Fwd: [u2p-issues] UML Universe of Discourse

Date view Thread view Subject view Author view Attachment view

From: Joaquin Miller (miller@joaquin.net)
Date: Sun 17 Nov 2002 - 17:38:33 GMT

>From: Trygve Reenskaug <trygve.reenskaug@ifi.uio.no>
>X-eGroups-Approved-By: ckobryn <cris@ieee.org>
>Delivered-To: mailing list u2p-issues@yahoogroups.com
>UML Universe of Discourse
>The question of the UML Universe of Discourse came up during the Helsinki 
>meeting. I am afraid a general discussion of this question could be 
>inconclusive and last forever. But I do believe it to be fruitful for each 
>of us to contemplate our own position.
>There is no room for actual drive trains for automobiles and boats in my 
>Universe of Discourse. Nor is there room for a VCR or a boy of flesh and 
>blood. But there is room for models of automobiles, boats, VCRs, and boys.
>My concepts build on the notion of a system as defined in 1975 in the 
>Delta project:
>A system is a part of the world which we choose to regard as a whole, 
>separated from the rest of the world during some period of consideration, 
>a whole which we choose to consider as containing a collection of 
>components, each characterized by a selected set of associated data items 
>and patterns, and by actions which may involve itself and other components.
>We make mental and manifest models of a considered system. A system 
>description is expressed in a system description language. A model is 
>generated from a system description.
>To me, UML is a system description language. My Universe of Discourse is 
>the set of models that can be expressed by a subset of UML; namely the 
>subset that can generate object models. This means that the UML 1.x 
>metaclasses Association and AssociationEnd are excluded. The reason is 
>that instances of these metaclasses are not reachable from the model 
>objects; i.e., instances of metaclass Class. (The problem is solved by the 
>NavigableEnd metaclass that is subclass of both AssociationEnd and Attribute.)
>         Erik Holbæk-Hanssen, Petter Håndlykken, Kristen Nygaard:
>         System Description and the Delta Language
>         Norwegian Computing Center 1977
>I have scanned the relevant pages from this report. I have posted them 
>with permission at
>         http://www.ifi.uio.no/~trygver/UML2.0-U2P/DeltaLanguage.doc
>It can also be worth while to look at the original Simula book:
>         Graham M. Birthwistle, Ole-Johan Dahl, Bjørn Myrhaug,
>         Kristen Nygaard:
>         Simula Begin.
>         Auerbach, Philadelphia 1973
>         ISBN 91-44-06211-7
>This book sports the world's first UML-like class diagrams!
>Trygve Reenskaug      mailto: trygve.reenskaug@ifi.uio.no
>Morgedalsvn. 5A       http://www.ifi.uio.no/~trygver
>N-0378 Oslo           Tel: (+47) 22 49 57 27


Very well said!

i will agree with everything you write here, and add:

My Universe of Discourse is the set of models that can be expressed by a 
subset of UML; namely the
subset that can generate object models (in the classic sense, started by 
Simula, continued by SmallTalk and other languages, and commonly understood 
today in terms of Java) and object models (in the strict RM-ODP sense) and 
object models in the sense that we model the environment of a system with 
objects, too.

This means that the UML 1.x metaclasses Association and AssociationEnd are 
included by me (even if not "navigable").

The reason is that the model objects (i.e., instances of metaclass Class) 
in a link are always reachable from the associations, er, links (i.e., 
instances of metaclass Association) and are often reachable directly from 
other model objects themselves (as with "navigable ends").

Forgive my catholicism in this matter.

I enjoyed our conversations in Helsinki.  Hope to see you again soon.



Date view Thread view Subject view Author view Attachment view