BIBO version 1.3
10 views
Skip to first unread message
Frederick Giasson
Nov 4, 2009, 5:03:14 PM11/4/09
to Bibliographic Ontology Specification Group
Hi all,
As mentioned previously, BIBO 1.3 has been released. I also added the
prism linkage according to the "prism" thread on the mailing list.
Here is the new Documentation:
http://bibotools.googlecode.com/svn/bibo-ontology/trunk/doc/index.html
(much smaller and simpler)
Here is the new OWL 2 xml file of the ontology:
http://bibotools.googlecode.com/svn/bibo-ontology/trunk/bibo.xml.owl
Here is the new OWL 2 turtle file of the ontology:
http://bibotools.googlecode.com/svn/bibo-ontology/trunk/bibo.n3.owl
Note that these XML and Turtle files have been generated using Protégé 4
Please tell me if you find any issues/error with this new release.
Thanks,
Take care,
Fred
Egon Willighagen
Nov 7, 2009, 11:46:01 AM11/7/09
to bibliographic-ontolog...@googlegroups.com
Hi Frederick,
On Wed, Nov 4, 2009 at 11:03 PM, Frederick Giasson <fr...@fgiasson.com> wrote:
> Here is the new Documentation:
> http://bibotools.googlecode.com/svn/bibo-ontology/trunk/doc/index.html
> (much smaller and simpler)
Looks nice. How was it generated? Are you using a custom tool, or some
library/program?
Egon
--
Post-doc @ Uppsala University
Homepage: http://egonw.github.com/
Blog: http://chem-bla-ics.blogspot.com/
PubList: http://www.citeulike.org/user/egonw/tag/papers
Niklas Lindström
Nov 8, 2009, 12:43:32 PM11/8/09
to bibliographic-ontolog...@googlegroups.com
Hi Fred!
Great to see a new release!
I have some questions though:
1. The SKOS namespace is still the non-reverted ("2008") one.
According to <http://www.w3.org/TR/skos-reference/#vocab>, it should
be <http://www.w3.org/2004/02/skos/core#>. (The reversion is described
e.g. at <http://www.w3.org/TR/2009/CR-skos-reference-20090317/#changes>.)
Shouldn't this be changed in bibo (or is bibo "locked" to the 2008
draft..)?
2. Every use of rdfs:isDefinedBy (and the two rdfs:seeAlso) have
xsd:anyURI-datatyped literals as objects instead of actual URI:s.
Isn't that an error?
3. The (bibo) "abstract" property is also stated to be
rdfs:isDefinedBy "http://purl.org/dc/terms/". I suspect it should be
<http://purl.org/ontology/bibo/>?
4. There are a lot of statements about terms from other vocabularies.
I suspect many are "imported" for reasoners and/or Protegé to
function? But at least rdf:value being an owl:ObjectProperty seems
wrong, doesn't it?
Best regards,
Niklas
Great to see a new release!
I have some questions though:
1. The SKOS namespace is still the non-reverted ("2008") one.
According to <http://www.w3.org/TR/skos-reference/#vocab>, it should
be <http://www.w3.org/2004/02/skos/core#>. (The reversion is described
e.g. at <http://www.w3.org/TR/2009/CR-skos-reference-20090317/#changes>.)
Shouldn't this be changed in bibo (or is bibo "locked" to the 2008
draft..)?
2. Every use of rdfs:isDefinedBy (and the two rdfs:seeAlso) have
xsd:anyURI-datatyped literals as objects instead of actual URI:s.
Isn't that an error?
3. The (bibo) "abstract" property is also stated to be
rdfs:isDefinedBy "http://purl.org/dc/terms/". I suspect it should be
<http://purl.org/ontology/bibo/>?
4. There are a lot of statements about terms from other vocabularies.
I suspect many are "imported" for reasoners and/or Protegé to
function? But at least rdf:value being an owl:ObjectProperty seems
wrong, doesn't it?
Best regards,
Niklas
Frederick Giasson
Nov 16, 2009, 9:14:24 AM11/16/09
to bibliographic-ontolog...@googlegroups.com
Hi Egon,
>> Here is the new Documentation:
>> http://bibotools.googlecode.com/svn/bibo-ontology/trunk/doc/index.html
>> (much smaller and simpler)
>>
>
> Looks nice. How was it generated? Are you using a custom tool, or some
> library/program?
