Re: Contexts and views

sowa@west.poly.edu (John F. Sowa)
Date: Thu, 1 Jun 1995 12:16:54 +0500
From: sowa@west.poly.edu (John F. Sowa)
Message-id: <9506011616.AA03312@west>
To: cg@cs.umn.edu, kr-advisory@isi.edu, macgregor@isi.edu, srkb@isi.edu
Subject: Re: Contexts and views
Cc: fikes@sumex-aim.stanford.edu, mrg@cs.stanford.edu, vishal@cs.stanford.edu
Content-Length: 1510
Sender: owner-srkb@cs.umbc.edu
Precedence: bulk
Bob,

Some comments:

>Our notion of contexts would seem to be rather different than John's.
>In our system, its not possible to look inside of a context from
>outside.  Instead, to "see" the facts, definitions, etc. in a context,
>you must be in it, or in one of its children.  This is in accord
>with normal scoping rules.

I agree with this restriction.  It's a basic assumption of CG contexts
as well.

>Also, we support
>multiple inheritance -- our context hierarchies are not limited
>to trees.  We have quite a few users (mostly in the NL domain)
>who use the multiple inheritance feature.  One way to reason
>with two different views/contexts is to create a new context
>that inherits both of them.

I'm not against multiple inheritance.  I suggested the Algol-like
tree structure because of its simplicity and the 30 years of practical
experience with it in conventional programming languages.  I think
that the tree-structured references of Algol are the minimal necessary
for knowledge sharing.  But the modules in Modula and the packages in


We have also identified (but not yet implemented) a need for
an export feature (lifting).  Often, one would like to reference some
of the objects in a context without inheriting that context.
That provides a second means for interviewing the contents
of a context without being in or below it.

- Bob



Robert M. MacGregor                                     macgregor@isi.edu
USC/ISI, 4676 Admiralty Way, Marina del Rey, CA 90292      (310) 822-1511