[plt-scheme] two tool ideas

On Friday, February 14, 2003, at 11:53  AM, Gregory Cooper wrote:

>
>>
>> I'm not sure we're talking about the same thing when we say 'tail
>> position,' in particular because I don't see how either argument to a
>> cons could ever be in tail position.
>>
>
> It depends on the evaluation strategy.  The runtime could first 
> allocate
> an uninitialized cons cell, and evaluation of the arguments could then
> "return" into the fields of the cell.  In this case, the second 
> argument
> of the cons will sometimes behave as if in tail position.  Wadler's
> "listless" work deals with optimizations for such programs, which are 
> said
> to be "tail-recursive modulo cons".

I wouldn't call this tail recursive.  Tail-recursive modulo cons, 
maybe, but not tail-recursive.  In particular, I think that this 
program is going to grow without bound, no matter how you implements 
cons--even omitting the memory taken by the cons cells:

(define (loop x)
   (cons 3 (loop x)))

(loop 5)

Perhaps you could allocate extra space in every cons cell for the 
activation record?  Yikes.

john