>
It is a free protege addon called OWLDOC. You can get it from here:
http://code.google.com/p/co-ode-owl-plugins/downloads/list
Thanks,
Fred
>> Here is the new Documentation:
>> http://bibotools.googlecode.com/svn/bibo-ontology/trunk/doc/index.html
>> (much smaller and simpler)
>>
>
> Looks nice. How was it generated? Are you using a custom tool, or some
> library/program?
>
http://code.google.com/p/co-ode-owl-plugins/downloads/list
Thanks,
Fred
Frederick Giasson
Nov 16, 2009, 9:25:58 AM11/16/09
to bibliographic-ontolog...@googlegroups.com
Hi Niklas!
for noticing!
I never understood why they were putting dates in their URIs,
particularly if they switch back-and-forth :)
I guess it is why I like the new OWL 2 versioning system...
to read a couple of things, but it should means that this value is a
URI, so the URI of a resource (in the sense of an object property).
I think that everything will depends on how triple store & such system
would interpret a xsd:anyURI datatype (as a resource URI or a literal of
type xsd:anyURI?)
But to make it un-ambiguous, I will simply remove these datatypes :)
re-defining re-used terms within BIBO. The reason is that weird things
were imported, and imported ontologies were badly described in some
cases. For these reasons, I removed any import statement, and I
re-defined terms needed within BIBO.
The ObjectProperty/DataProperty errors have been created when I swtiched
between two prot�g� version. I thougt I fixed all of them, but
apparently I didn't! So I will fix that as well :)
So, I will fix this stuff, and will update soon!
Thanks,
Take care,
Fred
> Great to see a new release!
>
Yup :)
>
> I have some questions though:
>
> 1. The SKOS namespace is still the non-reverted ("2008") one.
> According to <http://www.w3.org/TR/skos-reference/#vocab>, it should
> be <http://www.w3.org/2004/02/skos/core#>. (The reversion is described
> e.g. at <http://www.w3.org/TR/2009/CR-skos-reference-20090317/#changes>.)
> Shouldn't this be changed in bibo (or is bibo "locked" to the 2008
> draft..)?
>
Yeah, I didn't notice that they switched back to their old NS. Thanks
>
> 1. The SKOS namespace is still the non-reverted ("2008") one.
> According to <http://www.w3.org/TR/skos-reference/#vocab>, it should
> be <http://www.w3.org/2004/02/skos/core#>. (The reversion is described
> e.g. at <http://www.w3.org/TR/2009/CR-skos-reference-20090317/#changes>.)
> Shouldn't this be changed in bibo (or is bibo "locked" to the 2008
> draft..)?
>
for noticing!
I never understood why they were putting dates in their URIs,
particularly if they switch back-and-forth :)
I guess it is why I like the new OWL 2 versioning system...
> 2. Every use of rdfs:isDefinedBy (and the two rdfs:seeAlso) have
> xsd:anyURI-datatyped literals as objects instead of actual URI:s.
> Isn't that an error?
>
This is a good question. I am not sure if it is an error. I should have
> xsd:anyURI-datatyped literals as objects instead of actual URI:s.
> Isn't that an error?
>
to read a couple of things, but it should means that this value is a
URI, so the URI of a resource (in the sense of an object property).
I think that everything will depends on how triple store & such system
would interpret a xsd:anyURI datatype (as a resource URI or a literal of
type xsd:anyURI?)
But to make it un-ambiguous, I will simply remove these datatypes :)
> 3. The (bibo) "abstract" property is also stated to be
> rdfs:isDefinedBy "http://purl.org/dc/terms/". I suspect it should be
> <http://purl.org/ontology/bibo/>?
>
Right
> rdfs:isDefinedBy "http://purl.org/dc/terms/". I suspect it should be
> <http://purl.org/ontology/bibo/>?
>
> 4. There are a lot of statements about terms from other vocabularies.
> I suspect many are "imported" for reasoners and/or Proteg� to
> function? But at least rdf:value being an owl:ObjectProperty seems
> wrong, doesn't it?
