W3C home > Mailing lists > Public > www-rdf-rules@w3.org > June 2005

Re: Web Rule Language - WRL vs SWRL

From: Bijan Parsia <bparsia@isr.umd.edu>
Date: Wed, 29 Jun 2005 23:18:42 -0400
Message-Id: <b9683803d0f10f7a5190ce2f1c04548c@isr.umd.edu>
Cc: www-rdf-rules@w3.org
To: drew.mcdermott@yale.edu

On Jun 29, 2005, at 10:29 PM, Drew McDermott wrote:

>> [Bijan Parsia]
>> The old Layering Story has been bankrupted in several different
>> ways.
>
> I'll take advantage of this crack in the orthodox Wall of Opprobrium

I don't know how you determine your factions, but this really makes no 
sense.

> here to jump to Michael Kifer's defense.  Someone should.

And this less. I think you tried to be too clever :)

(The Old Layering Story is that every semantic web language should be a 
same syntax semantic extension of RDF. In this current thread, I think 
only Jim ever strongly endorsed that approach, and most of the major 
proponents, afaict, have weakened or abandoned it. Peter has a paper in 
ICJAI showing that it is paradoxical if extended to FOL. I find it a 
PITA for OWL and have argued vehemently, with major grade vehemence for 
the truth of my PITA pain.

I suppose one could think that the family of views Ian, Jim, and I have 
been groping for represent a similar folly. I'd like to see some sort 
of evidence of that since I so dislike the other folly. I'm very 
skeptical.)

> For one thing, I find the argument against his position to sound a
> little like this one:
>
> "Don't mix coal dust with the pixie dust --- it burns smoky and
> pollutes the air."

Sigh. To clarify, it's two points:

	Don't claim that you are really mixing pixie dust with pixie dust if 
you aren't
	If we're going to mix pixie dust and coal dust, let's get it in the 
proportions which burn cleanest (or, better, get the most bang for the 
polluting buck)

> Yes, pixie dust is much better than coal, except that it doesn't
> exist.  I believe that there are applications in which the logical
> purity of Owl, DLP,

Sigh.

> etc. can be preserved, but (as Michael has
> implied, I think) there are many domains for which people want rules
> in order to express problem-solving strategies, and I have serious
> doubts that those will be expressible in pure FOL.  The reason why
> nonmonotonicity is (regrettable or not) ubiquitous in real-world
> applications is that most inference is not deductive.  Any attempt to
> reduce an inference task to deduction is either completely misguided,
> or ends up compromising the purity of the deductive notation.  Feel
> free to outlaw this practice; you won't succeed.  If the semantic web
> succeeds, one reason will be that someone implements a "killer app"
> rule system that solves a lot of realistic problems, and it will
> surely have a quasi-procedural --- and nonmonotonic ---
> interpretation.

Well, this is familiar from you, Drew, and good to have it on the 
table. However, I don't think either Ian nor I are arguing against 
non-mon *per se*. I might argue against quasi-proceduralism, but that 
doesn't put me on a different side than Michael, as far as I know.

[snip]

> This point may be obscured by the way LP is presented nowadays.  In
> the past two decades logic programming and logic have drifted apart,
> to the point where their are routinely characterized in basically
> different ways.

I believe model theory is quite standard for each. I believe the 
proposal on the table from Michael et al is for the Well Founded 
semantics. I'd be interested in what you thought of all that.

>   I think this is unfortunate.  I prefer to think of
> nonmonotonic systems as being "basically" ordinary logic, with an
> escape hatch here or there.  I grant that this is hard to formalize,
> but that's because it's precisely the nondeductive parts of inference
> that are hard to find a uniform formal framework for.

I wonder what you think of all the work in formal semantics for LP. 
Minimal model, perfect model, Well founded, answer set....

> But consider
> this hypothetical example: a logical system with procedural hooks to
> connect to a Bayesian inference system.  It might have a nonmonotonic
> rule to the effect that "If the best estimate of the probability of P
> is > 0.9, infer P."

General annotated logic programs...

> One could develop a pretty formal account of how
> this thing worked, including an account of what inferences were
> licensed under what circumstances.  But it would have nothing to do
> with the _semantics_, which could be specified in advance,
> independently of the details of the inferential mechanisms.

I'll be interested to know if Michael thinks this is a defense of him :)

I still stand with trying for a clean, expressive integration 
framework. If we are just going to throw code at things, fine, we can 
throw code. But then I don't really see what we're specifying (there 
are plenty of programming languages).

I guess I just find this too vague to get a grip on. What's your 
proposal? Is ISO Prolog a reasonable starting place? Some more modern 
logic/functional programming language? Should we just add a few things 
to XQuery?

I just don't know how to take your POV and generate useful 
standardization activities. Do you think there are none? (That's not 
unreasonable.)

Cheers,
Bijan Parsia.
Received on Thursday, 30 June 2005 03:18:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:53:12 GMT