UML22 — attribute Activity.isSingleExecution

The attribute Activity.isSingleExecution creates, for activities, something akin to what the difference is between behavior as BehavioredClassifier.classifierBehavior vs. behavior as method of a behavioral feature. In the former case, there is exactly one execution during the life of the classifier (and the behavior would accept many tokens), while in the latter there is one execution for every invocation. Activity should not confusingly add another means of doing the same thing. At minimum, it should be explained what the effect of this attribute is when dealing with classifier behavior vs. method. In particular, it is very confusing what it would mean to have a behavior that is its own context and uses a separate execution for each token. Further, the definition of the attribute: "If true, tokens from separate invocations of the activity may interact" is completely meaningless as a definition.

see above