>
Yes and not. I stopped using the owl:import statement and was simply
> wrong, doesn't it?
>
re-defining re-used terms within BIBO. The reason is that weird things
were imported, and imported ontologies were badly described in some
cases. For these reasons, I removed any import statement, and I
re-defined terms needed within BIBO.
The ObjectProperty/DataProperty errors have been created when I swtiched
between two prot�g� version. I thougt I fixed all of them, but
apparently I didn't! So I will fix that as well :)
So, I will fix this stuff, and will update soon!
Thanks,
Take care,
Fred
Niklas Lindström
Nov 17, 2009, 9:22:22 AM11/17/09
to bibliographic-ontolog...@googlegroups.com
Hello Fred!
>> 1. The SKOS namespace is still the non-reverted ("2008") one.
>> According to <http://www.w3.org/TR/skos-reference/#vocab>, it should
>> be <http://www.w3.org/2004/02/skos/core#>. (The reversion is described
>> e.g. at <http://www.w3.org/TR/2009/CR-skos-reference-20090317/#changes>.)
>> Shouldn't this be changed in bibo (or is bibo "locked" to the 2008
>> draft..)?
>>
> Yeah, I didn't notice that they switched back to their old NS. Thanks
> for noticing!
>
> I never understood why they were putting dates in their URIs,
> particularly if they switch back-and-forth :)
Yes, switching is certainly confusing. (I can buy the "time of
minting" as a way to partition a URI space, but other than that
ontology URI:s reasonably have to be static and persistent until
something different/backwards-incompatible comes along to replace that
vocabulary.)
> I guess it is why I like the new OWL 2 versioning system...
>> 2. Every use of rdfs:isDefinedBy (and the two rdfs:seeAlso) have
>> xsd:anyURI-datatyped literals as objects instead of actual URI:s.
>> Isn't that an error?
>>
> This is a good question. I am not sure if it is an error. I should have
> to read a couple of things, but it should means that this value is a
> URI, so the URI of a resource (in the sense of an object property).
As they're given, they are only typed literals -- i.e. a self-denoting
thing which doesn't work as a resource identifier as real URI:s to
(the ones enclosed in <...> in turtle). It's a subtle but important
difference.
> I think that everything will depends on how triple store & such system
> would interpret a xsd:anyURI datatype (as a resource URI or a literal of
> type xsd:anyURI?)
>
> But to make it un-ambiguous, I will simply remove these datatypes :)
Sounds good. :) That is, if you mean change them from literals to
URI:s? I.e., in turtle, from:
rdfs:isDefinedBy "http://purl.org/ontology/bibo/"^^xsd:anyURI
to:
rdfs:isDefinedBy <http://purl.org/ontology/bibo/>
>> 4. There are a lot of statements about terms from other vocabularies.
>> I suspect many are "imported" for reasoners and/or Protegé to
recently came upon this interesting approach to it:
<http://sparqlpedia.org/rdfex/>
I figure one could just as well manually "cherry-pick" terms into a
custom resource and import that. I'm not sure about the soundness of
constructing "subset ontologies" and importing those though. Just some
thoughts. *If* it seems ok, perhaps Bibo would benefit from extracting
the external parts into just that, a "bibo-ext" vocabulary, and import
that instead of directly including the parts in the source of bibo? (I
think one should ask the semweb list or pendantic-web or similar..
Wasn't there some de-facto official "rdfs/owl vocabulary developer
list" set up somewhere?)
The strangeness in either case is that the imported stuff aren't
defined by the imported resource, so there would be no semantic
relation to it at all. And I don't really think owl:imports
semantically means an (imperative) "include", but more like "uses
definitions from". (The same thing really goes for the currently
inlined stuff; albeit they're part of the same "graph" resulting from
parsing the document..)
However, I think the weird things you've found in those ontologies
should be reported to their authors (and CC:ed to the
<http://pedantic-web.org/> group, which is dedicated to all things
strange in public RDF), to ideally be able to relate bibo to them via
just owl:imports.
> So, I will fix this stuff, and will update soon!
Great; thanks! (I'd be happy to check out a draft or similar if you want to.)
