[plt-scheme] Re: Semantics of ec?

On Feb 16, 2006, at 8:45 AM, Matthias Felleisen wrote:

>
> On Feb 16, 2006, at 11:38 AM, David Van Horn wrote:
>
>> Matthias Felleisen wrote:
>>> On Feb 16, 2006, at 7:33 AM, David Van Horn wrote:
>>>> Is a reduction semantics for "escape only" continuations  
>>>> available in the literature somewhere (or any other semantics  
>>>> for that matter)?
>>> Cartwright, Curien, Felleisen: Fully abstract models of  
>>> observably sequential languages. Info. Comp. 111(2). 1994. 297--401.
>>
>> Thanks for the reference.  I was wondering if there is something  
>> that gives the reduction semantics for call/ec, which I could  
>> contrast with the usual reduction semantics of call/cc, ie. I'm  
>> interested in how the notion "an escape continuation is only valid  
>> when the current continuation is an extension of the escape  
>> continuation" is formalized.
>
> That's precisely what the op sem in this paper for catch does. If  
> you're asking for an error-message version of the op sem, you need  
> to mix and match the catch the semantics with the call/cc  
> semantics. You may end up with Mike Sperber's rewriting semantics  
> for dynamic-wind in the end. -- Matthias

For a half-baked alternative, you might consider using continuation  
marks to model both call/cc and call/ec.

John Clements

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2430 bytes
Desc: not available
URL: <http://lists.racket-lang.org/users/archive/attachments/20060216/933910c4/attachment.p7s>