W3C home > Mailing lists > Public > public-owl-wg@w3.org > May 2009

Re: versioning in NF&R

From: Peter F.Patel-Schneider <pfps@research.bell-labs.com>
Date: Thu, 14 May 2009 14:15:10 -0400
Message-ID: <20090514.141510.176255795.pfps@research.bell-labs.com>
To: <ian.horrocks@comlab.ox.ac.uk>
CC: <public-owl-wg@w3.org>
From: Ian Horrocks <ian.horrocks@comlab.ox.ac.uk>
Subject: Re: versioning in NF&R
Date: Thu, 14 May 2009 17:27:41 +0100

> Looks good to me. I suggested a couple of possible changes below, but
> neither is critical.
> 
> Ian
> 
> 
> On 14 May 2009, at 15:44, Peter F.Patel-Schneider wrote:
> 
>> LCC J4 mentions that versioning is good and asks for a place where users
>> can find out about it.  However, NF&R doesn't mention this.
>>
>> I propose to add something like the following to NF&R.  If there is no
>> problem, I'll just go ahead and do it.
>>
>> peter
>>
>>
>>
>>
>>
>>
>> 2.6.3: Locating and Versioning OWL 2 Ontologies
>>
>> In OWL 1 ontologies can be stored as Semantic Web documents, and
>> ontologies can import other ontologies.  In OWL 2 this is slightly
>> cleared up, making it clear that ontology importing is by location.
> 
> "slightly cleared up, making it clear ..." sounds strange. How about:
> "This is the same in OWL 2, and it is made clearer that importing is by
> location."

I've fixed this up.

>> OWL 2 also clears up the relationship between an ontology name (IRI) and
>> its location and, in response to several requests, provides a simple
>> versioning mechanism by means of version names (IRIs).  Each OWL 2
>> ontology may have an ontology IRI, which is used to identify the
>> ontology. OWL 2 ontologies may also have version IRIs, which are used to
>> identify the version of the ontology.
>>
>> An OWL 2 ontology should be retrievable by using its version IRI, if it
>> has one.  One of the OWL 2 ontologies that share an ontology IRI (the
>> ontology versions) should be retrievable at the ontology IRI - this
>> ontology is the current one of the versions.  If it doesn't matter which
>> of the versions is desired then importing can use the ontology IRI, but
>> if a particular version is desired then the version IRI is used when
>> importing.
> 
> Is this last paragraph needed? Seems to me that it could be replaced by
> a pointer to the relevant section of SS&AS.

I think that NF&R has to say something about where ontologies are
supposed to be stored.  The above paragraph is too "dry", so I suggest
the following instead:

An OWL 2 ontology is stored at its version IRI and one of the ontologies
that share the ontology IRI is stored at the ontology IRI.  If it
doesn't matter which of the versions is desired then importing can use
the ontology IRI, but if a particular version is desired then the
version IRI is used.

peter
Received on Thursday, 14 May 2009 18:14:37 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:42:12 UTC