Ontology Definition Metamodel Avatar
  1. OMG Specification

Ontology Definition Metamodel — Open Issues

  • Acronym: ODM
  • Issues Count: 58
  • Description: Issues not resolved
Open Closed All
Issues not resolved

Issues Summary

Key Issue Reported Fixed Disposition Status
ODM12-131 ODM does not specify how to use triples ODM 1.1 open
ODM12-130 Serializing a self-contained model is problematic ODM 1.1 open
ODM12-129 Section 10.2.4's description of an RDFSResource is misleading ODM 1.1 open
ODM12-128 OWL metamodel needs an association between CardinalityRestriction and DataRange ODM 1.1 open
ODM12-127 OWLDatatype should specialize RDFSDatatype ODM 1.1 open
ODM12-126 Erroneous association in definition of Document? ODM 1.1 open
ODM12-125 Section 11.6.1 uses "ClassExpressions" instead of "DataRanges" ODM 1.1 open
ODM12-124 Conflicting role names in OWL metamodel ODM 1.1 open
ODM12-123 Issues in Annex G ODM 1.1 open
ODM12-122 Annex D needs complete models ODM 1.1 open
ODM12-32 There are still mentions of OWL Full and OWL DL ODM 1.1 open
ODM12-118 Figure 10.8 has role name ambiguity ODM 1.1 open
ODM12-24 Some values of the rdf:esource attribute are incorrect ODM 1.1 open
ODM12-114 ODM specification uses OWLDataRange instead of DataRange ODM 1.1 open
ODM12-113 Section D.4.5.2 has incorrect table reference ODM 1.1 open
ODM12-112 Typos in Section 10.8.5 ODM 1.1 open
ODM12-111 Invalid association in RDF metamodel, Section 10.8.1? ODM 1.1 open
ODM12-110 Section 10.4.3 has a typo in a property name ODM 1.1 open
ODM12-107 The Course ontology in Annex D is imprecisely defined ODM 1.1 open
ODM12-22 Annex mixes UML and OWL terminology ODM 1.1 open
ODM12-99 ODM does not use most recent version of UML ODM 1.1 open
ODM12-4 Representation of Axioms ODM 1.1 open
ODM12-3 Representation of Axioms ODM 1.1 open
ODM12-39 OWLOntology::owlUniverse is not documented – in fact it’s in 11.7.2 which should be moved ODM 1.1 open
ODM12-38 In 10.6.3, RDFSisDefinedBy should be a subproperty of seeAlso ODM 1.1 open
ODM12-41 In 10.6.2 the property “resource” does not seem right ODM 1.1 open
ODM12-40 In 10.6.3 the constraints should not refer to IRIs but linked objects ODM 1.1 open
ODM12-42 On Triple (mis-named RDF Triple in the spec), statement, document are not documented ODM 1.1 open
ODM12-43 On RDFSResource, groupingNamespace, RDFSLabel, RDFSComment, are not documented ODM 1.1 open
ODM12-29 OWLAllDifferent should specialize ClassExpression not OWLClass ODM 1.1 open
ODM12-35 Question section 11.2.3 RDFSLiteral (Augmented) ODM 1.1 open
ODM12-33 10.9.8 does not document bNode ODM 1.1 open
ODM12-36 In 10.9.7, graphForNG should be just graph ODM 1.1 open
ODM12-34 In 10.9.4 the attribute namespacePrefix should be just prefix ODM 1.1 open
ODM12-37 In 10.9.1, Document::xmlBase should be deleted (it’s on Source now) ODM 1.1 open
ODM12-16 Multiplicity Conflict ODM 1.1 open
ODM12-19 Class Node should be abstract ODM 1.1 open
ODM12-1 ODM metamodel needs a Package concept for managing a structure for ontologies ODM 1.1 open
ODM12-49 Remove sub-packages of OWL metamodel ODM 1.1 open
ODM12-45 The section numbers refer to the 1.1 convenience document ODM 1.1 open
ODM12-21 Annex uses old stereotype name ODM 1.1 open
ODM12-30 OWLOntologyProperty is obsolete ODM 1.1 open
ODM12-31 Constraint [1] of AnnotationProperty still has URIReference and RDFLLiteral ODM 1.1 open
ODM12-11 Typos, grammatical errors, and style issues ODM 1.1 open
ODM12-14 Incorrect relationship between OWL and RDF packages? ODM 1.1 open
ODM12-15 Graphical representation of «equivalentClass» conflicts with text ODM 1.1 open
ODM12-17 UML class name typo ODM 1.1 open
ODM12-18 Conflicting property names for stereotype "statement" ODM 1.1 open
ODM12-20 Update annotation stereotype in OWL profile ODM 1.1 open
ODM12-6 The ODM Metamodel and Profile should support OWL2 property chains ODM 1.1 open
ODM12-7 The ODM Metamodel and Profile should support OWL2 inverse object property expressions ODM 1.1 open
ODM12-5 The ODM Metamodel and Profile model libraries should support OWL2 extra built-in datatypes (rational, real) ODM 1.1 open
ODM12-10 The ODM Metamodel and Profile model libraries should support OWL2 universal and empty object and data properties ODM 1.1 open
ODM12-8 The ODM Metamodel and Profile should support OWL2 pairwise disjoint and disjoint union classes ODM 1.1 open
ODM12-9 WL2 n-ary datatype hook and universal and existential restriction on n-ary data ranges ODM 1.1 open
ODM12-12 Stereotypes associated with Restrictions in the OWL Profile are incompletely defined ODM 1.1 open
ODM12-13 Self Object Property Restrictions are only partially supported in the ODM 1.1 Specification ODM 1.1 open
ODM12-2 Issues on ODM 1.1 UML XMI file for RDFLibrary ODM 1.1 open

