Re: MISC: Internet Media Type registration: proposed TAG finding

On 22 May 2002, Dan Brickley wrote:
>On 22 May 2002, Dan Connolly wrote:
>
>>  >  I ask
>>  > that the WebOnt WG discuss whether to send a polite note back rejecting
>>  > this interpretation of our work.
>>
>>  I don't think we should.
>
>FWIW, Peter's dissatisfaction with my note (which wasn't addressed here)
>is noted.
>
>I continue to regard the WebOnt language (and the RDF 1.0 syntax, and it's
>MT, and RDFS) as a component of the wider Resource Description Framework,

What "wider RDF"?? I've heard phrases like this before, but they seem 
to refer to a secret W3.org ritual, because nobody is able to tell 
the rest of us what they are supposed to mean.

I take the phrase 'Resource Development Framework' to refer to a 
rather limited database language based on triples, as defined in the 
documents being produced now by the RDF Core WG. If it means 
something else, will someone PLEASE say CLEARLY what that other thing 
is?  I would like to know in case I'm supposed to be writing a model 
theory for it.

>but don't propose we take time up discussing labels here.
>(<onlyhalfjoking>We used to call this effort the Platform for Internet
>Content Selection; maybe we could go back to that name if folks really
>don't like the RDF TLA?</onlyhalfjoking>)
>
>[...]
>
>>  A consumer of the above document either or does or doesn't grok
>>  DAML+OIL semantics; it can come to more of the relevant conclusions
>>  if it applies DAML+OIL axioms, but since everything is monotonic,
>>  there's no harm done if it doesn't apply those axioms.
>>
>>  This is the principle of partial understanding in action.
>>  I have tried to make this point in the past...
>  >   http://lists.w3.org/Archives/Public/www-webont-wg/2002Mar/0339.html
>>  but I'm not having much luck.
>
>How about we try to think about this issue in forward-looking rather than
>backward-looking terms?
>
>Given RDFS and WebOnt, we're looking at partial understanding in terms of
>RDFS-aware tools dealing with with WebOnt-enriched RDF Schemas (er,
>Ontologies).

I'm not sure what that means, but I think it is wrong. That is, I 
would not expect an RDFS-aware tool (which, by the by, is more than 
an RDF-aware tool) to be able to handle WebOnt. (If it could, why are 
we bothering to develop WebOnt? We could just all use RDFS.) So an 
RDFS-aware tool will NOT be able to handle WebOnt-enriched RDF 
Schemas, even if (as seems highly unlikely) WebOnt could even be 
expressed as 'enriched' RDF Schemas.

>So how about we forget the past and look to the future?
>
>Imagine you're in the WebOnt v3.0 WG, looking back on the products of this
>group, balancing v3.0's backward compatibility with present-day
>requirements and opportunities. Presumably WebOnt v1.0 isn't the one true
>ontology language to end them all? We might expect a version 1.1 or 2.0 at
>least. Or perhaps people will take to describing their RDF Schemas and Web
>Ontology vocabularies using one of the various RDF-oriented rule
>languages. Maybe W3C will even do a REC-track spec or two for such a rule
>language. And what about datatyping? The XML Schema WG is still active,
>and might well produce refinements of the XML Schema datatyping system,
>which will at some point manifest itself in the RDF and Web Ontology
>world. The future looks busy.

Indeed.

>
>Partial understanding in action: people will write tools to work with the
>WebOnt 1.0 language, just as they're writing tools to work with RDF Schema
>vocabulary descriptions now. We need to think about how these new WebOnt
>tools will, or won't, be suprised by documents that draw on features
>defined in specs subsequent to WebOnt 1.0. Is a WebOnt ontology that draws
>upon some additional (webont v2, rdf-rules-1.0?) namespace still really a
>WebOnt doc?

Short answer, no. Full answer: you could take it either way. If it is 
really a WebOnt document then it won't be able to 'draw on' the 
additional (ie outside WebOnt) meaning used to define the terms in 
the other namespace. If it can 'draw on' that extra meaning, then it 
won't be a WebOnt document. The languages are different, and it has 
to be in one or the other. You can't have it both ways.

>Is it an RDF Schema for that matter? (re the latter, yes, imho).

No, similarly.  I would suggest that these answers are obvious. In 
CANNOT be an RDF Schema, in general, if WebOnt is an extension of 
RDFS; and if it isn't an extension of RDFS, then why are we even 
bothering to do it?

>At the instance data level, all this shouldn't matter. (Thankfully, for
>the poor end users...)

