If I understand your question correctly, this is done using the stereotype
mechanism. Basically, stereotypes are an extension mechanism whereby
existing model elements can be given new meaning. There are a number of
standard stereotypes, such as <<derived>> and <<friend>>, but it's possible
to define new stereotypes beyond hese standard ones.
Another mechanism in UML that might address some of your concerns is the
Object Constraint Language (OCL). The graphical language provides no
mechanism for describing constraints (such as a class being allowed to have
only one instance). To model a singleton, it would probably be prudent to
just introeduce a new stereotype <<singleton>>, but for more complex
constraints, it may be appropriate to annotatate diagrams using OCL.
----
Gregory Woodhouse
[EMAIL PROTECTED] / http://www.wnetc.com/home.html
"An atheist staring from his attic window is often nearer to God than the
believer caught up in his own false image of God."
--Martin Buber
----- Original Message -----
From: John S. Gage <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, January 26, 2000 4:04 PM
Subject: Re: Patient control
> How do UML models describe themselves? Is there a concept of meta-data
> in UML?
> John
>
