Andreas said:
> I felt the same way, when I contributed to the UML2 transition, as you
> know. But you came across like you wanted to do some experimental
> branch of the UML2 transition to see if (or how) you could do this
> with those diagram modules. It seemed to me, that you were not
> interested in contributions at that point.

I am interested in contributing to the euml model implementation I
simply don't feel qualified to lead. However in the absence of anyone
else I'm doing my best.

The reason I got involved with creating these diagrams was in response
to some emails I saw saying that there were too many conditions being
spread around regarding UML2/UML1.4. I simply wanted to demonstrate
that with the correct patterns in use that shouldn't be necessary.

In fact all the Figs I've committed so far extend the UML1.4 Figs with
no change and class and use case diagrams operate reasonably well. The
things I think will be a problem I can't test yet because of things
that are not yet implemented in euml. Hence my work there.

I'm also trying to make sense of how association navigability is
managed. The navigability seems to be reversed in euml and I'm trying
to understand why. Different docs describe navigability and different
UML2 tools generate different XMI for the same A->B model.

If this must be changed on the GUI side then its a good example for me
to demonstrate in the UML2 class diagram.

But if all the GUI needs to change because of this then I don't see
why we don't change the euml behaviour. Even if the euml
implementation is different the model interface should be consistent.

- I'm diverging - I'll start another thread for this.

I had got the impression that there'd be more people working on UML2
after 0.28 and it made sense for me to work to complete the XML
property panels at the same time. Right now though it feels like I'm
doing most everything (built on top of the hard labour of others of
course)

> At this point, we wanted to target the latest official release
> (read: UML 1.4). Everything else is too far away to depend on...

In that case I would advise that you simply continue that way.

All my suggestions have been for UML1.4 diagrams to continue to be
supported. Even if those Diagrams and Figs move to a module then they
would stay with the same package names and visibility. So just
including that jar means anything built against the current diagrams
should continue to work in the next release.

I'm hoping my previous suggestions will allow module developers such
as yourself some stability when working on your modules. We can then
deprecate old classes with instructions of how to upgrade to UML2 at
some later date.

Hopefully my suggestions will allow anyone to contribute whatever they
wish whether it is UML2, GEF separation or any other new feature or
bug fix to ArgoUML without having to wait for any milestone or fork
work. However, here's hoping we can drum up some more help with euml.

I'll post back later with some specific areas I've been looking at
that I'm unsure of.

Regards

Bob.

------------------------------------------------------
http://argouml.tigris.org/ds/viewMessage.do?dsForumId=450&dsMessageId=2382584

To unsubscribe from this discussion, e-mail: 
[[email protected]].
To be allowed to post to the list contact the mailing list moderator, email: 
[[email protected]]

Reply via email to