Re: Type checking (was: The strengths of FORTH)

In article <d6b19$47fd301e$cdd08510$19751@[EMAIL PROTECTED]
>,
Doug Hoffman  <dhoffman@[EMAIL PROTECTED]
> wrote:
<SNIP>
>But I am curious to know more about the "20 OOP extensions" you mention.
>  If we consider it a requirement that the extension be available now,
>be loadable on top of any ANS compliant Forth, and have the ability to
    ^^^^^^^^^^^^^^^^^^^

This is an unreasonable requirement. Make that
" implementable in any reasonable".

>survive use in a stand-alone non-source application, I am hard pressed
>to think of 2, let alone 20.  Where are these 20 or so that you mention?

The OOP system can only survive if carnal knowledge of the Forth is
incorporated. This means that it is inherently unportable.
A portable implementation of SAVE-SYSTEM is inherently impossible.

>  Do you have the URLs?  Context here is resource rich environment Forth
>programming (e.g., I wouldn't consider mini-oof, as clever as it is, a
>serious objects extension for this context).

The reasonable requirement of an OOP system is that its syntax
is re-implementable on other ANSI-94 systems, not that its code
is itself portable.

It may very well be that those 20 implementations all do.
Certainly my 2 blocks class definition does.
Still I cannot foresee what kind of problems with turnkeys
e.g. iforth would have if it uses a port of those blocks to
implement an OOP.

>
>-Doug


--
-- 
Albert van der Horst, UTRECHT,THE NETHERLANDS
Economic growth -- like all pyramid schemes -- ultimately falters.
albert@[EMAIL PROTECTED]
 &=n http://home.hccnet.nl/a.w.m.van.der.horst