Issues Descriptions

ODM does not specify how to use triples

  • Key: ODM12-131
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Triples are necessary to state certain semantics in the metamodel (as opposed to the profile). For example, property assertions must be expressed using triples. The metamodel sections do not state how and when triples are to be used.

    This issue requests that documentation be added to metamodel sections describing how to use triples.

  • Reported: ODM 1.1 — Wed, 3 May 2017 16:36 GMT
  • Updated: Wed, 26 Jul 2017 16:45 GMT

Serializing a self-contained model is problematic

  • Key: ODM12-130
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Because of the composition relationships in the current RDF and OWL metamodels, it is not possible to create a model that:

    1. contains an instance of class Namespace, and
    2. has an XMI serialization with a single root element that owns, directly or indirectly, all the other elements in the model.

    Class NamedGraph is the only class that can own Namespace (that is, the association between NamedGraph and Namespace is the only composite association involving Namespace). Therefore, a model with an instance of Namespace must either be serialized with a root element of Namespace or NamedGraph. In either case, the serialization cannot contain a Source or a NamespaceDefinition. No associations in the metamodel allow a Namespace or a NamedGraph to own a Source or a NamespaceDefinition, either directly or indirectly.

    Conversely, a model with a Source has no serialization in which the Source can own, directly or indirectly, a Namespace.

    The serialization of an ontology represented using the metamodel must contain (at least) two instances of metamodel classes. This is problematic for ECore, where the persistent representation of an ontology must exist in multiple files.

  • Reported: ODM 1.1 — Wed, 12 Apr 2017 17:36 GMT
  • Updated: Wed, 12 Apr 2017 17:36 GMT

Section 10.2.4's description of an RDFSResource is misleading

  • Key: ODM12-129
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Section 10.4.2 states:

    All things described by RDF are called resources. This is the class of everything. All other classes are subclasses of this
    class.

    This is true of RDF, but it is not true of the RDF metamodel. Class RDFSResource is not the superclass of all classes in the RDF metamodel. The paragraph should be rewritten to clarify that it applies to RDF but not the RDF metamodel.

  • Reported: ODM 1.1 — Wed, 12 Apr 2017 17:07 GMT
  • Updated: Wed, 12 Apr 2017 17:07 GMT

OWL metamodel needs an association between CardinalityRestriction and DataRange

  • Key: ODM12-128
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    In the OWL metamodel, class CardinalityRestriction has an association with ClassExpression but not with DataRange. Consequently, the metamodel can represent a cardinality restriction such as:

    objProp exactly 1 cls

    and

    dataProp exactly 1

    but not

    dataProp exactly 1 xsd:string

    The metamodel does associate SomeValuesFromRestriction and AllValuesFromRestriction with both ClassExpression and DataRange. It needs the corresponding association for CardinalityRestriction.

  • Reported: ODM 1.1 — Mon, 10 Apr 2017 13:55 GMT
  • Updated: Mon, 10 Apr 2017 13:55 GMT

OWLDatatype should specialize RDFSDatatype

  • Key: ODM12-127
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    OWLDatatype specializes DataRange, which specializes RDFSClass. But the OWL 2 Mapping to RDF Graphs states in Table 1 that an OWL datatype maps to the triple:

    	T(DT) rdf:type rdfs:Datatype .
    

    OWLDatatype should specialize RDFSDatatype rather than RDFSClass.

    Classes DataIntersectionOf, DataUnionOf, DataComplementOf, DataOneOf, and DatatypeRestriction also map to triples of this form. Together with OWLDatatype, that's all the subclasses of DataRange. Should DataRange specialize RDFSDatatype instead of RDFSClass?

