[Modeling] Role: [was Comments on interaction diagram modeling doc]

[James Odell]

On 3/19/03 5:47 PM, "Renato Levy" indited:

> I second Nodine, IF roles are correctly defined then one should be able to
> describe a protocol in a open system based only on role playing.

I would further add that you cannot define a protocol involving a particular
role, then it is not a role.  By definition -- IMO -- the primary purpose of
a role *is* interaction. Ergo: no interaction, no role.

-Jim