Path: utzoo!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!ames!lll-tis!CUNYVM.CUNY.EDU!pv%sun
From: pv%sun@CUNYVM.CUNY.EDU
Newsgroups: comp.protocols.iso.x400
Subject: Re:  Is P3 implementable?
Message-ID: <8811162143.AA11436@polya.sun.com>
Date: 17 Nov 88 00:21:00 GMT
Sender: root@tis.llnl.gov
Distribution: inet
Organization: The Internet
Lines: 16
Approved: post-x400@tis.llnl.gov

>Posted-Date: Wed, 16 Nov 88 13:43:06 PST

>  I think that you are correct about your statements reagarding
>  P3. P3 is not a true client server protocol. To overcome most
>  of these problems you stated P7 was developed. My understanding
>  is that with P7 the client has full control.

P7 has a similar problem but on a smaller scale: the Alert operation is
invoked asynchronously by the message store and so, if implemented, the
client would have to act as slave for alerts at the same time it acted
as master for the rest of the operations.

Also, P7 implies a message store which must talk P3 to the MTS (unless
co-located) so you've really only pushed the problem back a notch.

Pete