Sure I just did not want Stefan to be disappointed when changes come forth. But also it might
be good for him to track changes in the bigbang where he can also give feedback. Stefan if you
are interested in that we'd be more than happy to help you while you do this.
I would say that we can let Stefan do the shemas, because as we will
have access to the .xmi files, we will be able to modify them
directly, not by redesign them.
Also, I would suggest we save the xmi files on subversion.
On 10/4/07, Alex Karasulu <firstname.lastname@example.org> wrote:
> Stefan this is a great thing to try to do however you might what to wait
> before you delve
> a bit deeper since many things may change especially where JNDI is concerned
> due to
> the big bang.
> Just a heads up so you don't waste cycles.
> On 10/3/07, Stefan Zoerner <email@example.com> wrote:
> > Hi all!
> > I have tried to start with a high level architecture overview of
> > ApacheDS with our new UML tool.
> > Currently it contains only a very simplistic System Context and a
> > partition into two sub systems. For the network sub system, I have
> > identified some components. For one of these components (the NTP
> > protocol provider), I have created a diagram which explains how the
> > artifacts (=jar files) manifest the components, and on which other
> > artifacts (third party and own artifacts) they depend on.
> > For those interested in Poseidon. I have attached the UML model itself
> > as well (file "ApacheDS Overview.zuml").
> > Questions to the architects:
> > Is the partition into two subsystem for the first level valid? I
> > discussed with Emmanuel about three as well (network, core, backend),
> > but at least in the Java sources, the partitions abstractions are part
> > of the core.
> > Do you think, the deployment diagrams with the artifact dependencies are
> > useful? I think they are, because the illustrate the information from
> > the pom.xml files in a very nice way ...
> > Greetings from Hamburg,
> > Stefan