(responding to Eric)

> Here's the bet:  When you use boundary, entity, and control classes, the
> controls should contain the business rules. 

I'll have to take the role of Devil's Advocate here, because I tend
not to use the terms boundary, entity and control.  But assuming
I did, and assuming they are the sort of thing I think they are:

Hmm, let's hope we're all working from equivalent definitions
here...

> THIS IS OBVIOUSLY TRUE!  This is a thing that you can see readily in a
> sequence diagram.  The natural thing to do in *lots* of business
> applications is to pass messages between boundaries and entities through
> the control.  This is analogous to three-tiered architecture, with a
> Graphical User Interface (GUI), a Business Object Model (BOM), and a
> Relational Database (RDB).  The control is analogous to the BOM, that's
> obvious.

...I suspect I'm not using the same definitions as this chap, for a start.
The three tiers have different aspects of the same object.  An
object that logically resides in the system has its core behavioural
aspects in the middle, Business tier, its presentational aspects 
(or some of them, anyway) in the GUI tier, and its persistent
aspects in the RDB.  This object, if it has all these aspects,
is probably an Entity object, but could possibly be a Boundary
or Controller object.  So I'd be happy to agree that the
Business Rules (assuming these are the same as my
Domain requirements as opposed to Use requirements or
Non-Functional requirements) belong in the middle tier, along
with all the other business logic.

> THIS IS OBVIOUSLY CRAP!  As Freud said, there are times when a cigar is
> just a cigar.  You can see an analogy between boundaries/GUI and
> entities/RDB, but that doesn't prove that there is a similar
> relationship between controls and the BOM.  Controls should be just
> smart enough to route messages between objects in a subsystem (which
> might mean a business area).

Hmm - while I think, despite my previous superficial agreement with 
the other chap, this one is supporting the right side of the
argument, he's making a poor job of it.

Of course, I fully support the position taken by Todd Dunavant,
but I think it will take just a bit more than that to earn me the
virtual beer that's mine by rights (unless Huseyin dives in
again and claims his share, that is...).

> As usual, the bar bet is on a controversial question that should be
> obvious to everyone, dammit, but the instructors at Seneca College
> disagree with each other.  I will try to state opposing positions
> even-handedly, and will confess my "side" only after others have stated
> some opinions. 
>
> OK, folks, what do you think?  There is a virtual beer in it for a
> correct answer, but you have to make a clear winning argument! 

Those of you with a better memory and access to the right
forum will have heard my argument on this before - the rest
can wait while I see what other folk say, and try and find what
I wrote before.  I'll be back ...  ;-)

Paul Oldfield

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
www.aptprocess.com

any opinions expressed herein are not necessarily those of
Mentors of Cally or the Appropriate Process Movement
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

************************************************************************
* Rose Forum is a public venue for ideas and discussions.
* For technical support, visit http://www.rational.com/support
*
* Post or Reply to: [EMAIL PROTECTED]
* Subscription Requests: [EMAIL PROTECTED]
* Archive of messages:
*    http://www.rational.com/support/usergroups/rose/rose_forum.jsp
* Other Requests: [EMAIL PROTECTED]
*
* To unsubscribe from the list, please send email
*    To: [EMAIL PROTECTED]
*    Subject: <BLANK>
*    Body: unsubscribe rose_forum
*************************************************************************

Reply via email to