 |
 |
Ah, forget the first paragraph. I just found the section in the (very confusing - OWL DL? 2? ugh) documentation where they specifically allow ObjectProperty and class. But I do want to continue (or at least emphasize) the question of constraining the relators to ObjectProperties. I honestly do think that such a choice should be up to the folks using the vocabulary, based on their needs. If BIBFRAME wants to require IRIs as objects that's fine. But I see the LoC vocabularies as not being limited to BIBFRAME - or at least, I think that would be a good approach. YMMV. kc On 10/23/23 7:19 AM, Karen Coyle wrote: > Thanks, Kevin. My question, originally, was whether the typing > assigned can be seen as "OR" or "AND". I know that you can define SKOS > entities as objects and as properties and these are not seen as being > in conflict, but SKOS is very clear in defining this, making sure that > it is open. In the LoC case, it is an OWL declaration of > ObjectProperty and the class Role, a kind of punning. It seems to me > that all of the declarations are always attached to the subject, and > therefore using them as objects would trigger inferencing > inconsistencies (OWL tends to be strict). Have you tried that? Or are > you eschewing inferencing, as one often does. > > In any case, the big question was using the relators as properties and > the object as a string. There are folks who need to do that, and it is > a shame that there isn't an unconstrained version that would allow > this, since the LoC list is the most complete of all lists we can > find. Declaration as an rdf:Property would do that, and that would > entail less "rule" on the property definition, while users could > define their own more strict rules for their application. Again, this > brings up how far you can go with punning - adding rdf:Property to the > mix would probably just make things more confusing. > > I vote for simpler and less constrained at the vocabulary level, > leaving constraints to the application profile level, so everyone can > have the usage they need. > > kc > > > On 10/20/23 11:23 AM, Ford, Kevin wrote: >> Hi Karen, >> >> Steve is not wrong, but I think you are talking about two different >> things. >> >> Using a string with a Relators property would not conform to how they >> have been defined at ID.LOC.GOV. So, the answer to your specific >> question is: no, it is not our expectation Relator URIs would be used >> as properties with the object of the triple being either a URI or a >> string. Only URIs. >> >> But the Relators URIs have also been defined such that they can be >> used as a Property or as an Object, which is what Steve was driving >> at. We use them as Objects in Bibframe, hence their (additional) >> typing as a bf:Role. >> >> HTH, >> Kevin >> >> -- >> Kevin Ford >> Network Development and MARC Standards Office >> Library of Congress >> Washington, DC >> >> >> -----Original Message----- >> From: Code for Libraries <[log in to unmask]> On Behalf Of Karen >> Coyle >> Sent: Friday, October 20, 2023 11:41 AM >> To: [log in to unmask] >> Subject: Re: [CODE4LIB] [External] [CODE4LIB] Question about multiple >> declarations >> >> CAUTION: This email message has been received from an external >> source. Please use caution when opening attachments, or clicking on >> links. >> >> Steve, the list doesn't need to hear this, but you are not correct here. >> The relators are defined as owl:ObjectProperties (not just >> "properties") which means that they cannot take text as objects. >> However, I want LoC to confirm that, because this is their doing. >> >> kc >> >> >> On 10/17/23 8:17 AM, McDonald, Stephen wrote: >>> It is an inherent problem when creating a vocabulary--should this >>> set of traits be properties or types? Whichever choice you make, you >>> face the problem that other vocabularies may choose differently. I >>> believe this vocabulary defines relators as properties. But they >>> also want to show how the terms are related to terms in OWL and >>> BIBFRAME where they are defined as types. >>> >>> Steve McDonald >>> [log in to