Best regards,
Niklas
>> 1. The SKOS namespace is still the non-reverted ("2008") one.
>> According to <http://www.w3.org/TR/skos-reference/#vocab>, it should
>> be <http://www.w3.org/2004/02/skos/core#>. (The reversion is described
>> e.g. at <http://www.w3.org/TR/2009/CR-skos-reference-20090317/#changes>.)
>> Shouldn't this be changed in bibo (or is bibo "locked" to the 2008
>> draft..)?
>>
> Yeah, I didn't notice that they switched back to their old NS. Thanks
> for noticing!
>
> I never understood why they were putting dates in their URIs,
> particularly if they switch back-and-forth :)
minting" as a way to partition a URI space, but other than that
ontology URI:s reasonably have to be static and persistent until
something different/backwards-incompatible comes along to replace that
vocabulary.)
> I guess it is why I like the new OWL 2 versioning system...
>> 2. Every use of rdfs:isDefinedBy (and the two rdfs:seeAlso) have
>> xsd:anyURI-datatyped literals as objects instead of actual URI:s.
>> Isn't that an error?
>>
> This is a good question. I am not sure if it is an error. I should have
> to read a couple of things, but it should means that this value is a
> URI, so the URI of a resource (in the sense of an object property).
thing which doesn't work as a resource identifier as real URI:s to
(the ones enclosed in <...> in turtle). It's a subtle but important
difference.
> I think that everything will depends on how triple store & such system
> would interpret a xsd:anyURI datatype (as a resource URI or a literal of
> type xsd:anyURI?)
>
> But to make it un-ambiguous, I will simply remove these datatypes :)
URI:s? I.e., in turtle, from:
rdfs:isDefinedBy "http://purl.org/ontology/bibo/"^^xsd:anyURI
to:
rdfs:isDefinedBy <http://purl.org/ontology/bibo/>
>> 4. There are a lot of statements about terms from other vocabularies.
>> function? But at least rdf:value being an owl:ObjectProperty seems
>> wrong, doesn't it?
>>
> Yes and not. I stopped using the owl:import statement and was simply
> re-defining re-used terms within BIBO. The reason is that weird things
> were imported, and imported ontologies were badly described in some
> cases. For these reasons, I removed any import statement, and I
> re-defined terms needed within BIBO.
I see. I've had thoughts about how to solve such things myself. I
>> wrong, doesn't it?
>>
> Yes and not. I stopped using the owl:import statement and was simply
> re-defining re-used terms within BIBO. The reason is that weird things
> were imported, and imported ontologies were badly described in some
> cases. For these reasons, I removed any import statement, and I
> re-defined terms needed within BIBO.
recently came upon this interesting approach to it:
<http://sparqlpedia.org/rdfex/>
I figure one could just as well manually "cherry-pick" terms into a
custom resource and import that. I'm not sure about the soundness of
constructing "subset ontologies" and importing those though. Just some
thoughts. *If* it seems ok, perhaps Bibo would benefit from extracting
the external parts into just that, a "bibo-ext" vocabulary, and import
that instead of directly including the parts in the source of bibo? (I
think one should ask the semweb list or pendantic-web or similar..
Wasn't there some de-facto official "rdfs/owl vocabulary developer
list" set up somewhere?)
The strangeness in either case is that the imported stuff aren't
defined by the imported resource, so there would be no semantic
relation to it at all. And I don't really think owl:imports
semantically means an (imperative) "include", but more like "uses
definitions from". (The same thing really goes for the currently
inlined stuff; albeit they're part of the same "graph" resulting from
parsing the document..)
However, I think the weird things you've found in those ontologies
should be reported to their authors (and CC:ed to the
<http://pedantic-web.org/> group, which is dedicated to all things
strange in public RDF), to ideally be able to relate bibo to them via
just owl:imports.
> So, I will fix this stuff, and will update soon!
