That sounds like a good approach. However how would I name the partition object or would I just leave it blank.
Example:
Actor - Site Contributor (SC), set Partition's classifier to Site Contributor and leave the partition's name blank... Partition's displayed name would then ":Site Contributor"
Previously, I just named the Partition the same name as the actor it was associated with (for clarity on the activity diagram), but that now would lead to rather odd redundant looking results (Site Contributor :Site Contributor.)
It just seems "wrong" to leave the object's instance name blank... Is that true? <Looking over my shoulder for the UML Modeling police!>
Right now, I am developing project models to support requirement engineering involving Use Cases, Business Object Classes, Features, Requirements, Traces to other items such as Stakeholder Needs, Acceptance Criteria, UC dependencies, logical UI (prototypes) screen, tests, and test cases, expressed as UML objects or instances... BUT the underlying goal is to make sure that the model and it's elements have value to the downstream consumers (customer, testers, designers, developers, trainers, etc.)
I want to make sure that an approach I choose now, will not lead to unexpected issues later...