Re: Oberon/F Version 1.3....

Stan Warford (warford@pepperdine.edu)
Mon, 3 Mar 97 9:40:53 PST

Re:

> --better support for independent threads (I want to be able to have a
> procedure run in the background without any special coding -- there's no
> way my algorithm will work with the current method)

I agree that eventually we need this in BlackBox. Support for true
concurrent processing is one of the things that most people ask about
when I tell them about the development environment. It is extremely
encouraging to read Gutknecht's upcoming paper on a Proposal for
Self-Active Objects in Oberon. His proposed model for concurrency
is much cleaner than the one for Java. I only wish that the proposed
syntax were more in keeping with Oberon-2 instead of Oberon.

> --various improvements to the framework in general (it seems that there
> are still lots of areas which could use improvement, but for all I know
> the functionality is there but the lack of documentation hides it...)

The other feature I wish for (although in the language rather than
in the framework) is the template idea as in C++. Here again, the
paper by Roe and Szyperski on Lightweight Parametric Polymorphism
for Oberon is encouraging.

It is especially good to see in these two papers that the proposals
are well thought out and maintain the minimalist approach in a way
that integrates the features seamlessly. These two issues are much
more important than issues such as enumerated types, overloading
operators, complex numbers, etc., because they provide capabilities
that are otherwise not possible with the current language. I hope
that Component Pascal can incorporate them eventually. It is research
activity such as this that keeps me in the Oberon camp, to know that
work is still being done and that the language will evolve to keep
getting better.

Stan Warford
Professor of Computer Science
Pepperdine University
warford@pepperdine.edu