Re: An ontology for KIF (James A. Fulton)
Message-id: <>
X-Sender: jfulton@grace
Date: Mon, 16 Aug 1993 11:41:37 -800
To: macgregor@ISI.EDU, interlingua@ISI.EDU
From: (James A. Fulton)
Subject: Re: An ontology for KIF
X-Mailer: <PC Eudora Version 1.1a4>
At 10:02 AM 8/16/93 -0800, macgregor@ISI.EDU wrote:
>It has been suggested that these extensions, to the extent that they
>can be agreed upon, should be placed in an ontology, rather than added
>by extending KIF itself (I happen to agree with that approach). For
>example, we could include in the ontology a definition for the concept
>"Type", defined as a partial (i.e., primitive) subrelation of "Unary
>Relation" (which would also be included in the ontology). Roles might
>be defined via the definition of a two-place function that maps from
>types/classes (if we had them) and relations to role objects.
>Let us hypothesize that we collectively agreed upon the specification
>of a (probably conservative) ontology that provided for at least some
>of the above-mentioned constructs. It would be a mistake to think of
>this ontology as an *optional* part of KIF, in the sense that it is just
>one of many possible auxiliary ontologies. To enable knowledge
>interchange, the ontology ought to be a fixed part of the interchange
>language. Why? Because translators for languages (such as Loom) that
>incorporate pieces of that ontology into their own scheme of
>representation would necessarily have to hardwire the connections
>between the ontology and their own language constructs. If there is a
>choice of more than one (most-fundamental) ontology, then we begin to
>revert to the "n squared" translators phenomenon whose avoidance was
>the whole point of KIF in the first place.
I am uncomfortable with this suggestion.  It seems equivalent to saying that 
formal logic is incomplete without a single ontology for its semantics.  
Both KIF and any other form of formal logic ought to be viewed as a language 
skeleton that gets fleshed out with a vocabulary of terms, predicates, etc., 
that are interpreted in terms of some ontology.  Communicating a set of KIF 
sentences requires users to share the ontology of the vocabulary used in 
those sentences.  An ontology is implemented in a language by axiomatizing 
the vocabulary.  A translation facility between KIF and another language 
must be able, for each transmission, to extract and implement both the 
axioms that formalize the ontology presupposed by the transmission and the 
factual content of the transmission. The more widely an ontology is shared, 
the less axiomatic reconfiguration will be needed between transmissions, but 
it is not necessary to wait on such an ontological nirvana for translators 
between KIF and other languages to be constructed. (I am a little skeptical 
that global ontologies will ever be precise enough to satisfy a significant 
portion of business communications.)  

I do not believe that this approach creates an "n-squared" problem. Each 
user in a network will presumably have (or evolve to) a single neutral 
ontology in terms of which all transmissions (in whatever languages the user 
is set up to receive) are interpreted and translated into the the various 
languages in the user's internal polyglot.  

a                                              h
v        Jim Fulton                            a
e        Research and Technology               v
a        The Boeing Company                    e
g        P.O. Box 24346, M/S 7L-10             a
o        Seattle, WA 98124                     g
o                                              o
d        Phone:          (206) 865-3151        o
d        Fax:            (206) 865-6903        d
a        Email:        d
y                                              a