[SCL] Re: new draft

[Jon Awbrey]

o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o

Bill Andersen wrote:
> 
> On Dec 19, 2003, at 1024, John F. Sowa wrote:
> 
> > Pat,
> >
> > The latest (Dec 19) draft looks much better, but
> > I haven't yet had a chance to study it in detail.
> 
> I agree.  It's much more readable than the previous version.
> Nice job,
> Pat!
> 
> > However, the use of the word "ontology" as a synonym
> > for any set of sentences still sets my teeth on edge.
> > I agree with you that many people use it as a meaningless
> > buzz word -- but just for that reason it is likely to
> > become obsolete as soon as the latest bubble (i.e. the
> > so-called semantic web) pops.
> >
> > I really wish you would do a global change of "ontology" to "module".
> 
> I agree with John here.  How about "theory" -- is that not the accepted
> term for a set of sentences intended to constrain the interpretation of
> a set of terms?  This might be too high-falootin a term for the general
> public, so I could live with "module".

i vote for "theory".  that is the most widely accepted term for an arbitrary
set of sentences of any order.  see Chang & Keisler for a standard account.
the relation "set of axioms for ___" is also very useful, and if one means
"finite set of axioms for ___", then it would probably be best to say so.

the term "module" bears the slight objection that it already has a standard
meaning in math, being the generalization of a "vector space over a field"
to a "module over a ring" .  but even math folk have learned to deal with
the fact that some words have context-sensitive meanings.  still, the
more that one uses category theory, the more this will be a problem.

ja

> > By the way, I am now writing a grammar for a very general
> > version of controlled English, which supports nearly the
> > full SCL semantics, including sequence variables.  The
> > only thing I haven't found a natural way of expressing
> > is the option of quantifying over relations.
> 
> John, would you send this to me?  I'm working with a parser generator
> to create SCL Java objects from concrete syntaxes and would like to
> see how this works.
> 
>    .bill
> 
> _______________________________________________
> SCL mailing list
> SCL at philebus.tamu.edu
> http://philebus.tamu.edu/mailman/listinfo/scl

o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
http://www.cs.bsu.edu/homepages/mighty/history.html
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o