ontolog-forum
[Top] [All Lists]

Re: [ontolog-forum] The Open Group SOA Ontology

To: edbark@xxxxxxxx, "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: "John F. Sowa" <sowa@xxxxxxxxxxx>
Date: Fri, 18 Jul 2008 14:59:14 -0400
Message-id: <4880E802.1010408@xxxxxxxxxxx>
Ed,    (01)

The main point of my previous note was the pi-calculus, which
I should have highlighted up front instead of toward the end.    (02)

EB> Message passing is a nominal model for pretty much every
 > variety of communication that is recipient-specific, as
 > distinct from "broadcast-like".    (03)

It can handle broadcast-like communications as well.  See the
following architecture, which I published in 2002, but which
we have implemented in very flexible systems at VivoMind:    (04)

    http://www.jfsowa.com/pubs/arch.htm
    Architectures for Intelligent Systems    (05)

The basic idea for the Flexible Modular Framework (FMF) is that
every message has a destination field, which may be empty.  If
it is empty, the message goes to a Linda blackboard, from which
any agent that may be interested can pick up the message, if
it matches that agent's pattern of interests.    (06)

That approach supports a bulletin board or an eBay-like system,
among others.  After two agents become aware of each other, they
can avoid the blackboard by sending messages with a specified
destination.    (07)

EB> The important idea is that the request "message" is targeted
 > to a specific "role" in the system that is nominally played by
 > a single correspondent, who may or may not be known to the requestor.
 > And the message is a request for a service that the correspondent
 > is expected to provide.  And the response is a message directed to
 > the requestor that may include the provision of the service, or
 > just the notification that the service has occurred, will occur
 > or won't occur.    (08)

Yes, that is an important manner of usage that the FMF supports.
The pi-calculus is a formal theory that is used to analyze such
systems.    (09)

JFS>> There are many kinds of message types that
 >> are neither requests nor responses,    (010)

EB> And their relationship to an SOA is undefined.    (011)

Yes.  We are supporting "Societies of Agents", of which SOA
systems are special cases.    (012)

EB> But from an SOA point-of-view....    (013)

As I said, we support that point of view very nicely.    (014)

John    (015)


_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (016)

<Prev in Thread] Current Thread [Next in Thread>