Relationships with the Tag vocabulary
Alexandre Passant
I just discovered SCOT via the post on sioc-dev about int.ere.st, and
have a few questions.
First, what's the current relationship between SCOT and Newman tag
ontology [1] ?
When looking at SCOT specs, you define:
(prefix tag in for Newman's ontology)
scot:synonym rdfs:subpropertyOf tag:equivalentTag
scot:synonym rdfs:domain SCOT:Tag
scot:synonym rdfs:range SCOT:Tag
And there's already in [1]
tag:equivalentTag a owl:ObjectProperty ;
rdfs:subPropertyOf owl:sameAs ;
rdfs:domain :Tag ;
rdfs:range :Tag ;
which implies that tag:Tag is sameAs or a top class of SCOT:Tag if I'm right ?
It would also help to reuse (or create a subproperty of ) tag:name for
the name of a tag.
Secondly, regarding equivalent properties, if I understood right, they
are supposed to link instances of Tags re. similarity / differences of
the tag:name they have.
So in their rdfs:comment, I think it would semantically be more right
to say something like
"A tag which the name is a variation in the way this tag's name is spelled "
instead of
"A variation in the way in which a word is spelled"
as you link "Objects", not words.
It may also be useful to add some kind of "caseVariation" property.
Regards,
Alex.
Kim, Haklae
====
First, what's the current relationship between SCOT and Newman tag
ontology [1] ?
==
As I know, Newman's ontology is to represent tagging activities on
resources. So the ontology might be focused on "who tags a resource?"
that is, the core concept of the ontology is "tag" itself. While the
SCOT ontology is to describe the result of one's tagging activity. So,
the core concept of the SCOT is "tagcloud". I think it is possible to
connect Newman's ontology and SCOT.
As you mentioned here, I have reused the equivalentTag property. If
other property can be reused, I will do.
===
When looking at SCOT specs, you define:
(prefix tag in for Newman's ontology)
scot:synonym rdfs:subpropertyOf tag:equivalentTag
scot:synonym rdfs:domain SCOT:Tag
scot:synonym rdfs:range SCOT:Tag
And there's already in [1]
tag:equivalentTag a owl:ObjectProperty ;
rdfs:subPropertyOf owl:sameAs ;
rdfs:domain :Tag ;
rdfs:range :Tag ;
which implies that tag:Tag is sameAs or a top class of SCOT:Tag if I'm
right ?
It would also help to reuse (or create a subproperty of ) tag:name for
the name of a tag.
===
Yes. You're right. Actually tag:Tag and SCOT:Tag are a type of
skos:Concept.
Both cases make sense. I don't decide yet which one is better. But I
will change the specs.
===
Secondly, regarding equivalent properties, if I understood right, they
are supposed to link instances of Tags re. similarity / differences of
the tag:name they have.
So in their rdfs:comment, I think it would semantically be more right
to say something like
"A tag which the name is a variation in the way this tag's name is
spelled "
instead of
"A variation in the way in which a word is spelled"
as you link "Objects", not words.
It may also be useful to add some kind of "caseVariation" property.
===
Right. So the specs has the "scot:spellingVariant" property for
representing the purpose you mentioned.
<rdf:Description
rdf:about="http://scot-project.org/scot/ns#spellingVariant"
rdfs:label="spelling variants">
<rdfs:comment>A variation in the way in which a word is
spelled</rdfs:comment>
<rdf:type
rdf:resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#Property"/>
<rdfs:domain
rdf:resource="http://scot-project.org/scot/ns#Tag"/>
<rdfs:range
rdf:resource="http://scot-project.org/scot/ns#Tag"/>
<rdfs:isDefinedBy
rdf:resource="http://scot-project.org/scot/ns#"/>
</rdf:Description>
===
I think "caseVariation" is good. I will think about the name for the
property(spellingVariant vs caseVariation).
The spec has scot:acronym, scot:singular, scot:plural, and
scot:delimited properties as subproperty of scot:spellingVariant. The
scot:delimited also has subproperties such as hyphenated , underscored,
slashed, and spaced.
I will update the spec and int.ere.st as soon as possible and then I
will let you know.
Regards,
haklae