  • Reported: ODM 1.1 — Fri, 17 Mar 2017 19:44 GMT
  • Updated: Fri, 17 Mar 2017 19:44 GMT

Erroneous association in definition of Document?

  • Key: ODM12-126
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    The second association for class Document in the RDF metamodel (Section 10.8.1) is to Namespace, with role name xmlBase. The association's name is DefaultNamespaceForDocument. There is no corresponding association for Namespace (10.8.2). Association DefaultNamespaceForDocument exists between classes Source and Namespace. That role name is defaultNamespace, not xmlBase. Source has an association with role name xmlBase, but the associated class is IRI.

    Given that an association to Namespace exists in a superclass of Document, the xmlBase association is either redundant or a specialization that needs to be further described.

    Also, should the association between Source and Namespace be named DefaultNamespaceForSource instead of DefaultNamespaceForDocument?

  • Reported: ODM 1.1 — Tue, 21 Feb 2017 14:50 GMT
  • Updated: Tue, 21 Feb 2017 14:50 GMT

Section 11.6.1 uses "ClassExpressions" instead of "DataRanges"

  • Key: ODM12-125
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Section 11.6.1 describes how the OWL DataRange concept is modeled. Its Constraint [1] has the phrase "ClassExpressions must be anonymous". The phrase should read "DataRanges must be anonymous".

  • Reported: ODM 1.1 — Tue, 7 Feb 2017 14:41 GMT
  • Updated: Tue, 7 Feb 2017 14:41 GMT

Conflicting role names in OWL metamodel

  • Key: ODM12-124
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    In Figure 11.4, both classes DataRange and ClassExpression have two associations whose role names are "restrictionClass". UML does not permit a class to have multiple associations with the same role name.

    This issue is being reported based on Figure 11.4, but it affects sections 11.3.1 and 11.6.1 as well.

  • Reported: ODM 1.1 — Fri, 13 Jan 2017 18:36 GMT
  • Updated: Wed, 18 Jan 2017 22:37 GMT

Issues in Annex G

  • Key: ODM12-123
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Annex G (MOF QVT: A Brief Tutorial) contains some statements that are incorrect or incomplete:

    • Section G.2, paragraph 3, sentence 2 states: "No work is done via a when clause". This is misleading. Relations invoked in a when clause can create instances.
    • Section G.2, paragraph 3, sentence 4: The sentence states that predicates can be connected by an or operator. Individual predicates can contain or operators, but the QVT specification doesn't say or connects predicates.
    • Section G.2, paragraph 4: The paragraph reads "OCL functions can be used." This is true, but incomplete. QVT permits a transformation to define functions, and to access externally-defined functions written in a language such as Java. The paragraph should be amended to describe these types of functions.
    • Section G.2, paragraph 5, sentence 1:
      • The sentence references OCL statements. The OCL specification describes a statement as something along the lines of an invariant (including the keyword inv) or a default attribute value. The correct term is "OCL expressions".
      • The sentence neglects to state that an OCL let expression can define a variable.
    • Section G.2, paragraph 7, sentence 2 reads "A top relation is executed by the system." A top relation may also be called in a where clause.
    • Section G.3, paragraph 6, sentence 2 reads "A process in a repository can link from one object to another via any property ..." The paragraph does not explain what "process" means.
    • Section G.3, paragraph 2, last sentence: the second clause should begin with "it".
  • Reported: ODM 1.1 — Fri, 13 Jan 2017 18:23 GMT
  • Updated: Fri, 13 Jan 2017 18:23 GMT

Annex D needs complete models

  • Key: ODM12-122
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Annex D illustrates transforming between UML and OWL using an example that shows courses, students, their relationships, and their properties. The examples are presented in fragments throughout section D.2. Readers would benefit from access to complete UML and RDF models. Furthermore, these models being definitive, rigorous, and complete, they would be the basis for the figures and tables in Section D.2 (and perhaps elsewhere in Annex D).

    This issue includes two attachments:

    • Course-Student.mdzip is a UML class model. It can be opened using MagicDraw.
    • AnnexDExamples.rdf is an OWL 2 model expressed in RDF/XML format.

    These models are informative, not normative.

  • Reported: ODM 1.1 — Fri, 6 Jan 2017 22:34 GMT
  • Updated: Fri, 6 Jan 2017 22:34 GMT
  • Attachments:

There are still mentions of OWL Full and OWL DL

  • Key: ODM12-32
  • Legacy Issue Number: 19111
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    There are still mentions of OWL Full and OWL DL

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 21 Dec 2016 15:47 GMT

Figure 10.8 has role name ambiguity

  • Key: ODM12-118
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    In Figure 10.8, class Graph has associations to classes Dataset and Source. Both these associations have role name "root". That is not valid, because it creates ambiguity in OCL expressions.

