[Modeling] The UML 2.0 concept of Collaboration

Marc-Philippe Huget M.P.Huget@csc.liv.ac.uk
Tue, 03 Jun 2003 10:50:31 +0100

Hello Stephen and all,

Stephen Cranefield wrote:

> It's probably a bit late to be introducing more notation, but I think
> it would be useful to adopt the UML 2.0 notation for a Collaboration
> (see Section 3.3, page 132, in the UML 2.0 Superstructure
> specification ad/2003-04-01, 10 April 2003):

It is never to late to propose something if it improves the quality of
the specification

> In AUML, interaction diagrams could be used
> for different purposes, e.g. to define interaction protocols or just
> to illustrate an example interaction between agents.  It seems to me
> that an interaction protocol corresponds well to a Collaboration and
> that IPs should be defined as a collaboration and an associated
> interaction diagram.  This also provides a way to associate several IP
> diagrams with one interaction protocol, which would be useful to
> illustrate different scenarios.

Well, well, I don't know what to answer, I agree with you that this
Collaboration is important in the context of agents and IPs but my
question is does this element need to be included in the interaction
diagram or somewhere else? My aim when writing interaction diagram spec.
is really to focus on IPs and let apart other notions, it is the reason
actions or combining protocols are light, it is because I think this is
the subject of another specification where we write combination of
different elements: roles and groups with IPs, IPs together, etc. in this
latter case, your idea about Collaboration can be included.

What's your opinion? Do we need to address this question in the spec?


Marc-Philippe Huget

Agent Applications, Research and Technology Group
Department of Computer Science
University of Liverpool
Chadwick Building, Peach Street
L69 7ZF Liverpool
United Kingdom

email: mph@csc.liv.ac.uk