Certainly it is good to study a wide variety of programming languages before designing an all new one! A lot of people have thought about the best way to start, and this has resulted in things like LOGO and Karel the Robot. REBOL seems like a very good replacement for LISP in teaching.

But most every course introduce students to programming via some variant of “I’m controlling a robot out in empty space”. Though it can feel empowering, I don’t think it’s a good foundation—it emphasizes the power of single lines of code, instead of imagining how those manipulations are often at odds with a greater architecture.

So I feel like early programming education should start with something like the game The Incredible Machine. Then perhaps move to solving simple problems with Turing Machines. These are better for teaching foundations of software because they emphasize problem-solving with patterns, rather than by direct control of the CPU.

Anyway…if you like the enum.r then please feel free to post it on the REBOL list (if you have something to say about it)! I myself am busy trying to push other agenda items and meddling in other ways…

Glad I found your recent post about your implementation of enum.r.

In fact as I recently began to learn more deeply about existing programming languages concepts, since I want to be able in the long run to write my own programming language to help newcomers to program in a high level pseudo-code language and I would like to use REBOL as my first experimental tool to support my learning in this domain - because it seems to have some advantages to me - but also some caveats.

During the last week-end I found that the Caml Programming language supports the feature of being able to define new user-defined data types, and especially there is an example about how to simply define an enumerated type like the one you just implemented in REBOL- since this feature is not natively supported by the language.

So I would like to thank you for your effort and let you know that it will be at least useful to me before someone else finds it in the library.

To promote its use by others, I suggest you leave some entry in the Mailing list too.

When I’l be ready to experiment with your enum.r I’ll let you know too about my feelings - but sure it will help to implement some of my work from now.

It’s fun you have such a clear image about what is needed to implement rightly the tools required for keeping to a minimum the efforts to debug implemented software. This is also one of my main concerns since it is rarely if not never implemented right on all the line…

If and when I will define other useful utilities I’ll let you know too.

Thanks again
Gerard