Re: Proposal to close issue 5.6 - owl:imports

Jim,

To me, this proposal has the same effect as postponing the issue. We
still have a fundamental problem:

If there is no form of imports (whether implicit in namespace usage or
explicit), then the use of terms from an ontology is meaningless.

That is, if I have two documents:

doc1:
Man subClassOf Mortal

doc2a:
xmlns:d1="doc1"
Socrates type d1:Man

Then is Socrates type Mortal? We need to take a position on this. With
implicit imports, then the answer is yes. With explicit imports, the
answers is no. To get the same effect you would have to say: 

doc2b:
xmlns:d1="doc1"
doc2b imports doc1
Socrates type d1:Man

However, if the semantics do not specify which position we take, then we
lose interoperability. Two users may look at doc2a and disagree about
whether or not there is an implicit imports, and thus whether or not
doc2a implies Socrates is Mortal. The whole point of a formal semantics
is to make it absolutely clear who is right is such situations.
Therefore, I cannot support your proposal.

There have been various proposals on the table that give an umabiguous
semantics for imports. It seems like the one thing that worries some
people is the time-varying nature of the Web. 

I propose that we go with either my latest proposal [1] or Pat's take on
it [2], with the following additional wording to address time
dependency:

The Web is highly dynamic, and pages can change or disappear at any
time. However, a model theoretic semantics is used to assign meaning to
a static set of statements. To address this issue, the semantics for OWL
give meaning to a set of docuements with respect to the state of the Web
at some fixed point in time.

Jeff

[1] http://lists.w3.org/Archives/Public/www-webont-wg/2002Nov/0004.html
[2] http://lists.w3.org/Archives/Public/www-webont-wg/2002Nov/0099.html


Jim Hendler wrote:
> 
> To be able to reach closure on Issue 5.6 we must have a SPECIFIC
> proposal to discuss.  Since we do not have a specific wording on a
> semantics for import, I offer the following compromise position
> instead - please note, this is NOT a proposal to POSTPONE (sorry Dan)
> but rather to CLOSE the issue accepting an imports mechanism, leaving
> the semantic details undetermined at this time.
> 
> Proposal to close issue 5.6, owl:imports
> 
> 1) Change the name to owl:includes
> 2) Accept the wording from the reference document (copied below from
> [1]) with this name change
> 3) Accept the wording from the Guide document (copied below from [2])
> with this name change
> 4) Place a pointer on the issues list to the WOWG thread entitled "MT
> for imports (was: Re: Imports Proposal)" [3] with a note that the
> Model Theory, because of the time varying nature of the web and the
> difficulty of defining entailment with respect to multiple documents,
> does not specify an exact semantics for owl:includes at this time
> 
> Rationale: the current documents explain an inclusion mechanism in
> terms that are clear enough for implementors without being overly
> constraining.  The definitions below are similar to those in
> programming textbooks for includes, and no one seems to have trouble
> implementing and understanding them without a formal model theory.
>  i. this would go with our general principle of if we cannot decide
> "do it how D+O does"
>  ii. this embraces Ian's philosophy of "if it's research, avoid it" -
> It is clear from the email thread cited in [3]  that an includes style
> mechanism is not research, but formalizing it in MT is
>  iii. this proposal is a compromise between those wanting to avoid or
> Postpone having an imports mechanism, and those insisting we must
> formalize it now.
> 
> Note: this proposal does provide an imports mechanism, does not
> POSTPONE the issue, but does, in essence, postpone the formal
> semantics of imports which we seem unready to agree to at this time.
> 
> ------------
> 
> from [1]: Reference
> 
> Imports
> 
> Each owl:imports statement references another OWL ontology containing
> definitions that apply to the current OWL resource. Each reference
> consists of a URI specifying from where the ontology is to be imported
> from. See the example above. Imports statements are transitive, that
> is, if ontology A imports B, and B imports C, then A imports both B
> and C. Importing an ontology into itself is considered a null action,
> so if ontology A imports B and B imports A, then they are considered
> to be equivalent.
> 
> Note that namespaces only provide a mechanism for creating unique
> names for elements, and do not actually include definitions in the way
> that imports does. Similarly, imports statements do not set up a
> shorthand notation for names. Therefore, it is common to have imports
> statements that correspond to each namespace. However, additional
> imports may be used for ontologies that provide definitions without
> introducing any new names.
> 
> from [2]: Guide
> 
> <owl:imports> provides an include-style mechanism. <owl:imports> takes
> a single argument, identified by the rdf:resource attribute.
> 
> Importing another ontology brings the entire set of definitions
> provided by that ontology into the knowledge base. In order to make
> best use of this imported ontology it would normally be coordinated
> with a namespace declaration. Notice the distinction between these two
> mechanisms. The OWL namespace declaration provides a convenient means
> to reference names defined in other OWL ontologies. Conceptually,
> owl:imports is provided to indicate your intention to include the
> assertions of the target ontology. These assertions define the meaning
> of the terms defined in that ontology, meanings that support reasoning
> about the terms.
> 
> Note that owl:imports may not always succeed. As you would expect when
> dealing with the Semantic Web, access to resources distributed across
> the Web may not always be possible. Tools will respond to this
> situation in an implementation defined manner.
> 
> [1] http://www.w3.org/TR/owl-ref/
> [2] http://www.w3.org/TR/owl-guide/
> [3]
> http://lists.w3.org/Archives/Public/www-webont-wg/2002Nov/0099.html
> 
> --
> 
> Professor James Hendler                           hendler@cs.umd.edu
> Director, Semantic Web and Agent Technologies         301-405-2696
> Maryland Information and Network Dynamics Lab.      301-405-6707 (Fax)
> Univ of Maryland, College Park, MD 20742      240-731-3822 (Cell)
> http://www.cs.umd.edu/users/hendler

Received on Thursday, 14 November 2002 11:15:02 UTC