Best regards,
Niklas
Jay Luker
Nov 18, 2009, 11:24:29 AM11/18/09
to bibliographic-ontolog...@googlegroups.com
Hi,
I noticed that the definition of bibo:identifier has changed and is no
longer a subPropertyOf dcterms:identifier. Was this an error?
v1.3
### http://purl.org/ontology/bibo/identifier
:identifier rdf:type owl:DatatypeProperty ;
rdfs:range rdfs:Literal ;
rdfs:domain [ rdf:type owl:Class ;
owl:unionOf ( :Collection
:Document
)
] .
v1.2
### http://purl.org/ontology/bibo/identifier
:identifier rdf:type owl:Class ,
owl:ObjectProperty ;
rdfs:subPropertyOf terms:identifier ;
owl:objectPropertyDomain :identifier .
Thanks,
--jay
> --~--~---------~--~----~------------~-------~--~----~
> You received this message because you are subscribed to the Google Groups "Bibliographic Ontology Specification Group" group.
> To post to this group, send email to bibliographic-ontolog...@googlegroups.com
> To unsubscribe from this group, send email to bibliographic-ontology-spec...@googlegroups.com
> For more options, visit this group at http://groups.google.com/group/bibliographic-ontology-specification-group?hl=en
> -~----------~----~----~----~------~----~------~--~---
>
>
I noticed that the definition of bibo:identifier has changed and is no
longer a subPropertyOf dcterms:identifier. Was this an error?
v1.3
### http://purl.org/ontology/bibo/identifier
:identifier rdf:type owl:DatatypeProperty ;
rdfs:range rdfs:Literal ;
rdfs:domain [ rdf:type owl:Class ;
owl:unionOf ( :Collection
:Document
)
] .
v1.2
### http://purl.org/ontology/bibo/identifier
:identifier rdf:type owl:Class ,
owl:ObjectProperty ;
rdfs:subPropertyOf terms:identifier ;
owl:objectPropertyDomain :identifier .
Thanks,
--jay
> You received this message because you are subscribed to the Google Groups "Bibliographic Ontology Specification Group" group.
> To post to this group, send email to bibliographic-ontolog...@googlegroups.com
> To unsubscribe from this group, send email to bibliographic-ontology-spec...@googlegroups.com
> For more options, visit this group at http://groups.google.com/group/bibliographic-ontology-specification-group?hl=en
> -~----------~----~----~----~------~----~------~--~---
>
>
Frederick Giasson
Nov 18, 2009, 2:10:51 PM11/18/09
to bibliographic-ontolog...@googlegroups.com
Hi,
> I noticed that the definition of bibo:identifier has changed and is no
> longer a subPropertyOf dcterms:identifier. Was this an error?
>
> v1.3
> ### http://purl.org/ontology/bibo/identifier
> :identifier rdf:type owl:DatatypeProperty ;
> rdfs:range rdfs:Literal ;
> rdfs:domain [ rdf:type owl:Class ;
> owl:unionOf ( :Collection
> :Document
> )
> ] .
>
> v1.2
> ### http://purl.org/ontology/bibo/identifier
> :identifier rdf:type owl:Class ,
> owl:ObjectProperty ;
> rdfs:subPropertyOf terms:identifier ;
> owl:objectPropertyDomain :identifier .
>
yes, will fix with other found issues.
> I noticed that the definition of bibo:identifier has changed and is no
> longer a subPropertyOf dcterms:identifier. Was this an error?
>
> v1.3
> ### http://purl.org/ontology/bibo/identifier
> :identifier rdf:type owl:DatatypeProperty ;
> rdfs:range rdfs:Literal ;
> rdfs:domain [ rdf:type owl:Class ;
> owl:unionOf ( :Collection
> :Document
> )
> ] .
>
> v1.2
> ### http://purl.org/ontology/bibo/identifier
> :identifier rdf:type owl:Class ,
> owl:ObjectProperty ;
> rdfs:subPropertyOf terms:identifier ;
> owl:objectPropertyDomain :identifier .
>
Thansk!
Fred
Niklas Lindström
Feb 6, 2011, 7:43:05 AM2/6/11
to bibliographic-ontolog...@googlegroups.com
Hello!
Is there any news on releasing a new Bibo version with the issues of
the old SKOS namespace and the use of rdfs:isDefinedBy with literals
fixed?
Best regards,
Niklas
2009/11/17 Niklas Lindström <linds...@gmail.com>:
Reply all
Reply to author
Forward
0 new messages