>From: Dik van Leeuwen POP [mailto:[EMAIL PROTECTED]]
>Why would you want the use cases to be ordered sequentially? In what
order?
>Order of creation? Order of ???
I do not specifically want them ordered sequentially. Although I do talk of
sequencing of use cases in my original post, that's really me complaining
about two things in the same posting ;)
Use cases can be numbered hierarchically, sequentially, randomly or using
names of famous french wine districts. It's really not important in terms of
achieving the "human readable unique identifier" for a use case. If me and
the people I work with were a little bit more machine-like, we could use the
RoseID. But that's really not human-readable. The actual name of the use
case "Buy coffee" is not formally precise enough: In different contexts, it
has a tendency to pop up in "everyday" prose in documents, and you cannot be
sure you're talking about the same artefact. And versioning is also quite
troublesome, although that's really another issue.
>I think it's more flexible to do without any kind of sequence numbers
>anywhere, including use cases.
Sequencing is really all a part of the "View" aspect of modelling, so I
quite agree with you on that. Solid human-readable identifiers are
attributes of the actual artefact in question, and there may actually be
cases where these could be useful for other artefacts too.
>All things are changing during a project; use cases get different names, a
>use case is added/inserted, etc.
>Sequence numbers are always a pain. At certain moments you would want to
>have a list of elements (be it use cases) listed in a certain order, but
the
>next day you would want to have them listed in another order, be it
>alphabetically, or (primary-)actor related and then alphabetically, or ...
Someone here has already pointed out that the intention is not automatic
tool support. We number the use cases manually. We may start out at UC100
for a specific role, then use UC200 for the next significant role, but
that's just because we use a lot of role modelling. If we want to version,
the next version may use 1.100 for new instance of the same use case. Maybe
we just call it 1000 or 433 instead ;) If we delete use case UC202, well
that's no longer a part of the project, we don't reuse the label for
something different.
So what I really would want was really just an additional ID field.
Unfortunately, it's not as simple as that, because I'd like to have them
show up on diagrams, and work as references for other tools too. Conceptual
integration in a project is an important idea here. *Tools* like strings
like 0x432ab64cd8484 as indentifiers, *people* need stuff like "2001.1.24
Buy Coffee" OR "24 Buy Coffee".
For those of you that know COM+, it's very much the same:
*A given COM+ component has a name (use case name)
*The same component has a globallu unique identifier (roseid)
*The component has a version dependent and a version independent progid,
which is a "probably-unique" identifier of the component.
Regards,
Kristian
************************************************************************
* 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
*
*************************************************************************
