Hi Roger -

Thanks for your quick reply (below) to my note about an example of writing a Data Interoperability agent in a more human-readable notation than OWL.

What about generating OWL from the cleaner notation ?  Do you think that would be useful ?

BTW, here are the details of how to run the example OntologyInterop1 agent, the text of which is also appended below.

  1.  Point Netscape 7 or Mozilla to www.reengineeringllc.com

   2.  Click on the big font  Internet Business Logic   link

   3.  Click the "GO"

   4.  Select OntologyInterop1

   5.  At the top of the same page, "Choose an Agent and go to View or Change It"

   6.  On the next page choose "Go to the Question Menu"

From there on, please use the Help button to see how to navigate.

The SemanticWebOntology1 agent may also be of interest.

                Cheers,  -- Adrian
 
 
|    An example based on
|
|    'Enhancing Data Interoperability with Ontologies, Canonical Forms, and Include Files'
|
|              by Roger L. Costello     costello@mitre.org       August 10, 2003
|
|             (http://www.xfront.com/interoperability/CanonicalForms.html)

 
some-agent measures some-quantity in some-unit
not : the canonical measure of that-quantity is that-unit
one unit of that-quantity measured in that-unit converts to a canonical some-number some-standard-unit
-------------------------------------------------------------------------------------------------------------------
that-agent sending one unit of that-quantity in that-unit should first convert it to that-number that-standard-unit


this-agent measures this-quantity in this-unit
===============================================
Agent A             length           kilometers
Agent A             time             seconds
Agent A             speed            kph
Agent B             length           miles
Agent B             time             hours
Agent B             speed            mph


one unit of some-quantity measured in some-unit converts to a canonical some-number some-standard-unit
------------------------------------------------------------------------------------------------------
the canonical measure of that-quantity is that-standard-unit


one unit of this-quantity measured in this-unit converts to a canonical this-number this-standard-unit
======================================================================================================
             length                   miles                            1.609344      kilometers                    
             length                   meters                           0.001         kilometers                    
             time                     minutes                          60            seconds 
             time                     hours                            3600          seconds


one unit of length measured in miles converts to a canonical some-number1 kilometers
one unit of time measured in hours converts to a canonical some-number2 seconds
that-number1 / that-number2 = some-quotient
that-quotient rounded to 5 places after the decimal point is some-number
-------------------------------------------------------------------------------------
one unit of speed measured in mph converts to a canonical that-number kph



 

At 05:41 AM 8/11/03 -0400, you wrote:

Hi Adrian,

I believe that you are objecting to the OWL notation.  Yes, it is a bit
verbose.  Nothing I can do about that.  Your notation is very nice and
clean, but not standard.  Best stick with standards, aye?  /Roger

Adrian Walker wrote:
>
> Roger --
>
> Thanks for the pointer to your paper on Enhancing Data Interoperability.
>
> The idea seems very nice.  Yet the notation kind of overwhelms even the
> simple illustrative examples.  So, one wonders what real world examples
> would look like.
>
> I understand that the notation is intended to be machine- rather
> human-readable.  But someone, or something has to write the content into it
> in the first place.
>
> Attached is an example, based on your paper, in a different notation.
>
> I'm guessing that there is a way of generating the Owl/rdf notation from
> the attached example.  Do you think that would be useful ?
>
>                             Cheers,  -- Adrian
>
> PS: The example can be run by pointing a Netscape or Mozilla browser to
> www.reengineeringllc.com , then selecting
> the agent called OntologyInterop1 .
>
> At 10:19 AM 8/10/03 -0400, you wrote:
>
> >Hi Folks,
> >
> >Several weeks ago we had an excellent discussion on expressing the
> >relationship of entities that use different units-of-measure.  The more
> >general problem is - how do we express relationships that require a
> >transformation between the entities?
> >
> >Based upon our discussions I have synthesized an approach to expressing
> >transformation relationships, and an approach that applications can take
> >to use this relationship information to enhance interoperability.  Here
> >is the URL to the paper I have written:
> >
> >    http://www.xfront.com/interoperability/CanonicalForms.html
> >
> >Comments are welcome.  /Roger
>
>                             INTERNET BUSINESS LOGIC
>
>          Your English Business Rules Using Your Oracle Database
>
>                              www.reengineeringllc.com
>
> Adrian Walker
> Reengineering LLC
> PO Box 1412
> Bristol
> CT 06011-1412 USA
>
> Phone: USA 860 583 9677
> Cell:    USA  860 830 2085
> Fax:    USA  860 314 1029
>
>   ------------------------------------------------------------------------
>
>    OntologyInterop1.txtName: OntologyInterop1.txt
>                        Type: Plain Text (text/plain)


                           INTERNET BUSINESS LOGIC
 
        Your English Business Rules Using Your Oracle Database

                            www.reengineeringllc.com
             
Adrian Walker
Reengineering LLC
PO Box 1412
Bristol
CT 06011-1412 USA

Phone: USA 860 583 9677
Cell:    USA  860 830 2085
Fax:    USA  860 314 1029