  • Reported: ODM 1.1 — Mon, 24 Oct 2016 21:07 GMT
  • Updated: Mon, 24 Oct 2016 21:07 GMT

Some values of the rdf:esource attribute are incorrect

  • Key: ODM12-24
  • Legacy Issue Number: 19125
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Annex D has several OWL examples that use the rdf:resource attribute in which the attribute's value is, I think, meant to be prefixed with a "#". For example, in Table D-7,

    <rdfs:domain rdf:resource="Course"/>

    should be:

    <rdfs:domain rdf:resource="#Course"/>

  • Reported: ODM 1.1 — Tue, 26 Nov 2013 05:00 GMT
  • Updated: Fri, 2 Sep 2016 00:28 GMT

ODM specification uses OWLDataRange instead of DataRange

  • Key: ODM12-114
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    In 14 places, the ODM specification uses OWLDataRange: 12 times in the metamodel and twice in Annex D. The metamodel defines class DataRange, not OWLDataRange.

  • Reported: ODM 1.1 — Mon, 8 Aug 2016 14:49 GMT
  • Updated: Mon, 8 Aug 2016 14:49 GMT

Section D.4.5.2 has incorrect table reference

  • Key: ODM12-113
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    The first paragraph in section D.4.5.2 says Table A.4 in Annex A has the UML elements for OWL. These elements are now in Table A.6. The text needs to be brought up to date.

  • Reported: ODM 1.1 — Fri, 5 Aug 2016 21:45 GMT
  • Updated: Fri, 5 Aug 2016 21:45 GMT

Typos in Section 10.8.5

  • Key: ODM12-112
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    In the Associations for NamedGraph (Section 10.8.5, p. 55):

    • The first role is named graphForNG. In Figure 10.8, it's "graph".
    • The second role is graphname. In Figure 10.8, it's "graphName".
  • Reported: ODM 1.1 — Tue, 21 Jun 2016 21:26 GMT
  • Updated: Tue, 21 Jun 2016 21:26 GMT

Invalid association in RDF metamodel, Section 10.8.1?

  • Key: ODM12-111
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    The second association for Document (Section 10.8.1, p. 53) has role xmlBase. Is this a legacy of a previous version? The role name isn't in Figure 10.8, and the description references association DefaultNamespaceForDocument, which is between Source and Namespace.

  • Reported: ODM 1.1 — Tue, 21 Jun 2016 21:14 GMT
  • Updated: Tue, 21 Jun 2016 21:14 GMT

Section 10.4.3 has a typo in a property name

  • Key: ODM12-110
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    In section 10.4.3, in the Associations section, the 2nd bullet has "RDFSProperty" as the type of the RDFpredicate role. It should be RDFProperty.

  • Reported: ODM 1.1 — Tue, 21 Jun 2016 19:56 GMT
  • Updated: Tue, 21 Jun 2016 19:56 GMT

The Course ontology in Annex D is imprecisely defined

  • Key: ODM12-107
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Annex D uses a simple example, comprising courses, students, instructors, and their relationships. It presents this example in both UML and OWL. The OWL is fragmentary. A non-normative, machine-readable OWL file would resolve questions about the ontology's exact content.

  • Reported: ODM 1.1 — Tue, 19 Apr 2016 21:13 GMT
  • Updated: Tue, 19 Apr 2016 22:04 GMT

Annex mixes UML and OWL terminology

  • Key: ODM12-22
  • Legacy Issue Number: 19126
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    The second sentence of Annex D.2.2 reads as follows:

    A class in OWL is a set of zero or more instances.

    An OWL class consists of individuals, not instances. The sentence should be changed to use OWL terminology.

  • Reported: ODM 1.1 — Tue, 26 Nov 2013 05:00 GMT
  • Updated: Fri, 27 Nov 2015 06:48 GMT

ODM does not use most recent version of UML

  • Key: ODM12-99
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    ODM's metamodels are based on UML 2.4.1, which was formalized in August 2011. UML 2.5 was formalized in June 2015. ODM should switch to UML 2.5 as soon is as practical.

  • Reported: ODM 1.1 — Fri, 13 Nov 2015 15:58 GMT
  • Updated: Fri, 13 Nov 2015 16:45 GMT

Representation of Axioms

  • Key: ODM12-4
  • Legacy Issue Number: 19416
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    Representation of Axioms

    OWL2 uses Axioms to represent several capabilities, whereas ODM has them either as binary associations or not at all.

