[Modeling] UN and Agents
Mon, 7 Apr 2003 15:15:21 +0200
> in UN-Solution-Rado1
> I don't understand the cardinality for member on Figure 1, I thought the
> cardinality is fixed and cannot evolve
Aha... I did know it, it was not described in the case study description. OK
if there are always for instance 16 members, then the multiplicity should be
fixed to 16 (not 1..*).
> why do you give on Figure 4 <<role>>? I don't understand this point,
> normally we said that chair:Chair that we have an agent named chair that
> exhibits the role Chair, so why do you duplicate this information? Do
> you mean something else?
Usually instances display stereotypes of their classifiers in UML diagrams.
It improves readability of the diagrams. <<role>> is a stereotype of the
Chair classifier. That's the reason. <<role>> is not a stereotype of
lifelines. I draw (or my CASE tool draws) stereotypes for instances
automatically... I'm sorry that I did not explain it in the document...
> what confused me on Figure 4 is that we have proponents that propose and
> members that receive proposals, of course, proponents are members but it
> does not appear on the diagram that when the chair sends a proposal to
> members, she sends it as well to proponents. Don't we need to add this
Figure 4 says that the proposal (together with the date) is sent to all
members (k members of UN-SC). If k is unknown at the modeling time, we could
add a special constraint or description/comment to the lifeline saying that
the "members" lifeline represents all instances of :Member role belonging to
UN-SC. Now, if we know that there are 16 members, I'd rather put
multiplicity 16 to the members multi-lifeline.
> I don't understand conclusion c) about attaching roles
I wanted to express that modeling of roles I used in my UN-SC solution is
little bit semantically "richer". E.g. in your original proposal it would be
difficult (impossible???) to express that one agent plays several roles in
one interaction, or that an agent started to play new role while all old
roles remain, etc. Probably I used wrong or too short formulation, without
Radovan Cervenka | email@example.com
Whitestein Technologies | www.whitestein.com
Panenska 28 | SK-81103 Bratislava | Slovak Republic
Tel +421(2)5443-5502 | Fax +421(2)5443-5512
If you are not the intended recipient of this email,
you are not authorized to make any use of it;
please delete it and notify us by return email.