SUO: Re: Comment #12 - 'Syntax for SUO' -- Proposed Resolution
Schoening, James R CECOM DCSC4I wrote:
>
> All,
>
> We appear to have consensus that SUO should be specified in
> a 'formal language,' so I propose we simply use those words
> and drop the word 'syntax.'
>
> Any objections?
>
> I still need to work this into the overall Scope.
>
> Jim Schoening
>
> -----Original Message-----
> From: Schoening, James R CECOM DCSC4I
> [mailto:James.Schoening@mail1.monmouth.army.mil]
> Sent: Friday, September 08, 2000 17:55
> To: Standard-Upper-Ontology (E-mail)
> Subject: SUO: Comment #12 - 'Syntax for SUO'
>
> SUO,
> Let's now discuss the issue of 'Syntax,' which received the
> following comments:
>
> The draft Scope read: "This standard will specify the syntax and semantics
> of a general-purpose upper level ontology"
> ----------------------------
> Bill Burkett commented:
> I think the SUO should be specified in a formal language to facilitate
> data processing. I don't think terms and definitions are enough for
> practical applications.
> -----------------------------
> Patrick Cassidy responded to a comment by Adam Pease:
> Adam's point regarding "language issues" needs some discussion:
> > [Adam Pease]
> > A language to emphasize frame statements. Personally, I'd like us not
> > to get hung up on language issues. The real point of this effort is
> > semantic content, not syntactic sugar. If folks want to create translation
>
> > tools to re-express the SUO automatically in various formats (frames,
> > pictures of graphs etc), I'm sure that will be helpful but doesn't really
> > seem central to our efforts.
> It's true that much unproductive time can be wasted in nit-picking about
> issues that are not central to a project, but the ease with a standard can
> be understood and used is an important matter when the standard is to be
> used for communication. The SUO standard that is proposed, if it is
> successful, will not be used merely by a few dozen adepts deeply versed in
> the arcana of philosophy and deontic logic, but tens of thousands of
> programmers who need to get up to speed and start using it for very
> practical purposes as quickly as possible. Any communication standard gains
> from widespread use, the more the better. It follows that *unnecessary*
> complications should be avoided, but if any notational device helps the
> learning process, it should be integrated with the standard itself, and not
> left for hapless initiates to try to discover and implement by themselves.
> Pat
> ---------------------------------
> Chris Menzel commented:
> "I find this a bit ambiguous. One might take the specification of a syntax
> and a semantics to be only the specification of a formal *framework* for an
> ontology, a specification of the general syntactic structure of an
> appropriate language along with a specification of the corresponding
> semantic structures. This could be accomplished with very little
> specification of any sort of *content*.
> So the question is: Is this clause calling for the specification of a
> general syntax and semantics only, or the specification also of the
> *content* of the ontology. In any case, I think these are separate tasks and
> should definitely be distinguished."
>
> -------------------------------------
> Adam Pease responded to some of these comments with:
>
> "I take syntax in this context broadly to mean the choice of a particular
> language for expressing the ontology. I'd interpret semantics to mean the
> definitions in logic of the terms in the ontology."
> -------------------------------------
> Matthew West; Matthew.R.West@is.shell.com commented:
> MW: "The syntax ..." suggests only one. I think it is the semantics that are
> important specified in such a way that they can be represented in one or
> more syntaxes. I think the discussion we have had reflects the need for
> this. There might be a base syntax selected (but there might not, most ISO
> standards are developed simultaneously in two languages, and this is seen as
> having a benefit in identifying hidden assumptions).
>
> -----------------------------------
>
> Let's discuss.
>
> Jim Schoening
>
<...>
¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤
Jim,
Not really an objection, as it depends on whether
your intended audience cares about the distinction --
But there is a shade of an idiomatic difference between
"specifying a formal language" and "specifying the syntax".
The way I learned it, a formal language is just a subset L
of A*, where A is a finite alphabet, but when people talk
about "specifying the syntax" they usually have in mind
the specification of a "formal grammar", and there is
a many-to-one relationship between formal grammars and
formal languages, strictly speaking. So, it depends
on whether you think anybody will care about that.
Jon Awbrey
¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