Hi everybody,   
      
   I've got a few questions about the KQML. I hope you can help me with   
   some of them or direct me to the appropriate information source:   
      
   -If I want two agents to communicate, they belong to different   
   platforms,   
   and one of them uses sockets for communication and the other one uses   
   RMI... Although both of them understand KQML, how can I make them talk?   
   It   
   is said that KQML is independent of the protocol use for message   
   exchange,   
   and I agree. But from a practical point of view, I think that not   
   specifying a "standard" network protocol makes it difficult for agents   
   of   
   different vendors to interoperate. Am I missing something here?   
      
   -I'm a bit confused about the fact that KQML does not enforce the   
   implementation of all the standard performatives (according to papers   
   by   
   the authors). How's this possible? How can then we ensure   
   interoperability? An agent could try to send to another one a   
   performative   
   it cannot understand.   
      
   -About the "completion condition" in KQML. I don't clearly see which is   
   the difference between this and the postconditions. Can anybody clarify   
   this to me?   
      
   -As far as I understand, KQML does not offer any support for agent   
   management, except for registering and unregistering agents (it has two   
   performatives for this). On the contrary, FIPA devotes a specification   
   to   
   these issues and also for creating and executing agents, etc. Am I   
   right?   
      
   Thanks in advance,   
      
   S.   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)