OpenEuphoria

Re: The "perfect" language...

• Posted by Everett Williams <rett at GVTC.COM> Mar 19, 2000
• 449 views

Martin Nilsson wrote:

snip
>
>Short description of Kafka=A8:
>
>It's *probject-oriented*. Probjects are mixtures between functions and
>OOP-objects. Sequences can be regarded as a special kind of probjects.
>Everything evaluates to something: perhaps Kafka=A8 can be viewed as some
>kind of remote Lisp-clone.
>
>There is virtually no special syntax for OOP: no classes or interfaces;
>just two built-in probjects.
>The transition between OOP-style- and none-OOP-style programming will be
>easy and almost unnoticeable.
>
snip
>So my questions are:
>
>Is there any interest for this among Euphorians (or RDS); or do you feel
>Mike Nelson's and other people's OOP-libraries are simply enough?
>
>Are there any similar projects going on?

As far as I am concerned, the more the merrier.

>Is Euphoria 2.3 on it's way? Will it have namespaces?

Rob hints that it will, but who knows what that means and when it will come.=

If anybody else has heard more hints on time than I have, I would welcome
the knowledge.

>If the answer is "yes",
>   I might concentrate my energies on writing a good language description
>   before doing the implementation; so everybody can read it and get some
>   ideas.

I would always recommend writing a good language description before
implementation. That way, you can compare what you have done with
what you intended and trim one or the other, but at least you will know
where you were originally headed when you get buried down in the
"probject" 

>Otherwise,
>   I prefer to release everything at the same time, because it is a rather
>   unique language with many new concepts (probjects, cells etc) and I thin=
k
>   it will be better recieved if people can play around trial/error-wise
>   with the Kafka=A8-compiler and some example programs.

If the language desciption comes out first, at least you can get a comment
or three before committing to code. There might be one or two dead-ends
that could be avoided with more than one set of eyes on the design.

I know that a lot of people on the list would prefer the whole ball of wax o=
r
even the compiler without the language description. I have yet to see a
really clean project that was coded first and designed later. I have seen
some interesting "proof of concept" code done that way, but something as
large as a language has way too many gotcha's for that kind of "only real
men can code without a design" macho.

Everett L.(Rett) Williams
rett at gvtc.com