Hi Guys,
I have a big question, at what point do you want to break up use cases... For example lets say your an administrator of a web-based application that requires users to "register" and "sign in" in order to use your web application. For an administrator of the web application there will have to be some sort of user management interface.
For instance I figure I would have the following system use cases...
Admin --> Ban User UC
Admin --> Add a User UC
Admin --> Delete a User UC
Admin --> Change User Password UC
or should it something a little bit more high level:
Admin --> Manager Web Application Users UC
The thing is, you could probably can do it both ways but have the second one with multiple Scenarios.. The problem with that is I see a Scenario as a way to accomplish the same thing in a different way (alternative flow) or to describe if something is wrong with the system (exceptional flow) and of course the main scenario being the basic flow.
My use case template looks like this...
1. Brief Description
2. Flow of Events
2.1 Basic Flow
2.2 Alternative Flow - Name
2.3 Exceptional Flow - Name
3. Pre conditions
4. Post Conditions
5. Special Requirements
Note that this is a SoDA templates and more information is imported from the model. What I think is funny is you normally have a "happy" scenario. If you go with the second option, you really can't just have one happy scenario (unless you use subflows??). Each use case does bring back an observable result to the actor.
Another quick example is say there are system configuration parameters. Should we have the following
Admin --> Enable Option 1
Admin --> Disable Option 1
Admin --> Configure Transaction Log Size
etc...
or
Admin --> Modify System Configuration
Again same idea... I have read many books and papers on use cases... I am just trying to check with other practitioners to see what their comments would be on this.
Sincerely,
Paul
Paul Vermette
Software Architect
Spielo Gaming International
1.506.852-7450
mailto:[EMAIL PROTECTED]
