FW: Heartburn with OWL and related technologies

>  -----Original Message-----
> From: 	Hardgrave, Terry (Contractor)  
> Sent:	Friday, May 20, 2005 5:01 PM
> To:	'public-webont-comments@w3.org'
> Subject:	Heartburn with OWL and related technologies
> 
> 
> OWL Team--
> 
> First of all, these comments are my personal opinion and not
> the opinion of either my employer (Pearson) or my client (DoED/FSA).
> 
> Furthermore, this is not just a critique of OWL, but of computer-science
> and information-technology in general.
> 
> The critique falls into two major categories:
> 
>       1.  Mis-use (and re-use) of common English terminology.
> 
>       2.  Failure to appreciate mathematics.
> 
> 
> Mis-use and re-use of English terms:  One problem with loading up English terms
> with new meanings is that it creates confusion both within the IT community and the
> English-speaking population at large.  Some other sciences (e.g. chemistry) invent
> brand new terms using Greek, Latin or some other methodology.  Computer scientists 
> would be well-advised to do something similar.
> 
> In your particular case, the English definition for "ontology" from the online Webster's 1828
> dictionary is:
> 
>          "That part of the science of metaphysics which investigates and explains the nature  
>           and essence of all beings, their qualities and attributes."
> 
> Here is your definition:
> 
>           "An ontology defines the terms used to describe and represent an area of knowledge."
> That's a far cry from the original definition. One "area of knowledge" is not "the nature and essence of
> all beings". A better definitional choice from English would have been either "classification" or
> "taxonomy", but better yet would have been an invented word e.g. "knowledgedomain" or some foreign
> phrase with similar connotation.
> 
> 
> Failure to appreciate mathematics:  Computer scientists constantly have the urge to re-invent
> set theory without really understanding the basics.  The whole object-oriented technology is
> a perfect example, but fouling it up in the definition of "classes".  Another example is the
> definition of "bags".  OWL  falls into this trap with the definition of "axioms" without really
> understanding the true reason why axioms need to exist in the definition of metalanguages.
> That is, OWL could be defined on top of set-theory, but as it stands, it is a poor substitute
> for set-theory -- and it does not set up a basis for proofs such as "Are two websites equivalent?"
> 
> My view is that it is just sloppy science.  But mathematics fell out of favor with computer science
> professionals sometime in the mid 1970s.  One reason I am writing this now its that it looks like we,
> as computer-scientists, are just digging ourselves into a bigger and bigger hole as new professionals 
> reinvent old technologies under new names and we roll the new "buzzwords" out on the community
> at large.
> 
> If w3.org is going to be setting the standards, perhaps it can enforce some discipline as well.
> 
> Just my humble opinion.
> 
> Thanks,
> 
> W. Terry Hardgrave
>