Re: re. Rick's musings

[Richard Beton <rdb@xxxxxxxxxx>]

Brian O'Neill wrote:
> Are there any surveys of occam users to identify its main usage?

A good question. But market research of its potential usage is
also important, especially because of the 'divorce' between 
Inmos transputers and occam.


> It is nice to have an upgrade to the language which adds to its flexibility
>  of use.  At the same time it is important that these changes do not have
> any side effects on current usage.  The best way to minimise this is by
> Beta testing of the compiler with a signification number of end users.

I can't imagine any side effects *except* by the removal of features or the 
alteration of semantics. So we should know what to expect from any 
proposed changes. But Beta testing is clearly desirable, for a variety
of reasons.


> Will the rest of the world embrace a new version of occam or will it carry
> on ignoring it?
>
> Yes lets have some modest changes to occam but I would still like to view
> it as a application specific language for embedded processors.

That depends on whether OFA cultivates the Unix market sufficiently. I
think the possibilities are moderately good, especially if we can offer
novel approaches for (i) SMP programming (a credible alternative to Posix
lightweight threads), (ii) GUI programming, (iii) animation.


> At the same time the group must push for the development of a better model
> of concurrency in a main stream language  i.e. lets have our own version
> of JAVA.

There is the distinct possibility that our own version of Java would have
even less impact than occam.


> I would like to propose
> 
> (1) That the group is committed to small but modest changes to occam.  The
> understanding is that this is a good thing to do but will not have a
> dramatic impact on the wider use of occam.
> 
> (2) That the group is committed to supporting a variation of JAVA that
> embraces a good model of concurrency.

(1) I disagree with the understanding. (2) may be a flier, but my reservation
is that doing our own thing with Java is that of heading off in our own
direction again, rather than having a significant impact on the rest of the
world. Now if we could influence Sun to change the de facto Java standard,
that would be exciting!

All the best,
Rick
-- 
              Richard Beton BSc MInstP CPhys
      Roke Manor Research, Romsey, Hampshire SO51 0ZN
-------  Standard disclaimer about my own views etc  -------
           See http://www1.roke.co.uk/WHR/WHR.html