By separating them in two (as my second example) any part of the ArgoUML GUI can call the Notation subsystem whenever it chooses. The NotationGui subsystem, or whatever we should call it, can depend on GUI components without destroying the layered architecture.
Now I'm even more confused. You show a complex subsystem like diagrams as a package inside some GUI subsystem. Yet the few simple GUI classes require for notation, that should be accessible by the ArgoUML GUI, are shown as a subsystem that can't be accessed from that GUI. Why would the existing NotationGUI classes rely on the diagrams? Or the other way around? I can't see that they have any relationship with each other at all. Unfortunately all the suggested architecture diagrams I submitted so long ago I can't find any more. They were something along the lines of the attached, I've colour coded with pink as non-gui. In here I would consider those few GUI notation classes to be among the common available Actions and dialogs contained in some package within that subsystem. It may be that all GUI subsystems need to have access to some set of GUI utility classes. I've left this off for clarity. I've added things like ArgoNetbeans and ModelEMF to demonstrate the flexibility we should be prepared for. Regards Bob
architecture.zargo
Description: Binary data
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
