Reactive Systems
- Topic:
-
Specification of reactive systems, including consideration of states
- Coordinator:
- Egidio Astesiano (Genova) and
Heinrich Hussmann (Munich)
- Mailing List:
-
cofi-reactive
- Notes Series:
- see the CoFI Reactive Systems Home Page below
- Home Page:
-
http://www.disi.unige.it/person/ReggioG/COFI/index.html,
mirror
An area of particular interest for applications is that of reactive,
concurrent, distributed and real-time systems. There is considerable
past work in algebraic specification that tackles systems of this
kind, but nonetheless the application of Casl to such systems is
speculative and preliminary in comparison with the rest of CoFI.
The aim here is to propose and develop one or more extensions of Casl to deal with systems of this kind, and to study methods for developing
software from such specifications. Extensions in three main
categories are currently being considered:
- Combination of formalisms for concurrency (e.g. CCS, Petri
nets, CSP) with Casl for handling classical (static) data structures;
- Formalisms built over Casl, where processes are treated as
special dynamic data; and
- Approaches where Casl is used for coding, at the meta-level, some formalism
for concurrency, as an aid to reasoning.
Since object-oriented methods have a dominant role in concrete
developments, the group will address OO aspects insofar they are
needed for reaching the above goals, but object orientation is not
addressed as an independent topic in itself.
Work in this area begun only after the design of Casl was complete
and so it is still in its early stages. Presently, the work is
organized in two tracks: autonomous and coordinated extensions.
Proposals for such extensions are autonomously submitted to the Group;
they are required to follow a suggested submission procedure,
and are subject to approval [16].
Currently a number of proposals have been announced.
It has been decided to start an effort centered around UML, the Unified
Modeling Language.
We are pursuing two tracks.
- The basic idea is to adopt Casl, or an extension of it, for
annotating UML, thus possibly replacing OCL (work led by
Heinrich Hussmann, Dresden).
- We have joined the activity of the free group Precise
UML; the goal is to provide a formal/rigorous underpinning of
UML, possibly exploiting CoFI-related techniques. A sketchy
proposal for a general approach to the problem has been presented at
a workshop at OOPSLA'98 [2]. Two other draft papers
are available, one relating the ADT approach to UML [23]
and the other proposing an underlying model for UML state machines
[46].
CoFI
: CoFI -- Version: -- November 29, 2004.
Comments to pdmosses@brics.dk