I've followed the conversations around actors, and i'm really curious to see what the end-result will look like, considering the actor model seems to have been

I've followed the conversations around actors, and i'm really curious to see what the end-result will look like, considering the actor model seems to have been designed independantly from the execution infrastructure. In particular, the concepts of mailbox, executor, etc don't seem to be very apparent so far. By contrast, i see a lot of talk about actor "isolation", inheritance, reentrancy, etc... Which i don't see mentionned often in erlang / OTP style actor systems...