It has to matter. If someone marks up their webpage using WebOnt, 
then an RDF engine isn't going to be able to understand it, right? 
This isn't rocket science: all of computation is like this. If I 
write something in LISP, your Java interpreter isn't going to be able 
to make any sense of it either. Some web browsers can't handle CSS 
style sheets properly. And so on.

>
>A question. Or maybe even test case...
>
>Is the following XML doc 'mere RDF', or a 'WebOnt instance document'? (or
>a DAML+OIL doc). What changes in the Web might change our answers to this
>question?
>
><web:RDF xmlns:web="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>	xmlns:wn="http://xmlns.com/wordnet/1.6/" 
>xmlns="http://xmlns.com/foaf/0.1/">
>
>  <wn:Person>
>   <name>Dan Brickley</name>
>   <mbox web:resource="mailto:daniel.brickley@bristol.ac.uk"/>
>   <mbox web:resource="mailto:danbri@w3.org"/>
>   <homepage web:resource="http://purl.org/net/danbri/"/>
>   <dateOfBirth>1972-01-09</dateOfBirth>
>   <depiction 
>web:resource="http://rdfweb.org/people/danbri/2000/01/01/Image1.gif"/>
>  </wn:Person>
>
></web:RDF>
>
>Note that currently the RDF schema at the http://xmlns.com/foaf/0.1/
>namespace asserts that the 'mbox' property used here is a
>daml:UnambiguousProperty.

OK, that is *completely meaningless* in RDF(S). There are no 
unambiguous properties in those languages; the concept does not 
exist, and there is no way to express that idea. So apparently, you 
are using the above as though it were DAML (ie your engine is smart 
enough to interpret daml:UnambiguousProperty properly.) Good for you 
(though see below): but don't go around boasting about how smart 
RDF(S) is, please, or you may mislead some poor soul who has only 
read the RDF(S) specs and thinks that he is going to somehow get 
unambiguous property reasoning.

That document makes the following claim:

"
FOAF is expressed as an RDF Schema, annotated with DAML to express 
the fact that a foaf:mbox uniquely picks out an individual.

FOAF:mbox is an UnambiguousProperty, in the DAML+OIL sense:
"

How is a machine supposed to know that FOAF:mbox is unambiguous?  If 
the DAML is merely an annotation, I would think that is supposed to 
just be something like a comment; but the document claims to be 
"intended for machine consumption."  WHAT KIND OF MACHINE, Dan??

The only place that the machine-readable part of the document refer 
to daml is with this line in the definition of foaf:mbox:

<rdf:type 
rdf:resource="http://www.daml.org/2001/03/daml+oil#UnambiguousProperty"/>

This looks like an RDF assertion, not a DAML assertion. To an RDF(S) 
engine, this merely asserts that foaf:mbox is in a certain class with 
a name beginning with 'http://www.daml.org....'. It does NOT transfer 
the DAML content of that class name into RDF. In fact, it couldn't 
possibly do so, since that content is (provably) inexpressible in 
RDFS: there is no RDFS expression that could possibly restrict the 
RDFS interpretations so that a property was required to be 
unambiguous in all satisfying interpretations. So if this is RDFS and 
the engine reading your page is a conforming RDFS engine, it will not 
grok your intended meaning.  Even if this line somehow causes the 
entire daml+oil source page to be imported into the reading engine, 
it will appear there as RDFS, not as DAML+OIL, and that will not be 
enough to convey what you intend to convey.

On the other hand, if it were a conforming DAML+OIL engine, and if 
could figure out that you wanted this RDFS to be treated like DAML, 
it presumably would understand you better. So it appears that your 
document is actually written in DAML+OIL. Maybe it would be more 
honest, therefore, and also more likely to convey your intended 
meaning to some dumb software agent, if you had actually written it 
in DAML+OIL rather than cleverly disguising it as RDFS. (What were 
you trying to do, trick the machine into thinking it was simpler than 
it really is? )

>At some point it'll probably use WebOnt 1.0
>vocab instead. And eventually I'll use whatever ontology, rules and schema
>language best capture the intended meaning of the classes and properties
>in my namespace.

Well, if you use WebOnt then it will be a WebOnt document, rather 
than an RDFS document. (Why do I even need to say things like this, 
for God's sake?)

>Maybe I won't change the document you get at the
>namespace; I might send digitally signed RDF to a usenet group instead.
>But the intention should be clear: describe the vocabulary as accurately
>as possible with the machinery currently to hand.

But the whole point of this entire discussion is that 'the machinery 
currently to hand' keeps changing, and will continue to change. I 
thought that you were making that point yourself earlier (??)

Pat
-- 
---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes