W3C home > Mailing lists > Public > public-xg-lld@w3.org > July 2010

Re: MARC Codes for Forms of Musical Composition

From: Young,Jeff (OR) <jyoung@oclc.org>
Date: Sun, 04 Jul 2010 18:35:36 -0400
Message-ID: <002d01cb1bc9$31572000$d71dae84@oa.oclc.org>
To: "Antoine Isaac" <aisaac@few.vu.nl>
Cc: "Karen Coyle" <kcoyle@kcoyle.net>, "Ross Singer" <ross.singer@talis.com>, <public-xg-lld@w3.org>, "List for Working Group on Open Bibliographic Data" <open-bibliography@lists.okfn.org>, "public-lld" <public-lld@w3.org>
Oops, sorry. I do seem to have some wires crossed. Let me talk to Andy
Houghton to see if he can help sort out why I  believe multiple rdf:types
are bad for instances.

Jeff

Antoine Isaac <aisaac@few.vu.nl> wrote:

(moving this thread from [1] to the wider LLD list, this can be of
interest beyond the XG!)

Hi Jeff,

I'm not sure exactly why Ross' file is OWL Full, at least in the OWL 1
sense. In the validator you point to, I get the following output:


>     * Untyped Object Property: http://umbel.org/umbel#isAbout
>     * Untyped Object Property: http://purl.org/ontology/mo/wikipedia
>     * Untyped Data Property:
http://www.w3.org/2004/02/skos/core#prefLabel
>     * Untyped Class: http://purl.org/ontology/mo/Genre
>     * Untyped Individual: http://dbpedia.org/resource/Symphony
>     * Untyped Individual:
http://id.loc.gov/authorities/sh85131473#concept
>     * Untyped Individual: http://en.wikipedia.org/wiki/Symphony


It could be a aide effect of that specific validator's implementation,
which expects all ontological data to be present in the source at
validation time--remember that we're checking instance data here, while
the primary purpose of this validator is, I expect, ontologies.
Maybe if Ross had pulled the definitions for all the above constructs in
his file, the problem would have vanished.

Note that if you want to validate against the latest OWL2-DL, you can use
http://owl.cs.manchester.ac.uk/validator/
I've tried it, and it gives roughly the same results: in OWL2-DL you also
have to declare explicitly the resources that you're using...

Cheers,

Antoine

[1] http://lists.w3.org/Archives/Public/public-xg-lld/2010Jul/0000.html


> The http://purl.org/NET/marccodes/muscomp/sy.rdf example assumes OWL
> Full.
>
> http://www.mygrid.org.uk/OWL/Validator
>
> I think it would be better as OWL DL. This could be done by separating
> the various types into separate identities using hash URIs. If anyone is
> interested, I could amend the example to show how.
>
> As a rule, I like using OWL DL better than OWL Full because my brain
> doesn't fall out nearly as often.
>
> http://www.w3.org/TR/owl-ref/#Sublanguage-def
>
> Jeff
>
> -----Original Message-----
> From: public-xg-lld-request@w3.org [mailto:public-xg-lld-request@w3.org]
> On Behalf Of Karen Coyle
> Sent: Friday, July 02, 2010 10:05 AM
> To: Ross Singer
> Cc: public-xg-lld@w3.org; List for Working Group on Open Bibliographic
> Data
> Subject: Re: MARC Codes for Forms of Musical Composition
>
> Quoting Ross Singer<ross.singer@talis.com>:
>
>> Hi everybody,
>>
>> I just wanted to let people know I've made the MARC codes for forms of
>> musical compositions (
>> http://www.loc.gov/standards/valuelist/marcmuscomp.html) available as
>> http://purl.org/ontology/mo/Genres.
>
> Thanks, Ross. I looked at the RDA terms [1] and interestingly type of
> composition isn't one of the vocabularies that was defined in RDA. I
> don't know whether that was an oversight or not -- type of composition
> is included in the RDA rules, there's just no list to accompany it. So
> this one may end up doing double duty: MARC and RDA.
>
>      kc
> [1]http://metadataregistry.org/rdabrowse.htm
>
>>
>> http://purl.org/NET/marccodes/muscomp/
>>
>> They follow the same naming convention as they would in the MARC 008
> or 047,
>> so it's easy to map (that is, no lookup needed) from your MARC data:
>>
>> http://purl.org/NET/marccodes/muscomp/sy#genre
>>
>> etc.
>>
>> The RDF is available as well:
>> http://purl.org/NET/marccodes/muscomp/sy.rdf
>>
>>
>> I'd love any feedback/suggestions/corrections/etc.
>>
>> Also, you can look around to see MARC country codes, geographic area
> codes
>> and language codes.  Eventually I would like to get all of the MARC
> codes
>> (not already modeled by LC) in there (
>> http://www.loc.gov/standards/valuelist/).
>>
>> Thanks,
>> -Ross.
>>
>
>
>
Received on Sunday, 4 July 2010 22:36:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 4 July 2010 22:36:24 GMT