Re: [moca] Combinators i Have Known And Loved

[Martin Berger <martinb@xxxxxxxxxxxxxx>]

> Sum is needed to represent arbitrary transition systems, and are often
> convenient abstractions

I don't deny that sums are convenient and powerful. But they are also
problematic, for example when one wants to implement or type them. Their
representative power can also be overestimated: I find value passing
more intuitive than pure synchronisation with infinite summation over
the value domain, and i similarly prefer parallel composition as a
basic operator, rather than having to implement it a la expansion theorems.

Be that as it may, what I want to understand is how much of the power of
the sum is really needed: I'd like play a bit of the old divide-and-conquer
and stratify the sum into simpler combinators like timers or value passing.
Such a decomposition seems to help with developing interaction types.
But I'm a bit worried that I might overlook a crucial application of
sums, hence my request to the community.

Martin (http://www.dcs.qmul.ac.uk/~martinb)

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