[Top] [All Lists]

Re: [ontolog-forum] Start thinking about the 2008 Ontology Summit

To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: "John F. Sowa" <sowa@xxxxxxxxxxx>
Date: Fri, 21 Dec 2007 23:11:31 -0500
Message-id: <476C8E73.6090500@xxxxxxxxxxx>
Pat,    (01)

There are two primary roles for CL:    (02)

  1. To serve as a target for translations to and from a wide
     range of logic-based languages.    (03)

  2. To provide a model-theoretic semantics for any notation
     that can be translated to CL (i.e., for any sentence s in
     any language L, and for any model M, the truth value of s
     in the model M shall be defined as the truth value in M
     of the CL statement to which s is translated).    (04)

 > Do you have a list of the open-source reasoning engines that
 > support any of the implementations of CL?    (05)

Any reasoning engine for any language that is translatable to CL
is a reasoning engine for that subset of CL.    (06)

Therefore, every reasoning engine for plain vanilla FOL, for OWL,
for Datalog, for any UML diagrams, etc., etc., etc., can be used
as a reasoning engine for CL.    (07)

And I mean that point *very* seriously.  Different subsets of FOL
can be supported by different reasoning engines with different
levels of efficiency.  Therefore, many reasoning engines are
actually implemented by a collection of specialized reasoning
engines for different subsets.  That technique has some advantages:    (08)

  1. A syntactic test to determine which subset of FOL a particular
     problem belongs to can be performed very rapidly -- usually in
     time proportional to the length of the problem statement.    (09)

  2. But reasoning engines usually take more time, sometimes polynomial
     time and sometimes even exponential time.    (010)

  3. Yet many important special cases can often be done very fast.    (011)

Therefore, it is useful to perform the syntactic check in order
to determine which reasoning engine to use for a given problem.    (012)

The beauty of having a common semantics for a wide range of different
dialects of logic is that a reasoning engine tailored to a dialect D1
is guaranteed to produce the same results as a reasoning engine for a
dialect D2 on any problem expressible in the intersection of D1 and D2.    (013)

Therefore, given a problem stated in any source dialect, syntactic
checks can be used to choose the best reasoning engine for that
problem -- independent of the dialect for which that reasoning
engine was originally designed.    (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>