    In OWL these axioms are represented as a blank node so should have a metamodel element: that is necessary to allow the statement itself to be associated with an ontology/document.

    ODM has OWLAllDifferent which inherits from OWLClass which does not make sense.

    In the metamodel there should be a top level class representing Axiom with the following subclasses:

    • DifferentIndividuals (replaces association DifferentIndividual and class OWLAllDifferent)
    • SameIndividual (replaces association SameIndividual)
    • EquivalentClass (replaces association EquivalentClass)
    • DisjointClasses (replaces association DisjointClass)
    • EquivalentProperty(replaces association EquivalentProperty)
    • DisjointProperties (new)
  • Reported: ODM 1.1 — Wed, 14 May 2014 04:00 GMT
  • Updated: Wed, 5 Aug 2015 17:08 GMT

Representation of Axioms

  • Key: ODM12-3
  • Legacy Issue Number: 19421
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    Representation of Axioms

    OWL2 uses Axioms to represent several capabilities, whereas ODM has them either as binary associations or not at all.

    In OWL these axioms are represented as a blank node so should have a metamodel element: that is necessary to allow the statement itself to be associated with an ontology/document.

    ODM has OWLAllDifferent which inherits from OWLClass which does not make sense.

    In the metamodel there should be a top level class representing Axiom with the following subclasses:

    • DifferentIndividuals (replaces association DifferentIndividual and class OWLAllDifferent)
    • SameIndividual (replaces association SameIndividual)
    • EquivalentClass (replaces association EquivalentClass)
    • DisjointClasses (replaces association DisjointClass)
    • EquivalentProperty(replaces association EquivalentProperty)
    • DisjointProperties (new)
  • Reported: ODM 1.1 — Wed, 14 May 2014 04:00 GMT
  • Updated: Wed, 5 Aug 2015 17:07 GMT

OWLOntology::owlUniverse is not documented – in fact it’s in 11.7.2 which should be moved

  • Key: ODM12-39
  • Legacy Issue Number: 19109
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    OWLOntology::owlUniverse is not documented – in fact it’s in 11.7.2 which should be moved

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 17:04 GMT

In 10.6.3, RDFSisDefinedBy should be a subproperty of seeAlso

  • Key: ODM12-38
  • Legacy Issue Number: 19103
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    In 10.6.3, RDFSisDefinedBy should be a subproperty of seeAlso

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 17:03 GMT

In 10.6.2 the property “resource” does not seem right

  • Key: ODM12-41
  • Legacy Issue Number: 19102
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    In 10.6.2 the property “resource” does not seem right

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 17:03 GMT

In 10.6.3 the constraints should not refer to IRIs but linked objects

  • Key: ODM12-40
  • Legacy Issue Number: 19104
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    In 10.6.3 the constraints should not refer to IRIs but linked objects

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 17:03 GMT

On Triple (mis-named RDF Triple in the spec), statement, document are not documented

  • Key: ODM12-42
  • Legacy Issue Number: 19101
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    On Triple (mis-named RDF Triple in the spec), statement, document are not documented. Actually statement is in 10.4.4 which should be moved.

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 17:03 GMT

On RDFSResource, groupingNamespace, RDFSLabel, RDFSComment, are not documented

  • Key: ODM12-43
  • Legacy Issue Number: 19100
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    On RDFSResource, groupingNamespace, RDFSLabel, RDFSComment, are not documented. Actually the latter are in 10.3.2 which should be moved

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 17:02 GMT

OWLAllDifferent should specialize ClassExpression not OWLClass

  • Key: ODM12-29
  • Legacy Issue Number: 19114
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    OWLAllDifferent should specialize ClassExpression not OWLClass

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 17:01 GMT

Question section 11.2.3 RDFSLiteral (Augmented)

  • Key: ODM12-35
  • Legacy Issue Number: 19110
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    Question section 11.2.3 RDFSLiteral (Augmented)

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 17:00 GMT

10.9.8 does not document bNode

  • Key: ODM12-33
  • Legacy Issue Number: 19108
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    10.9.8 does not document bNode

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 16:59 GMT

In 10.9.7, graphForNG should be just graph

  • Key: ODM12-36
  • Legacy Issue Number: 19107
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    In 10.9.7, graphForNG should be just graph

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 16:59 GMT

In 10.9.4 the attribute namespacePrefix should be just prefix

  • Key: ODM12-34
  • Legacy Issue Number: 19106
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    In 10.9.4 the attribute namespacePrefix should be just prefix

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 16:59 GMT

In 10.9.1, Document::xmlBase should be deleted (it’s on Source now)

  • Key: ODM12-37
  • Legacy Issue Number: 19105
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    In 10.9.1, Document::xmlBase should be deleted (it’s on Source now)

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 16:58 GMT

Multiplicity Conflict

  • Key: ODM12-16
  • Legacy Issue Number: 19170
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    The association between Document and Triple in Figure 10.8 shows multiplicity 0..* for the "triple" role. In Section 10.8.1, the association's multiplicity is 1..*.

