RE: How does RDF/OWL formalism relate to meanings?

> From: Thomas B. Passin
> Sent: Wednesday, April 07, 2004 9:20 PM
> 
> John Black wrote:
> 
> >>From: Thomas B. Passin
> > 
> > As I was asking Dan, why do I need to remember that a URI stands 
> > in some way (what way?) for a "resource"?  How does that relate 
> > to the meaning of the URI? 
> > 
> 
> We are talking primarily RDF here, and RDF uses URIs to identify the 
> things that it is interested in, and the things that can be 
> said about 
> them.  The RDF term for those things is "Resource".  By 
> "stands for", I 
> mean "identifies in an RDF statement".  Less formally, I 
> think of it as 
> "stands for the resource in a computer", though of course 
> people use the 
> URIs without computers, too, when they are talking about RDF data.

If you are saying that the Resource and my meaning are equivalent, and 
thus that the URI stands for my meaning, this would make sense.  Otherwise, 
this doesn't work.  I want my URIs to stand for what I intend to mean by 
them. But even so, I also want to know how the intelligent agents that 
receive the URIs in my RDF messages will correctly interpret my URIs 
and understand what I intended to signify by them.  How do I do that?

[snip]

> I would say at this point that it would help if you 
> came up with a small concrete example that could be a focal point for 
> more discussion, one that highlights why you have questions and more 
> about their nature.  The kind of responses you are seeking probably 
> depend on just what you really want to accomplish.

[snip]

I have written large systems in Prolog.  I have written parsers,  
a forward chaining production rules system, and I implemented 
a prototype of Conceptual Graphs, a system with many similarities to 
RDF.  And I implemented a prototype of an "agent" that ran 
continuously on computers in a network, waiting to receive messages, 
and then react to patterns in the message combined with data in a 
blackboard about the current state of the machine, by firing rules that 
caused actions, sent messages, and changed the state of the machine as 
recorded on the blackboard.  So I do know a little about some of this 
stuff.  And I have no illusions about the 'intelligence' of these programs 
or of their 'understanding' of the data copied back and forth, in spite 
of some of the neat things they could do.

But in my systems, both the sender and the receiver of all the data 
sent back and forth had the same parsers, and thus could interpret 
correctly any message sent or received.  To speak loosely with the 
same freedom about 'meaning' that Pat claims for himself, dangerous 
though it may be, each node understood the meaning of messages from 
the other nodes because I wrote the code for each node.

Now we are implementing a system that does not have that condition.  
It is created by vast numbers of different people from around the 
world. And we are planning to have lots of nodes on this wide network 
share documents with each other.  And we want to let these machines
process these documents in 'meaning preserving' ways.  This gives rise 
to certain questions about how those machines can preserve the 
that meaning.  Questions like these:

"Is a given inference engine expected to take into account a given 
document under given circumstances?"
http://www.w3.org/2001/tag/issues.html?type=1#rdfURIMeaning-39

and proposals like this:
"-  the architecture is that a single meaning is given to each URI (such 
as P), that the URI ownership system makes statements by owners 
authoritative weight, despite what other documents may say.
- the architecture does not permit the meaning of a URI to be changes 
by consistent misuse by others.
- that use of a URI in RDF implies a commitment to its ontology, and if 
there is doubt as to what ontology that is, the web may be used to 
resolve it."
http://lists.w3.org/Archives/Public/www-tag/2003Jul/0022.html

That have been hotly debated on this list.

I have thought a lot about this and came up with a specific proposal
in answer to some of these questions.  Namely that we work out how 
make this an optional facility, so that an author of some RDF can 
stipulate that in this case, this RDF will not make sense unless 
it is used in conjunction with exactly these ontologies.  And I offered 
as an example the history of RFC 2119, a natural language document 
that has been used in very similar way for many years, as a model.
In my opinion, part of what was wrong with the infamous Section 4.0,
sensibly removed, was that it proposed that these restrictions 
apply to *every* URI/RDF document/ontology.  What is new in my 
proposal is that this could be made explicit and optional.

John Black


> 
> Cheers,
> 
> Tom P
> 
>

Received on Thursday, 8 April 2004 15:05:03 UTC