RE: call/cc

["David Simmons" <address@hidden>]

> -----Original Message-----
> From: Guy Steele - Sun Microsystems Labs
[mailto:gls@labean.East.Sun.COM]
> Sent: Tuesday, December 18, 2001 5:01 AM
> To: gls@labean.East.Sun.COM; kragen@pobox.com; pixel@mandrakesoft.com;
> David.Simmons@smallscript.net
> Cc: ll1-discuss@ai.mit.edu
> Subject: RE: call/cc
> 
> 
>    From: "David Simmons" <David.Simmons@smallscript.net>
>    Date: Mon, 17 Dec 2001 15:58:33 -0800
>    ...
>    > For a really good test that will provide better separation
>    > of the languages, try curried addition of first one parameter,
>    > then the other, plus 1:
>    >
>    > (((lambda (x) (lambda (y) (+ x y 1))) 3) 5)     Scheme, 24 tokens
>    >
>    > --Guy
> 
>    SmallScript (superset of Smalltalk):
> 
>        [:x|[:y|x+y+1]](3)(5)		"13 tokens"
> 
>    -- Dave S. [SmallScript LLC]
> 
> How is that 13 tokens?  I'm a little unclear on how you would
> count ":x|" and ":y|", but surely
> 
> [  [  x  +  y  +  1  ]  ]  (  3  )  (  5  )
> 
> is already 15 tokens, so the whole thing must be somewhere
> between 17 and 21 tokens?  If 21, then it saves two "lambda"s
> and a pair of parens over Lisp, then loses one back by needing
> a second "+".
> 
> --Guy

Hi Guy.

I was taking my token count from the SmallScript parse tree.

    TBlock	                    [:x...]
	TVar                      :x|
	    TBlock                [:y...]
              TVar              :y|
                  TVarRef       x
                  TMsg          #+...
                      TVarRef   y
...stopped indenting  TMsg      #+...
...mail will      TLiteral      1
...screw up     TParenExpr      (3)
                  TLiteral      3
		    TParenExpr      (5)
                  TLiteral      5

I now see how you are counting (lexer) tokens. Which means 13 is the
wrong count. On that measure, as you suggest, it would make the count
21.

    1   [
    2   :
    3   x
    4   |
    5   [
    6   :
    7   y
    8   |
    9   x
    10  +
    11  y
    12  +
    13  1
    14  ]
    15  ]
    16  (
    17  3
    18  )
    19  (
    20  5
    21  )

OTOH, SmallScript doesn't really provide anything equivalent to the
simplicity or generality of s-expressions. Which is where, I suspect,
scheme would crush SmallScript in a more general form of this kind of a
comparison.

-- Dave S. [SmallScript LLC]

SmallScript for the AOS & .NET Platforms
David.Simmons@SmallScript.com | http://www.smallscript.org