Right now, the definitions of worlds of InXTargets are just those of the corresponding XTargets. This is fine, but it leads to a situation where we cannot partially select a handful of subcomponents of an X and reason about the connections to any other subcomponents not explicitly mentioned (without adding them to the lifelines).
We could extend the definition of these worlds to contain all subcomponents. This would fix this issue, but then mean that, with the way we currently resolve worlds, each subcomponent both exists in the world and also as a collection target. To fix this we would need to either:
make world resolution aware of which other actors are available in the current interaction (or I guess the group, but then we lose half of the point of this change); or
add a WFC that forbids references to things in the world which are already represented by another actor.
This seems a little clumsy, so I'd only want to go forward with this if it improves things.
Right now, the definitions of worlds of
InXTargets are just those of the correspondingXTargets. This is fine, but it leads to a situation where we cannot partially select a handful of subcomponents of anXand reason about the connections to any other subcomponents not explicitly mentioned (without adding them to the lifelines).We could extend the definition of these worlds to contain all subcomponents. This would fix this issue, but then mean that, with the way we currently resolve worlds, each subcomponent both exists in the world and also as a collection target. To fix this we would need to either:
This seems a little clumsy, so I'd only want to go forward with this if it improves things.