  • Reported: ODM 1.1 — Mon, 30 Dec 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 16:57 GMT

Class Node should be abstract

  • Key: ODM12-19
  • Legacy Issue Number: 19168
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    The Description in Section 10.2.2 states that ReferenceNode, BlankNode, and Literal form a complete and disjoint covering of Node. That means Node is an abstract class. For consistency with the «node» stereotype (Section 14.1.3.5), Section 10.2.2 should describe it as such.

  • Reported: ODM 1.1 — Sat, 28 Dec 2013 05:00 GMT
  • Updated: Wed, 5 Aug 2015 16:56 GMT

ODM metamodel needs a Package concept for managing a structure for ontologies

  • Key: ODM12-1
  • Legacy Issue Number: 19630
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    ODM metamodel needs a Package concept for managing a structure for ontologies, akin to the use of UML Packages in the Profile.

    In fact, the use of UML Packages in ontology models is outside the documented profile. There is nothing to say, for example, that the package structure should mirror the URI structure and/or whether it matters whether they differ.

    Despite the fact that the RDF/OWL languages do not have Package, they do make use of folders/directories in wither filestore or on the web.

    And ontologies like FIBO now have metadata at the package/folder level but neither the metamodel nor the profile provide any place to hang these.

    Finally packages provide a much-needed scoping mechanism for applying various operations – which become increasingly impractical as the number of leaf-level ontologies has already exceeded 400 for FIBO

  • Reported: ODM 1.1 — Mon, 6 Oct 2014 04:00 GMT
  • Updated: Wed, 5 Aug 2015 16:43 GMT

Remove sub-packages of OWL metamodel

  • Key: ODM12-49
  • Legacy Issue Number: 19079
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    Remove sub-packages of OWL metamodel

    With OWL 2 no longer making a distinction between OWL DL and OWL Full it does not make sense to have separate ODM packages for these: there should just be one OWL package/metamodel.

  • Reported: ODM 1.1 — Mon, 11 Nov 2013 05:00 GMT
  • Updated: Wed, 29 Jul 2015 16:04 GMT

The section numbers refer to the 1.1 convenience document

  • Key: ODM12-45
  • Legacy Issue Number: 19099
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    The section numbers refer to the 1.1 convenience document

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Annex uses old stereotype name

  • Key: ODM12-21
  • Legacy Issue Number: 19138
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    n table A-4, the cell in the second column of row 1 contains the stereotype «rdfsContainerMembershipProperty». This name has been changed to «containerMembershipProperty» in version 1.1.

  • Reported: ODM 1.1 — Thu, 12 Dec 2013 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

OWLOntologyProperty is obsolete

  • Key: ODM12-30
  • Legacy Issue Number: 19113
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    OWLOntologyProperty is obsolete

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Constraint [1] of AnnotationProperty still has URIReference and RDFLLiteral

  • Key: ODM12-31
  • Legacy Issue Number: 19112
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    Constraint [1] of AnnotationProperty still has URIReference and RDFLLiteral

  • Reported: ODM 1.1 — Tue, 19 Nov 2013 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Typos, grammatical errors, and style issues

  • Key: ODM12-11
  • Legacy Issue Number: 19279
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    This is a bulk report of minor problems I've identified in the specification. The following is a CSV-format list that can be read by a spreadsheet application.

    Page,Location,Problem,Fix

    38,"Section 10.2.6, Description, paragraph 1","""RFC3986"" should be ""RFC 3986"", consistent with ""RFC 3987"" in the same paragraph.",

    38,"Section 10.2.6, Description, paragraph 2","""RFC3986"" should be ""RFC 3986"".",

    38,"Section 10.2.5, Constraints, paragraph 1, sentence 2","""However, in any case"" is redundant.","Replace ""However, in any case,"" with ""However,""."

    44,"Section 10.5.2, Semantics, paragraph 2",The sentence beginning on the third line is run-in and the second part is incomplete.,"Change the sentence to ""Such features may in the future be provided to support, for example, more elaborate datatype conditions."""

    50,"Section 10.8, paragraph 1",Replace hyphen with dash in line 3.,

    50,"Section 10.8, paragraph 1",Missing comma.,"Change ""prefixes and URIs"" to ""prefixes and IRIs,""."

