search

USGCRP / gcis-ontology

Ontology for the Global Change Information System
4 stars 7 forks source link

possible update gcis:hasVersion #18

Closed zednis closed 9 years ago

zednis commented 9 years ago

The domain of gcis:hasVersion is gcis:Model.

gcis:hasVersion a owl:DatatypeProperty ;
    rdfs:label "Has Version" ;
    rdfs:comment "A model may have a version." ;
    rdfs:domain gcis:Model ;
    rdfs:range xsd:string ;
    rdfs:subPropertyOf dcterms:hasVersion .

From the property's comment it is clear this property is intended specifically for the model version, but it has a very broad name and label.

I think we should do one of the following: 1) updating the property to be gcis:modelVersion 2) broadening the definition of gcis:hasVersion to not be just for model versions 3) removing the property entirely and using dcterms:hasVersion directly in the instance data.

rewolfe commented 9 years ago

I agree that the version domain should be for more than just the model. It is also applicable to datasets and activities (e.g. software version).

On Tue, Jun 16, 2015 at 5:33 PM, Stephan Zednik notifications@github.com wrote:

The domain of gcis:hasVersion is gcis:Model.

gcis:hasVersion a owl:DatatypeProperty ; rdfs:label "Has Version" ; rdfs:comment "A model may have a version." ; rdfs:domain gcis:Model ; rdfs:range xsd:string ; rdfs:subPropertyOf dcterms:hasVersion .

From the property's comment it is clear this property is intended specifically for the model version, but it has a very broad name and label.

I think we should do one of the following: 1) updating the property to be gcis:modelVersion 2) broadening the definition of gcis:hasVersion to not be just for model versions 3) removing the property entirely and using dcterms:hasVersion in the instance data.

— Reply to this email directly or view it on GitHub https://github.com/USGCRP/gcis-ontology/issues/18.

Robert Wolfe, NASA GSFC @ USGCRP, o: 202-419-3470, m: 301-257-6966

bduggan commented 9 years ago

I think we should do one of the following: 1) updating the property to be gcis:modelVersion 2) broadening the definition of gcis:hasVersion to not be just for model versions 3) removing the property entirely and using dcterms:hasVersion in the instance data.

My impression is that model versions are very distinct from one another, so much so that the versions are often built into the identifiers. Different model versions may have different attributes. Some examples are at http://data.globalchange.gov/model and also the draft health report appendix, table A1.5.

For these reasons "modelVersion" may be more appropriate.

zednis commented 9 years ago

How about option 3, removing the gcis:hasVersion from the ontology and using dcterms:hasVersion in the RDF?

edit: dcterms:hasVersion does not currently have a range and is not specified as an object property or as a datatype property. How about we keep gcis:hasVersion as we have specified it as a datatype property with range of string and instead remove the domain entirely so anything can be given a version?

So, how about option 2 ?

bduggan commented 9 years ago

On Wednesday, June 17, Stephan Zednik wrote:

How about option 3, removing the gcis:hasVersion from the ontology and using dcterms:hasVersion in the RDF?

"A related resource that is a version, edition, or adaptation of the described resource."

Would this mean that we would be undertaking to trace which models are versions of which other models? That would be interesting but could be hard (e.g. is CCSM4 a version of CESM1?) Another notion of version would just be a particular label/tag applied to a set of algorithms.

zednis commented 9 years ago

@bduggan good catch. On further look it appears that the dcterms:hasVersion is similar to the frbr and fabio relationships that connect (work -> expression -> manifestation -> item) and is not a simple datatype to express version information.

With this in mind I would not recommend using dcterms:hasVersion.

We could use <http://dbpedia.org/property/version> (see http://dbpedia.org/data4/version.n3)

bduggan commented 9 years ago

On Wednesday, June 17, Stephan Zednik wrote:

We could use <http://dbpedia.org/property/version> (see http://dbpedia.org/data4/version.n3)

That looks good.

justgo129 commented 9 years ago

fine here

On Wed, Jun 17, 2015 at 3:13 PM, Brian Duggan notifications@github.com wrote:

On Wednesday, June 17, Stephan Zednik wrote:

We could use <http://dbpedia.org/property/version> (see http://dbpedia.org/data4/version.n3)

That looks good.

— Reply to this email directly or view it on GitHub https://github.com/USGCRP/gcis-ontology/issues/18#issuecomment-112917299 .

Justin Goldstein, Ph.D. Advance Science Climate Data and Observing Systems Coordinator US Global Change Research Program 1717 Pennsylvania Ave NW, Suite #250 Washington, DC 20006

O: (202) 419-3496 M: (202) 285-3005

e-mail: jgoldstein AT usgcrp Dot gov http://www.globalchange.gov

rewolfe commented 9 years ago

Looks good.

