Hi,
There is no need to mention playmaker here, you are looking for a higher level support. If it can be done without playmaker, it can be done with playmaker. So don't mention it, else people will get misleaded as to what the question is about. It's about the likely technics to apply to run a multi player physics project.
Your link is clear about the fact that this is not really a supported solution, and it doesn't work in all cases. I would definitly not recommand doing this unless you find a clear documentation saying this is the way to go... ( cause I don't think it is really)
Instead, I strongly believe in the technic that I am currently using personally in my R.mote project. Which is the one I exposed in the previous post. Basically, you add a level of inderection. the object you want to control is NOT networked, it's inert. and each player has it's own controller, that you decide wether or not it controls that object. The benefit within R.mote is that I don't expose internal logic to the SDK ( the iphone app widgets prefab used to control the app is not there in the sdk!
typically you could do this with a character, your "player" is an empty gameobject with an fsm that has a reference to that character and controls is if it's yours or not.
I appreciate it's very abstract, I don't have time currently to create a specific sample for this, but I have added this to my todos for photon related work.
bye,
Jean