Fw: [SysML-Evaluators] Another view of "parsimony"

0 views

Skip to first unread message

David Oliver

unread,

Dec 19, 2005, 5:23:49 PM12/19/05

to SysML Evaluators, Tim Tritsch, Sandy Friedenthal, Samantha F Brown, Roger Burkhart, Rick Steiner, Phil Spiby, Peter Shames, Michael Latta, Michael Dickerson, Mark Sampson, Julian Johnson, Jozef Bedocs, Joseph Skipper, John Nallon, Jim S. Schier, Jim Long, James Odell, James N Martin, James A U'Ren, Ian Bailey, Howard Lykins, Harold P. Frisch, Georg Siebes, Erik Herzog, Darold Smith, Daniel T. Cocks, Cris Kobryn, Conrad Bock, Chris Sibbald, Bob Cohen, Asmus Pandikow, Mike T Loeffler

The following inserts are my own opinion about this:

----- Original Message -----
From: "Thomas Weigert" <thomas....@motorola.com>
To: <SysML-Ev...@googlegroups.com>
Cc: "'Thomas Weigert'" <thomas....@motorola.com>
Sent: Thursday, December 15, 2005 1:28 AM
Subject: [SysML-Evaluators] Another view of "parsimony"

While it may not be of primary interest to somebody who is only interested
in a system engineering notation, I believe we should nevertheless consider
how well each submission fits into the overall context of UML.

The reason for this is that system engineering is not done in isolation, but
at least for software systems the specifications will be handed off to teams
who will implement the software artifacts comprising the specified system.
These teams will most likely leverage UML to drive their development.

Questions that are of interest in this context are, for example:

* Does the profile introduce concepts that are inconsistent to existing UML
concepts (or are likely to be confused with existing, but different, UML
concepts)?
"Yes SysML has concepts that existing UML does not support or contain. This
was the reason for creating SysML.
Some essential Systems Engineering concepts are still missing from both
submissions. I do not know UML well enough to discuss inconsistency." DWO

* Does the profile introduce a new concept where an existing UML concept
could have served equally well, or could have been easily extended to
address the perceived need? "Submitters have to answer this one."

* Does the profile introduce concepts in a manner that makes it difficult to
migrate the system specification into a UML specification of the software
subsystems during development?
"No. SysML can provide an executable black box specification of what is
needed for software. It cannot be in software language directly because
there is no consensus in that community on a single design and
implementation language. Many are used. This situtation is identical for all
the other engineering disciplines because they have their own special
languages. DWO"

I believe satisfactory answers to these types of questions are just as
important as to "does the profile support feature X" type questions, if we
want this profile to be used in real-life development.