On Wed, Jun 17, 2015 at 3:18 PM, justgo129 notifications@github.com wrote:

fine here

On Wed, Jun 17, 2015 at 3:13 PM, Brian Duggan notifications@github.com wrote:

On Wednesday, June 17, Stephan Zednik wrote:

We could use <http://dbpedia.org/property/version> (see http://dbpedia.org/data4/version.n3)

That looks good.

— Reply to this email directly or view it on GitHub < https://github.com/USGCRP/gcis-ontology/issues/18#issuecomment-112917299> .

Justin Goldstein, Ph.D. Advance Science Climate Data and Observing Systems Coordinator US Global Change Research Program 1717 Pennsylvania Ave NW, Suite #250 Washington, DC 20006

O: (202) 419-3496 M: (202) 285-3005

e-mail: jgoldstein AT usgcrp Dot gov http://www.globalchange.gov

— Reply to this email directly or view it on GitHub https://github.com/USGCRP/gcis-ontology/issues/18#issuecomment-112918764 .

Robert Wolfe, NASA GSFC @ USGCRP, o: 202-419-3470, m: 301-257-6966

zednis commented 9 years ago

I propose the following changes: 1) remove gcis:hasVersion from the GCIS ontology 2) update the RDF templates to use <http://dbpedia.org/property/version> (I need to check on the prefix)

edit - I think there is already general agreement on this, I just wanted to make the proposed change official.

bduggan commented 9 years ago

:+1:

justgo129 commented 9 years ago

I'm fine with it, but could someone confirm that the domains/ranges of < http://dbpedia.org/property/version> match our instance data? I don't see one listed there. Explicitly, are both floats and integers encompassed?

On Tue, Jun 23, 2015 at 4:00 PM, Brian Duggan notifications@github.com wrote:

:+1

— Reply to this email directly or view it on GitHub https://github.com/USGCRP/gcis-ontology/issues/18#issuecomment-114624305 .

Justin Goldstein, Ph.D. Advance Science Climate Data and Observing Systems Coordinator US Global Change Research Program 1717 Pennsylvania Ave NW, Suite #250 Washington, DC 20006

O: (202) 419-3496 M: (202) 285-3005

e-mail: jgoldstein AT usgcrp Dot gov http://www.globalchange.gov

zednis commented 9 years ago

@justgo129 This is what I got when I asked for the RDF of the property:

@prefix rdf:    <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix dbpprop:    <http://dbpedia.org/property/> .
dbpprop:version rdf:type    rdf:Property .
@prefix owl:    <http://www.w3.org/2002/07/owl#> .
dbpprop:version owl:sameAs  dbpprop:version .
@prefix rdfs:   <http://www.w3.org/2000/01/rdf-schema#> .
dbpprop:version rdfs:label  "version"@en .

It does not have a domain or range and is defined as a rdf:Property so it can be treated as an owl:ObjectProperty or owl:DatatypeProperty. it can have any rdfs:Literal as a value. The downside of this that it does not conform to OWL-DL.

If we want to use a property that confirms to OWL-DL then we should define the following:

gcis:version a owl:DatatypeProperty ;
  rdfs:label "version"@en ;
  rdfs:subPropertyOf dbpprop:version ;

That is OWL-DL and can have any rdfs:Literal value (string, integer, etc.)

justgo129 commented 9 years ago

Excellent. I'm fine with this then if @bduggan and @aulenbac are. Your explanation answered my question, Stephan -- thanks.

On Jun 23, 2015 4:16 PM, "Stephan Zednik" notifications@github.com wrote:

@justgo129 https://github.com/justgo129 This is what I got when I asked for the RDF of the property:

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dbpprop: http://dbpedia.org/property/ . dbpprop:version rdf:type rdf:Property . @prefix owl: http://www.w3.org/2002/07/owl# . dbpprop:version owl:sameAs dbpprop:version . @prefix rdfs: http://www.w3.org/2000/01/rdf-schema# . dbpprop:version rdfs:label "version"@en .

It does not have a domain or range and is defined as a rdf:Property so it can be treated as an owl:ObjectProperty or owl:DatatypeProperty. it can have any rdfs:Literal as a value. The downside of this that it does not conform to OWL-DL.

If we want to use a property that confirms to OWL-DL then we should define the following:

gcis:version a owl:DatatypeProperty ; rdfs:label "version"@en ; rdfs:subPropertyOf dbpprop:version ;

That is OWL-DL and can have any rdfs:Literal value (string, integer, etc.)

— Reply to this email directly or view it on GitHub https://github.com/USGCRP/gcis-ontology/issues/18#issuecomment-114630114 .

xgmachina commented 9 years ago

+1

zednis commented 9 years ago

resolved by #23