Ronald,

 

This sounds more like a business rule than a system requirement.
Although the system is supposed to implement this. We record these as
business rules, since it is a rule that also lives outside of the IT
context. These kinds of rules in our situation must be captured in a
central place and are maintained separately from the IT process. This
way we achieve reuse of this information without having to redefine
these rules for every project. This approach fits product based
companies best (financial etc).

 

With regards,

Ronald

 

________________________________

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On Behalf Of Ronaldo r
Sent: 05 June 2007 21:08
To: Eclipse Process Framework Project Developers List
Subject: Re: [epf-dev] system requirements

 

Hi Jim, 
we use conceptual model to get requirements such domain of informations.

When we put a entity as a class in our conceptual model, we can put some
atributes and write the domain for that attributes.

Example: we put a entity named Customer with an attribute named age. We
define in the constraints of age that the age must be between 18 and 70.
This kind of information will be used to test the system. It is a
requirement, isn't it? 

The example above is a a requirement? This kind of information should be
in other place, rather than conceptual model? Some authors tell to not
put in the uses case the attributes of the entities. What's the
recomendation of OpenUP for this kind of requirement? 

On 6/5/07, Jim Ruehlin <[EMAIL PROTECTED]> wrote:

Hi Ronaldo,



All the requirements are characterized as use cases (for functional
requirements) or supporting (for non-functional) requirements. Artifacts
like the glossary and conceptual model support the understanding of 
requirements (as well as development), but are not considered
requirements by themselves. For example, the definition of a term doesn'
t indicate how the system must perform, or how it can be tested. But
that term may be used in the context of a requirement, which is written
to be unambiguous, testable, and understandable.



Artifacts like prototypes prove technology, get feedback from the
customer, etc. But they don't describe, in an unambiguous way, how the 
final system should perform. For that you need some form of well written
statements that the stakeholders and development team can understand and
agree on. OpenUP uses use cases and supporting requirements to achieve 
this. There are other methods of course, such as user stories or writing
a bunch of discrete requirements.



- Jim



____________________

Jim Ruehlin, IBM Rational

RUP Content Developer 

Eclipse Process Framework (EPF) Committer www.eclipse.org/epf

email:   [EMAIL PROTECTED]

phone:  760.505.3232

fax:      949.369.0720



________________________________

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] ]
On Behalf Of "Ronaldo r" <[EMAIL PROTECTED]>
Sent: Tuesday, June 05, 2007 7:00 AM
To: [email protected] 
Subject: [epf-dev] system requirements



The Supporting requirements concept mention that supporting requirements
+ use cases define the requirements of the system. This means that the
Glossary, conceptual model (entities in a class diagram) and Prototypes 
doesn't define the requirements of the system? Why this other
requirements are out?


-----
Supporting requirements and Use Cases, together, define the requirements
of the system. These requirements support the features listed in the 
Vision statement. Each requirement should support at least one feature,
and each feature should be supported by at least one to requirement

_______________________________________________
epf-dev mailing list 
[email protected]
https://dev.eclipse.org/mailman/listinfo/epf-dev

_______________________________________________ 
epf-dev mailing list
[email protected]
https://dev.eclipse.org/mailman/listinfo/epf-dev

 

_______________________________________________
epf-dev mailing list
[email protected]
https://dev.eclipse.org/mailman/listinfo/epf-dev

Reply via email to