt (at a min), though I  
specialized them to "evaluation contexts" formerly known as  
"applicative contexts" and that became a standard tool to formulate  
reduction semantics.

> and this one is the first time I've seen overbars used like ... in  
> syntax-rules.)

Again, Barendregt used arrows for similar, simpler languages. I had  
seen it generalized to more complex things [almost] before I  
encountered ... . From what I understand Dan had picked up ... from Guy  
and passed it on to Eugene who put it into extend-syntax, the  
predecessor of syntax-rules. But deep down it was just an ASCII  
notation for arrows over patterns. So it is indeed the same idea.

Richard just pointed you to Featherweight. No, they don't deserve the  
credit either.

;; ---

On serializable continuations: They used to be in some funky form in  
Lisp machines (of course not for call/cc purposes) and they definitely  
showed up in Kali. I have also seen them in an SML/NJ dialect around  
1992 at the ERC in Munich; they used them for migrating computations  
from one machine to another, and they really loved the fact that SML/NJ  
had call/cc even if it was type unsafe at the time.

Our paper only shows a particularly useful approach given a novel set  
of constraints.

As Jay said, he'll show you more as things progress.

-- Matthias



Posted on the users mailing list.