Issue 7339: Property defines an association "datatype" which is redundant (uml2-rtf) Source: Missouri University of Science and Technology (Dr. Thomas Weigert, weigert(at)mst.edu) Nature: Clarification Severity: Significant Summary: Property defines an association "datatype". This association is redundant for the following reasons: (i) A DataType is a kind of classifier, so saying that a property can be owned by a DataType adds nothing new. (ii) as feature, one can navigate from the property to the featuringClassifier, and so the navigability to an owning data type is already given. Moreover, an association to a data type would be incorrect if the property would otherwise be owned by a different Classifier. Moreover, if this property is owned by a classifier, there is no guarantee that the datatype association references the same DataType. There are no consistency constraints. Anyway, this association is redundant, can possibly lead to inconsistent models, and should be deleted. The last sentence on p.92 "A property may be owned by and in the namespace of a datatype." is correct even if the association is deleted. However, this sentence adds no new information either and is best deleted also. Resolution: see pages 14 - 17 of ptc/2011-01-19 Revised Text: Actions taken: May 15, 2004: received issue April 25, 2011: closed issue Discussion: Disposition: Deferred to UML 2.4 RTF End of Annotations:===== m: webmaster@omg.org Date: 15 May 2004 17:14:50 -0400 To: Subject: Issue/Bug Report -------------------------------------------------------------------------------- Name: Thomas Weigert Company: Motorola mailFrom: thomas.weigert@motorola.com Notification: No Specification: UML Section: 7.11.4 FormalNumber: ptc/03-08-02 Version: 2.0 RevisionDate: 02/08/2003 Page: p.90 Nature: Clarification Severity: Significant HTTP User Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0) Description Property defines an association "datatype". This association is redundant for the following reasons: (i) A DataType is a kind of classifier, so saying that a property can be owned by a DataType adds nothing new. (ii) as feature, one can navigate from the property to the featuringClassifier, and so the navigability to an owning data type is already given. Moreover, an association to a data type would be incorrect if the property would otherwise be owned by a different Classifier. Moreover, if this property is owned by a classifier, there is no guarantee that the datatype association references the same DataType. There are no consistency constraints. Anyway, this association is redundant, can possibly lead to inconsistent models, and should be deleted. The last sentence on p.92 "A property may be owned by and in the namespace of a datatype." is correct even if the association is deleted. However, this sentence adds no new information either and is best deleted also. From: Steve Cook To: "uml2-rtf@omg.org" Subject: Resolution of 7339 in ballot 5 Thread-Topic: Resolution of 7339 in ballot 5 Thread-Index: Acq/s1XezJhALgs6Q1GyxQM7jCo87Q== Date: Tue, 9 Mar 2010 18:07:30 +0000 Accept-Language: en-GB, en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: Issue 7339 is .Property defines an association .datatype. which is redundant.. The resolution says .Revise figure 7.9 to remove association end Property::classifier.. Firstly I don.t see how it is possible to remove an association end. Secondly it is a MOF rule that every feature must have a name, so if this supposed to mean .remove the name. that would be illegal. Thirdly, the .classifier. end is owned by the association, so it is unsurprising that Property does not own it, and formally incorrect to refer to it as Property::classifier. In fact I think this whole issue is a special case of 14977 Matching subsettting across association ends. Because Property::class subsets featuringClassifier, it follows that Class::ownedAttribute subsets feature. I think there is no problem with making Classifier::/attribute subset feature, in which case the other end (classifier) would subset featuringClassifier automatically. But the second proposed amendment to figure 7.9 is wrong. -- Steve