    51,Last paragraph,Missing period at end of sentence.,

    53,"Section 10.8.2, Description, paragraph 1","The article modifying ""IRI"" is incorrect.","Change ""a IRI"" to ""an IRI""."

    54,"Section 10.8.2, Constraints",Remove the quotation marks around [XMLNS].,

    56,"Section 10.8.8, Associations, bullet 1","Change ""NameSpace"" to ""Namespace"".
    Missing period at end of sentence.",

    73,"Section 11.3.8.1, Description, First paragraph",Missing period at end of last sentence.,

    77,"Section 11.3.8.9, Constraints, paragraph 1",Missing period at end of sentence.,

    82,"Section 11.5, title",Replace hyphen with dash.,

    83,"Section 11.6, paragraph 2",The first sentence is run-in.,"Change ""data values,"" to ""data values:""."

    173,"Section 14.2.5, paragraph 1","On the last line, replace ""classes,"" with ""classes"".",

    174,"Section 14.2.5.1, Description, paragraph 2, sentence 1",The use of passive voice reduces specificity.,"Change
    owl:Class is defined as a subclass of rdfs:Class.
    to
    [OWL SS&FS] defines owl:Class as a subclass of rdfs:Class."

    199,"Section 14.2.6.3, Description, paragraph 1",The first sentence is missing a comma.,"Change ""in the profile"" to ""in the profile,""."

    207,"Section 14.2.7.5, Description, paragraph 1","The use of commas in the first sentence decreases readability: it's not immediately obvious how to read ""OWL properties"".","Change ""OWL properties"" to ""OWL, properties""."

    207,"Section 14.2.7.5, Stereotype and Base Class, paragraph 1",The first sentence is run-in.,"Change ""No stereotype,"" to ""No stereotype;""."

  • Reported: ODM 1.1 — Thu, 6 Mar 2014 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Incorrect relationship between OWL and RDF packages?

  • Key: ODM12-14
  • Legacy Issue Number: 19183
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Figure 14.19 shows an association labeled "Imports" between packages OWL and RDF. Should it be a dependency stereotyped «import»? That's what the UML specifications describe.

  • Reported: ODM 1.1 — Tue, 14 Jan 2014 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Graphical representation of «equivalentClass» conflicts with text

  • Key: ODM12-15
  • Legacy Issue Number: 19182
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    In the GraphicalRepresentation section of owl:equivalentClass, the text says the representation is a dashed line, but Figure 14.41 shows a solid line.

  • Reported: ODM 1.1 — Tue, 14 Jan 2014 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

UML class name typo

  • Key: ODM12-17
  • Legacy Issue Number: 19181
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    The "Stereotype and Base Class" section of 14.2.4.3 states that the base class of «owlImports» is UML::PackageImports. The class name is PackageImport (i.e., it doesn't end with "s").

  • Reported: ODM 1.1 — Tue, 14 Jan 2014 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Conflicting property names for stereotype "statement"

  • Key: ODM12-18
  • Legacy Issue Number: 19140
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Fig. 14.3 shows the first 3 properties of «statement» named RDFsubject, RDFpredicate, and RDFobject. Section 14.1.3.9 lists properties named subject, predicate, and object. The names need to be consistent.

  • Reported: ODM 1.1 — Wed, 11 Dec 2013 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Update annotation stereotype in OWL profile

  • Key: ODM12-20
  • Legacy Issue Number: 19139
  • Status: open  
  • Source: Institute for Defense Analyses ( Steven Wartik)
  • Summary:

    Line 87 of the OWL profile (http://www.omg.org/ODM/20130801/OWLProfile.xml) has an element named annotationProperty. This is the v1.0 name. In 1.1 the name is owlAnnotation (cf. Section 14.2.3.1).

  • Reported: ODM 1.1 — Thu, 12 Dec 2013 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

The ODM Metamodel and Profile should support OWL2 property chains

  • Key: ODM12-6
  • Legacy Issue Number: 19317
  • Status: open  
  • Source: Thematix Partners LLC ( Elisa Kendall)
  • Summary:

    Property chains were added as a feature of OWL 2 and need to be supported in the ODM metamodel and profile.

  • Reported: ODM 1.1 — Fri, 28 Mar 2014 04:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

The ODM Metamodel and Profile should support OWL2 inverse object property expressions

  • Key: ODM12-7
  • Legacy Issue Number: 19316
  • Status: open  
  • Source: Thematix Partners LLC ( Elisa Kendall)
  • Summary:

    Inverse property expressions were new in OWL 2 and support for them needs to be added to the ODM specification.

