This is a pretty common situation. Here's how I divide things up...
I would agree with the architect that the design classes should be in the packages
representing the subsystems for which they are a realization. The use case
realization diagrams don't belong in either of these packages because they are NOT
design level constructs. Rational pushes you to create a Use Case Realization package
in the logical view which is okay or alternatively you can create use case realization
diagrams directly attached to the use cases in the use case view. Either way the key
is that the use case realization diagrams should show the internals of any of the
subsystems. I only show the interfaces and the class that is acting as a surrogate
for the subsytem (a class in the subsystem package that is stereotyped as
<<subsystem>> and which has realization relationships to the interfaces that subsystem
will project). For each of the messages being sent to an interface I attach a note
the message and nd then drag the detailed sequence diagram (showing the subsyt!
!
!
!
!
em's realization of this message)into that note thus creating a hyperlink. So what I
have for the use case realization is a high-level illustration of the subsystem based
realization which is linked directly to the detailed realization of the subsystem.
This approach is VERY powerful and I highly recommend it.
Neal McWhorter
President
McWhorter Technology Management
Bussiness Process Modeling/Enterprise Architecture/Process Mentoring
DIEBEN Peter <[EMAIL PROTECTED]> wrote:
>
Hello,
While doing the architecture for our system I came across the following
problem, which I'll describe in general terms:
I have a use case UC1.
The sequence diagram of UC1 shows 2 analysis classes A, and B.
The architect decided to put class A in sub-system SSa and class B in SSb
(i.e. two separate packages in the Logical View, each with its own analysis
classes).
Now I want to make the use case realisation (diagram) of UC1. What do I do?
Put it in SSa and use class B in SSa? The consequence is that looking at
SSb it is very hard to see that one of its analysis classes participates in
the realisation of UC1.
Or is it possible that UC1 be realised by 2 -incomplete- use case
realisations, one in each subsystem?
Peter Dieben
(32)(0) 2 729 96 57
[EMAIL PROTECTED]
************************************************************************
* Rose Forum is a public venue for ideas and discussions.
* For technical support, visit http://www.rational.com/support
*
* Admin.Subscription Requests: [EMAIL PROTECTED]
* Archive of messages:
http://www.rational.com/products/rose/usergroups/rose_forum.jtmpl
* Other Requests: [EMAIL PROTECTED]
*
* To unsubscribe from the list, please send email
*
* To: [EMAIL PROTECTED]
* Subject:
* Body: unsubscribe rose_forum
*
*************************************************************************
************************************************************************
* Rose Forum is a public venue for ideas and discussions.
* For technical support, visit http://www.rational.com/support
*
* Admin.Subscription Requests: [EMAIL PROTECTED]
* Archive of messages:
http://www.rational.com/products/rose/usergroups/rose_forum.jtmpl
* Other Requests: [EMAIL PROTECTED]
*
* To unsubscribe from the list, please send email
*
* To: [EMAIL PROTECTED]
* Subject:<BLANK>
* Body: unsubscribe rose_forum
*
*************************************************************************
