Issue 197 in neologism: Support for named individuals

0 views
Skip to first unread message

neol...@googlecode.com

unread,
May 13, 2011, 5:17:07 AM5/13/11
to neologis...@googlegroups.com
Status: New
Owner: ----

New issue 197 by pmurray....@gmail.com: Support for named individuals
http://code.google.com/p/neologism/issues/detail?id=197

When building an ontology, one of the most important things you do is to
create enumeration instances. This means supporting named individuals.

neol...@googlecode.com

unread,
May 13, 2011, 9:06:22 AM5/13/11
to neologis...@googlegroups.com

Comment #1 on issue 197 by rich...@cyganiak.de: Support for named
individuals
http://code.google.com/p/neologism/issues/detail?id=197

Enumeration classes are rare in web vocabularies, therefore they are not
supported in Neologism.

See also Issue 42.

neol...@googlecode.com

unread,
May 14, 2011, 7:54:06 AM5/14/11
to neologis...@googlegroups.com

Comment #2 on issue 197 by pmurray....@gmail.com: Support for named
individuals
http://code.google.com/p/neologism/issues/detail?id=197

This is .... a surprising response.

I am working in the bioinformatics community. Lists of enumerations is one
of the basic things that a vocabulary needs. Here are some ontologies from
the Taxonomic Database Working Group (TDWG)

http://rs.tdwg.org/ontology/voc/TaxonRank
http://rs.tdwg.org/ontology/voc/CollectionType
http://rs.tdwg.org/ontology/voc/CyclicityTerm
http://rs.tdwg.org/ontology/voc/InstitutionType
http://rs.tdwg.org/ontology/voc/OccurrenceStatusTerm

the list goes on. As things stand, we cannot use neologism without support
for named individuals.


neol...@googlecode.com

unread,
May 15, 2011, 11:24:12 AM5/15/11
to neologis...@googlegroups.com

Comment #3 on issue 197 by rich...@cyganiak.de: Support for named
individuals
http://code.google.com/p/neologism/issues/detail?id=197

We surveyed many of the most widely used vocabularies on the Web, and very
few of them declare any individuals.

The major exception would be SKOS-based vocabularies, which are already
well supported by various tools and we have no plans of competing with them
by adding SKOS support. SKOS is great for managing lists of terms and
similar controlled vocabularies, and frankly, I would have advised to do
most of the term collections you link to as SKOS vocabularies.

Other than that, publishing of individuals is simply a different
problem—while vocabularies typically contain less than 100 classes and
properties, individuals are usually published and managed in much larger
collections, which calls for different tools with different features.
Database-to-RDF mappers, spreadsheet-to-RDF mappers, and RDF-enabled CMSes
(such as Drupal, which Neologism is based on) come to mind.

neol...@googlecode.com

unread,
Jul 20, 2012, 11:50:08 AM7/20/12
to neologis...@googlegroups.com

Comment #4 on issue 197 by dom.guar...@gmail.com: Support for named
individuals
http://code.google.com/p/neologism/issues/detail?id=197

I have the same need, and would cite Goodrelations as a huge ontology that
ships with individuals (the payments methods)
We it comes to having a few constants, I think we should have the ability
to just remove the "rdf:type owl:Class" from a class to make it an
individual resource, removing at the same time the whole inline selectors
for subclass, disjoint etc

Reply all
Reply to author
Forward
0 new messages