  • Reported: ODM 1.1 — Fri, 28 Mar 2014 04:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

The ODM Metamodel and Profile model libraries should support OWL2 extra built-in datatypes (rational, real)

  • Key: ODM12-5
  • Legacy Issue Number: 19315
  • Status: open  
  • Source: Thematix Partners LLC ( Elisa Kendall)
  • Summary:

    These datatypes were added in OWL 2, and should be supported in the metamodel, profile and related libraries.

  • Reported: ODM 1.1 — Fri, 28 Mar 2014 04:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

The ODM Metamodel and Profile model libraries should support OWL2 universal and empty object and data properties

  • Key: ODM12-10
  • Legacy Issue Number: 19314
  • Status: open  
  • Source: Thematix Partners LLC ( Elisa Kendall)
  • Summary:

    The OWL 2 top and bottom object and data properties should be represented in the metamodel and profile (possibly as additions to the libraries in Annex A).

  • Reported: ODM 1.1 — Fri, 28 Mar 2014 04:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

The ODM Metamodel and Profile should support OWL2 pairwise disjoint and disjoint union classes

  • Key: ODM12-8
  • Legacy Issue Number: 19313
  • Status: open  
  • Source: Thematix Partners LLC ( Elisa Kendall)
  • Summary:

    Pairwise disjoint classes and disjoint unions are only partially supported in the ODM 1.1. Representation should be completed for both the OWL metamodel and profile.

  • Reported: ODM 1.1 — Fri, 28 Mar 2014 04:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

WL2 n-ary datatype hook and universal and existential restriction on n-ary data ranges

  • Key: ODM12-9
  • Legacy Issue Number: 19312
  • Status: open  
  • Source: Thematix Partners LLC ( Elisa Kendall)
  • Summary:

    The ODM Metamodel and Profile should support OWL2 n-ary datatype hook and universal and existential restriction on n-ary data ranges

  • Reported: ODM 1.1 — Fri, 28 Mar 2014 04:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Stereotypes associated with Restrictions in the OWL Profile are incompletely defined

  • Key: ODM12-12
  • Legacy Issue Number: 19311
  • Status: open  
  • Source: Thematix Partners LLC ( Elisa Kendall)
  • Summary:

    A number of stereotypes are given in section 14.2.5.3 for dependencies to be used with restrictions are incompletely defined in ODM 1.1.

  • Reported: ODM 1.1 — Fri, 28 Mar 2014 04:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Self Object Property Restrictions are only partially supported in the ODM 1.1 Specification

  • Key: ODM12-13
  • Legacy Issue Number: 19310
  • Status: open  
  • Source: Thematix Partners LLC ( Elisa Kendall)
  • Summary:

    In ODM 1.1, partial support for hasSelf property restrictions were added to the metamodel and profile for OWL. This support needs to be fully integrated, and at a minimum, an additional stereotype should be provided in the profile.

  • Reported: ODM 1.1 — Fri, 28 Mar 2014 04:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT

Issues on ODM 1.1 UML XMI file for RDFLibrary

  • Key: ODM12-2
  • Legacy Issue Number: 19656
  • Status: open  
  • Source: Adaptive ( Pete Rivett)
  • Summary:

    I was working on the FIBO UML files and came across some XMI issues with the ODM XMI[PJR] for RDFLibrary: It has an erroneous import of OWLProfile and it uses .xml instead of .xmi for the RDFProfile.

    Proposed resolution:

    The lines at 402 of RDFLibrary.xmi are currently:

    <uml:Package xmi:id="_16_6_1_15100de_1266442802908_497608_344" name="rdf" URI="http://www.omg.org/spec/ODM/20131101/RDFLibrary.xmi">

    <profileApplication xmi:type="uml:ProfileApplication" xmlns:RDFProfile="http://www.magicdraw.com/schemas/RDFProfile.xmi" xmi:id="_RDFProfile-Application" xmlns:OWLProfile="http://www.magicdraw.com/schemas/OWLProfile.xmi">

    <appliedProfile href="http://www.omg.org/spec/ODM/20131101/RDFProfile.xml#_12_5_1_137b03ac_1193948135234_354269_2454"/>

    And should be as follows:

    <uml:Package xmi:id="_16_6_1_15100de_1266442802908_497608_344" name="rdf" URI="http://www.omg.org/spec/ODM/20131101/RDFLibrary.xmi">

    <profileApplication xmi:type="uml:ProfileApplication" xmi:id="_RDFProfile-Application">

    <appliedProfile href="http://www.omg.org/spec/ODM/20131101/RDFProfile.xmi#_12_5_1_137b03ac_1193948135234_354269_2454"/>

  • Reported: ODM 1.1 — Tue, 11 Nov 2014 05:00 GMT
  • Updated: Tue, 24 Mar 2015 23:05 GMT