Re: [moca] shooting from the hip: Mobility FAQ ?

[Tom Chothia <tomc@xxxxxxxxxxxxxxxxxxx>]

On Fri, 1 Aug 2003, Martin Berger wrote:

>  > Qu1: "Why the he.. do you have so many different calculi?
>
> (counter-question: why are there so many natural numbers?)


I'd like to suggest the counter question "Why are there so many different
programming languages?". I think the answer is the same in both cases,
particular languages and calculi will try to address different aspects of
computing. For instance, there are different calculi and languages that
are useful for data security, distributed environments, information flow,
fault tolerance, ...  the list is almost endless, which is exactly the
point. A calculi or language that try to include everything available in
all other languages or calculi would just be to big and cumbersome to be
much use.

Also as Martin mentioned another reason for the large number of calculi is
that this is still work in progress.  New calculi are often used to
demonstrate a new idea in the simplest possible setting. In this case it
is the new idea which is important, not the calculus itself. I doubt most
calculi will still be around in 10 years time but hopefully the ideas
presented by those calculi will out live them.


>  > Qu7: "Can it be implemented?
>  >       Has it been implemented?"


Pict and JoCaml are definitely worth a mention.


>  > Qu9: "Why all this fuzz about encodings?"


Exactly because we have so many calculi! Encodings between two calculi can
give us an insight into the exact differences between them. For instance,
the encoding of value passing CCS into plain CCS tells us that in the
setting of CCS the ability to pass value adds no real expressive power. It
also shows us that we can think of value passing as the same as passing a
guard or key to the value we wish to pass.


Tom
  
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The "models for mobility" mailing list     mailto:moca@xxxxxxxxxxxxxxx
 http://www-sop.inria.fr/mimosa/personnel/Davide.Sangiorgi/moca.html