Re: Importing the correct version of an ontology

Hi Michael,

Interesting point that we had not considered. There wouldn't be a conflict for 
us because we do use fragment identifiers so the class URIs are all something 
like

http://docs.oasis-open.org/plcs/refdata#MyClass

but I do agree you've raised a general possibility that could lead to 
confusion if not conflict. 

We tried to blend the Dublin Core approach to Web addresses for versions of 
our ontology with the recommendations of not actually using version ids in 
the URIs and came up with the approach I outlined. I'll discuss your 
suggestion with our team. It's not too late to change things in our OASIS TC 
if a better approach can be found ... so if anyone has any other/better 
ideas, now would be a perfect time to speak up!

Cheers,
David

On Friday 21 September 2007 16:57:46 Michael Schneider wrote:
> Hi, David!
>
> David Price wrote on Friday, September 21:
> >For example:
> >
> >http://docs.oasis-open.org/plcs/refdata is the URI for the
> >ontology in all cases. It is also the Web address for the
> >latest and greatest ontology.
> >
> >http://docs.oasis-open.org/plcs/refdata/V1 contains the same
> >ontology when version 1 is published.
> >
> >When version 2 is published, then
> >
> >http://docs.oasis-open.org/plcs/refdata contains the version 2 ontology
> >
> >http://docs.oasis-open.org/plcs/refdata/V2 contains the same
> >version 2 ontology
> >
> >http://docs.oasis-open.org/plcs/refdata/V1 contains the
> >version 1 ontology.
> >
> >This lets ontology importing tools/users decide when to use
> >ontology V1 vs. V2 or whether to always use the latest and
> >greatest.
>
> I am just considering: Wouldn't it be better to use for those "frozen"
> ontology versions an URI scheme like in
>
>   http://docs.oasis-open.org/plcs/V1/refdata
>
> with the "V1" /before/ the "refdata" part. Or with the publishing data
> contained as in:
>
>   http://docs.oasis-open.org/plcs/2007/09/21/refdata
>
> like it is used for the RDF and OWL namespaces.
>
> My concern with your proposed ".../refdata/V1" scheme is that I would
> regard this more to be an URI for an OWL class named "V1" within the
> "OASIS" ontology (if slash-URIs are used). So, this might (while very
> unlikely to happen) collide with such a class name in the ontology. And,
> perhaps, ontology tools might get confused by it? I don't know, just have a
> somewhat bad feeling.
>
> On the other hand, the Dubling Core also uses the version number at the
> end, and so does FOAF:
>
>   http://purl.org/dc/elements/1.1/
>   http://xmlns.com/foaf/0.1/
>
> So, perhaps still not a problem?
>
> This might be somewhat pedantic, I suppose. :-) But I would like to use the
> opportunity of this thread to discuss some "best solution" for such a
> certainly pretty common scenario. Or is there already some "best practice"
> document out covering this topic?
>
> Cheers,
> Michael
>
> --
> Dipl.-Inform. Michael Schneider
> FZI Forschungszentrum Informatik Karlsruhe
> Abtl. Information Process Engineering (IPE)
> Tel  : +49-721-9654-726
> Fax  : +49-721-9654-727
> Email: Michael.Schneider@fzi.de
> Web  : http://www.fzi.de/ipe/eng/mitarbeiter.php?id=555
>
> FZI Forschungszentrum Informatik an der Universität Karlsruhe
> Haid-und-Neu-Str. 10-14, D-76131 Karlsruhe
> Tel.: +49-721-9654-0, Fax: +49-721-9654-959
> Stiftung des bürgerlichen Rechts
> Az: 14-0563.1 Regierungspräsidium Karlsruhe
> Vorstand: Rüdiger Dillmann, Michael Flor, Jivka Ovtcharova, Rudi Studer
> Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus

-- 
UK +44 20 8747 3900
Mobile +44 7788 561308
Skype +1 336 283 0606

Eurostep Limited. Registered in England and Wales No.03049099
Registered Office: Cwttir Lane, St. Asaph, Denbighshire LL17 0LQ.

Received on Friday, 21 September 2007 18:44:26 UTC