W3C home > Mailing lists > Public > www-rdf-interest@w3.org > October 2002

Re: Meaning of URIRefs

From: Sandro Hawke <sandro@w3.org>
Date: Fri, 25 Oct 2002 18:17:49 -0400
Message-Id: <200210252217.g9PMHnq26405@wadimousa.hawke.org>
To: pat hayes <phayes@ai.uwf.edu>
cc: www-rdf-interest@w3.org


I need to call it a day and week, but let me leave with one small set
of questions for now:

> Well, it seems to me that there is a simple, clear story here.
> 1. RDF is a logic (an assertional language) , and entailment has its 
> usual meaning.
> 2. Urirefs 'belong' to their owners, and the owner is only 
> responsible for the sentences hse publishes. That is, owners are not 
> responsible for their urirefs, but they are responsible for their 
> triples (just like everyone else.)
> 3. In general, its good advice to only believe things published by 
> entities you trust; but it is not logically inconsistent to be either 
> paranoid or gullible.
> 
> The only special thing that needs to be said is that
> 
> 4. if you publish something which uses a uriref belonging to someone 
> else,  then it is your responsibility to ensure that the meaning you 
> intend it to have is consistent with the meaning its owner intends it 
> to have, ie expressed in whatever assertions hse makes using it.

So if animals says Cat and Dog are disjoint classes, and I say _:x is
an animal:Cat and an animal:Dog, what happens?  What does my message
mean?  Is my message false or in some way in error?

If I just say that _:spot is an animals:Dog, and animals (the document
at the URI) says that Dog is a subclass of Mammal, it in no way
follows that _:spot is an animals:Mammal?   Maybe the URI part of the
URIRef is a strong-hint for additional content to include?

If some other web document (not at the animals URI) says that
animals:Dog is a subclass of animals:Tree (disjoint from
animals:Mammals), does the reader have any basis for thinking that
_:spot is more legitimately an animals:Mammal than an animals:Tree?

(extra credit for answers which are easy to program into computers.  :-))

> 5. if you don't, and someone is misled as a result (ie if they draw 
> conclusions from what you wrote and what the author wrote which are 
> valid but which neither of you intended), then you are to blame (not 
> the owner of the uriref that you misused).

This kind of suggests the reader can use the content from the URIRef's
owner without knowing/trusting that owner, because the URIRef's user
is taking responsibility for any possible bad consequences of doing
so.   I'm not sure how that's different from what I'm suggesting,
except that it goes nowhere near words like the "true" and "entail".
Sigh.

     -- sandro
Received on Friday, 25 October 2002 18:18:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:56 GMT