Re: [stack] What does "concatenative" actually mean?

Le Fri, 6 Mar 2009 18:24:13 -0500,
John Nowak <[email protected]> s'exprima ainsi:

> Your point about naming it and being able to refer to it is confused.  
> Even if you define 'square', you still need to use quotation to form a  
> new function that pushes the 'square' onto the stack. It is not the  
> same as giving the name for a function in Scheme and getting that  
> function back as a result. You're explicitly creating a new function.

Isn't this issue implementation choice? What I mean is at the conceptual/programmer level.

* There must be a way to distinguish a reference to a word from running it (think eg at python "func" vs "func()" or Io "getSlot("meth")" vs "meth"). Why is it wrong to consider that "[square]" vs "square" plays this role?

* As I see it, the implementor has numerous options:
	1. Create a new func as explained above by John.
	2. Literally replace in code (~ macro) before processing (then we get an 		anonymous func again).
		[square] map ==> [dup *] map
	3. Store a reference (name, pointer, id) on stack when "[square]" is parsed.
	4. Store "[square]" as data on stack.
	5. Store square's definition on stack.
All would do the job?

Denis
------
la vita e estrany