unmask] >>> >>> >>>> -----Original Message----- >>>> From: Code for Libraries <[log in to unmask]> On Behalf Of Karen >>>> Coyle >>>> Sent: Tuesday, October 17, 2023 10:40 AM >>>> To: [log in to unmask] >>>> Subject: Re: [CODE4LIB] [External] [CODE4LIB] Question about multiple >>>> declarations >>>> >>>> tl;dr: Does LoC intend that its relator properties be used with both >>>> "thing" and "string" objects? >>>> >>>> kc >>>> >>>> >>>> On 10/10/23 8:02 AM, McDonald, Stephen wrote: >>>>> That is not correct. The statement >>>>> <rdfs:subPropertyOf >>>>> rdf:resource="http://purl.org/dc/elements/1.1/contributor"/> >>>>> >>>>> is a single predicate-object statement, enclosed within angle >>>>> brackets. >>>>> The following statement >>>>> <rdf:type >>>>> rdf:resource="http://www.w3.org/2002/07/owl#ObjectProperty"/> >>>>> >>>>> is also separate statement, enclosed within angle brackets. The OWL >>>> statement is not part of the subPropertyOf statement. The next >>>> statement is also a separate statement. So we have three statements: >>>>> subPropertyOf: DC contributor >>>>> type: owl ObjectProperty >>>>> type: BIBFRAME role >>>>> >>>>> The term you were looking up is the implied subject of the >>>>> statements, >>>> making these RDF triples. >>>>> Steve McDonald >>>>> [log in to unmask] >>>>> >>>>> >>>>>> -----Original Message----- >>>>>> From: Code for Libraries <[log in to unmask]> On Behalf Of >>>>>> Karen Coyle >>>>>> Sent: Monday, October 9, 2023 5:36 PM >>>>>> To: [log in to unmask] >>>>>> Subject: [External] [CODE4LIB] Question about multiple declarations >>>>>> >>>>>> All, >>>>>> >>>>>> I am looking at the LoC relators at id.loc.gov, and am trying to >>>>>> understand the implications of the multiple declarations for >>>>>> relator terms. >>>>>> >>>>>> <rdfs:subPropertyOf >>>>>> rdf:resource="http://purl.org/dc/elements/1.1/contributor"/> >>>>>> <rdf:type >>>>>> rdf:resource="http://www.w3.org/2002/07/owl#ObjectProperty"/> >>>>>> <rdf:type >>>>>> rdf:resource="http://id.loc.gov/ontologies/bibframe/Role"/> >>>>>> >>>>>> dct:contributor is not an Object Property; there is no object type >>>>>> given, so I suppose it is de facto an Annotation Property. I read >>>>>> the next statement as narrowing, so at statement 2 we have: >>>>>> subproperty of dct:contributor AND an owl:ObjectProperty >>>>>> >>>>>> If my reading is correct, it would be a violation of this to use >>>>>> the relator with a string rather than a thing. >>>>>> >>>>>> (Stop me here if I'm wrong.) >>>>>> >>>>>> Then the 3rd statement appears to say that the relator is a >>>>>> bf:Role, which is a BIBFRAME-specific class. I can't wrap my head >>>>>> around the functionality of this statement and would love a brief >>>>>> explanation. >>>>>> I'm undoubtedly not into BIBFRAME deep enough to grok this. >>>>>> >>>>>> Also, my reading is that each relator is ALL THREE OF THESE; this >>>>>> is an AND not at OR. Right? >>>>>> >>>>>> Thanks for any help, >>>>>> kc >>>>>> >>>>>> -- >>>>>> Karen Coyle >>>>>> [log in to unmask] >>>>>> https://urldefense.com/v3/__http://kcoyle.net__;!!EDx7F7x-0XSOB8YS_ >>>>>> BQ!eHPXLOmgHd34Nkhl7hC1y1HksSXx1U6hRMICVD7hgM2VshIAMS7KC8rwlhpiRDMc >>>>>> J39slRBrXwrxVIJV$ >>>>>> m: +1-510-435-8234 >>>>>> skype: kcoylenet/+1-510-984-3600 >>>>>> >>>>>> Caution: This message originated from outside of the Tufts >>>>>> University organization. Please exercise caution when clicking >>>>>> links or opening attachments. When in doubt, email the TTS Service >>>>>> Desk at [log in to unmask]<mailto:[log in to unmask]> or call them directly >>>>>> at 617-627-3376. >>>> -- >>>> Karen Coyle >>>> [log in to unmask] >>>> https://urldefense.com/v3/__http://kcoyle.net__;!!EDx7F7x-0XSOB8YS_BQ >>>> !eHPXLOmgHd34Nkhl7hC1y1HksSXx1U6hRMICVD7hgM2VshIAMS7KC8rwlhpiRDMcJ39s >>>> lRBrXwrxVIJV$ >>>> m: +1-510-435-8234 >>>> skype: kcoylenet/+1-510-984-3600 >> -- >> Karen Coyle >> [log in to unmask] >> https://urldefense.com/v3/__http://kcoyle.net__;!!EDx7F7x-0XSOB8YS_BQ!eHPXLOmgHd34Nkhl7hC1y1HksSXx1U6hRMICVD7hgM2VshIAMS7KC8rwlhpiRDMcJ39slRBrXwrxVIJV$ >> > -- Karen Coyle [log in to unmask] http://kcoyle.net