Re: "Semantics" vs. "No Semantics"

From: Francois Bry <bry@ifi.lmu.de>
Subject: Re: "Semantics" vs. "No Semantics"
Date: Tue, 09 May 2006 10:18:54 +0200

> 
> Gary Hallmark wrote:
> >
> > I agree that there are many ways to approach semantics.  While I'm not
> > sure I would characterize model-theoretic semantics as more or less
> > abstract than other approaches, I do fear that a model-theoretic
> > semantics will be of little help
> > to the implementors of RIF translators and associated rule engines. 
> > Looking at other W3C formal semantic specifications for guidance, I
> > find http://www.w3.org/TR/xquery-semantics/ to be a good approach. 
> > Formal XQuery semantics are specified using RULES.  Why can't we
> > specify RIF semantics using rules?  We could even write those rules
> > using RIF.
> >
> 
> Look, I do not want to start a religious war about "abstract semantics".
> All what I mean is the following:
> 
> If S1 is a  semantics specifying more aspects of a language than a
> semantics S2 for this language, then S1 is more abstract than S2.

How is S1 more "abstract"?  Isn't it just "incomplete", or "vague", or
"unfinished"? 

> This is the standard meaning of "abstract" in Computer Science.

Not in my view, at least not without a whole lot of other caveats.

> Whether a model theopry helps or not implementi ng rule engines, is an
> important issue. Many (among others myself) think that a model theory
> helps avoiding implementations that diverge as of what they compute.
> 
> Implementations diverging on "how they compute" is good. Implementations
> diverging on "what they compute" is bad. This is a common view in
> stardizing.

Sure.

> All we need is a good way to properly specify the "what" and leave out
> enough of the "how" for people to realize not only one implementations
> but severla ones imporving over the foremr ones. Find suchy a good way,
> and, I am confident, it won't be difficult to re-express ist as model
> theory.

Well, I don't see how this is going to get to you to anything that is more
"abstract" than model theory.

> Another point: A Tarskian model theory is also specified using rules...

I don't think that looking at Tarskian model theory as a set of rules is an
interesting approach.

> François

peter

Received on Tuesday, 9 May 2006 09:02:27 UTC