Issue 11105: Bi-directional URIReferenceForNamespace association (odm-ftf) Source: International Business Machines (Mr. Guo Tong Xie, xieguot(at)cn.ibm.com) Nature: Revision Severity: Critical Summary: In section 10.7.3, URIReferenceForNamespace should be changed to two uni-directional associations between URIReference and Namespace. In the current model, if two URIReference have the same Namespace, it could not be represented. Resolution: This issue really identifies the need for two relationships between these classes. It calls for an additional role that would yield links to all the URIReferences "contained" in a Namespace. The current association between these two classes is meant to represent URIReferences which identify namespaces, hence the association name URIReferenceForNamespace, while the URIReferences in question would identify elements in an owl ontology or rdfs vocabulary and not Namespaces. As the model stands, there is no easy way to derive this information. The change to address this involves the addition of a new bi-directional association, URIReferenceInNamespace, between the URIReference and Namespace classes, as shown below. Revised Text: 1. Replace Figure 10.7, page 55, with 2. Under Associations in section 10.7.6, URIReference (Augmented Definition), add the following additional association: § owningNamespace: Namespace [0..1] in association URIReferenceInNamespace - links a URI reference to the namespace that owns it. 3. Also under Associations in section 10.7.6, URIReference (Augmented Definition), within the association description for namespace:Namespace, replace text: links a URI reference to a namespace. with links a URI reference to an optional namespace it identifies. 4. Under Associations in section 10.7.3, Namespace, add the following additional association: § uriRefInNamespace: URIReference [0..*] in association URIReferenceInNamespace - links a namespace to the URI reference(s) it owns. Disposition: Resolved Actions taken: June 20, 2007: received issue January 15, 2008: closed issue Discussion: End of Annotations:===== s is issue # 11105 Bi-directional URIReferenceForNamespace association To: "Elisa F. Kendall" Cc: Bob Colomb , Juergen Boldt , odm-ftf@omg.org Subject: Re: Minutes from ODM-FTF telecon of 27 June X-Mailer: Lotus Notes Release 7.0 HF277 June 21, 2006 From: Guo Tong Xie Date: Tue, 10 Jul 2007 11:29:11 +0800 X-MIMETrack: Serialize by Router on D23M0037/23/M/IBM(Release 7.0.2HF32 | October 17, 2006) at 07/10/2007 11:29:13, Serialize complete at 07/10/2007 11:29:13 Please find my followup as below. 1. Issue 11105, chapter 10, section 10.7.3, URIReferenceForNamespace should be changed to two uni-directional associations between URIReference and Namespace. In the current model, if two URIReference have the same Namespace, it could not be represented. Actually, we use two uni-directional associations when implementing EODM (http://www.eclipse.org/modeling/mdt/?project=eodm#eodm). 2. Issue 11107, chapter 14, section 14.2.3.1, it is not clear on how to set values for annotation property. In the following example, an AnnotationProperty, creator, is defined. Then an owl:Class, MusicalWork, is defined and its creator is "N.N.". Actually, this is an exmaple from http://www.w3.org/TR/owl-ref/#Annotations. Musical work N.N. The question is that I have no idea on how to set creator of MusicalWork as N.N. by using the current profile. Regards, Guo Tong (Gordon) Xie (л¹úÍ®) Manager, Semantic Integration IBM China Research Lab Email: xieguot@cn.ibm.com Tel: 86-10-58748425 Tieline: 905-8425 Fax: 86-10-58748230 Homepage: http://www.research.ibm.com/people/x/xieguot Address: Building 19 Zhongguancun Software Park, 8 Dongbeiwang WestRoad, Haidian District, Beijing,P.R.C.100094 "Elisa F. Kendall" 2007-06-28 05:15 To odm-ftf@omg.org cc Juergen Boldt , Bob Colomb Subject Minutes from ODM-FTF telecon of 27 June [[Juergen: Would you please put this on the OMG server? Thanks.]] Minutes from ODM-FTF Telecon 27 June 2007 Scribe: Elisa Kendall Participants: Roy Bell Raytheon Elisa Kendall Sandpiper Software Evan Wallace NIST Chris Welty IBM Watson Research Guo Tong Xie IBM China Research Regrets: Saartje Brockmans (FZI/AIFB), Bob Colomb (Universiti Teknologi Malaysia) ***** AGENDA ***** 1. Admin - Approved: minutes of meeting held 20 June 2007 (recorded at http://www.omg.org/cgi-bin/doc?ptc/2007-06-07) 2. Review of outstanding actions Outstanding Action Items prior to telecon: Continues Elisa Kendall to talk with Chris Welty, Pete Rivett, and Bob Colomb to determine their plans wrt the ODM-FTF participation. - Chris Welty & Bob Colomb contacted; Pete Rivett remaining Continues Bobbin Teegarden to talk with Pete Rivett to determine what type of file is needed for the UML profiles in ODM and submit issues requesting these be included for the finalized specification. Continues Saartje to look at issue 10843 (issue regarding whether subclassing of OWL Class takes one out of OWL DL) Continues Chris will post the issues he and other IBM colleagues noted in reviewing the RDF and OWL metamodel chapters in October. Continues Elisa will contact Sreedhar Reddy regarding availability to assist with the QVT vetting for finalization purposes. Complete Elisa will revise the spreadsheet per discussion and upload it to the wiki, add URI to the agenda Complete Elisa will add content for issues outside of chapter 16 (by Roy Bell) Complete Roy will add issues related to chapter 16 Continues Evan will add a page to the wiki for managing issues 3. Discussion of outstanding issues / work items - Discussed WebEx access issues -- Elisa and Evan will attempt to determine the right link to include in future agenda email - Reviewed several issues submitted by Gordon, including 11103 through 11107, as follows 1. Issue 11103, chapter 10, section 10.4.1: the role name of superClass in ClassGeneralization is confusing. Discussion: agreed that superClass seems synonymous with RDFSsubClassOf, and that it should be changed, potentially to superClassOf. Elisa will propose a new figure and search the text to find any relevant references that require revision for resolution, to be discussed at the next telecon. 2. Issue 11104, chapter 10, section 10.5.1: the role name of superProperty in PropertyGeneralization is confusing. Discussion: agreed that superProperty seems synonymous with RDFSsubPropertyOf, and that it should be changed, potentially to superPropertyOf, similarly to 11103, above. Elisa will propose a new figure and search the text to find any relevant references that require revision for resolution, to be discussed at the next telecon. 3. Issue 11105, chapter 10, section 10.7.3, URIReferenceForNamespace should be changed to two uni-directional associations between URIReference and Namespace. In the current model, if two URIReference have the same Namespace, it could not be represented. Discussion: participants thought that we understood the issue, but for clarification and to provide more detail for discussion, Gordon will send examples, including the work-around adopted by IBM to use two uni-directional associations. 4. Issue 11106, chapter 11, section 11.4.5, the cardinality of OWLInverseOf in InverseProperty association should be changed from 0..1 to 0..*. One property can have multiple inverse properties Discussion: we used the example of manages and is managed by, as two inverse properties, with equivalent properties bosses and is bossed by to try to tease out this issue. What Gordon was saying is that is managed by should be capable of having both manages and bosses as inverses, and the current metamodel only allows for a single inverse. After discussion, we agreed that the cardinality should be changed from 0..1 to 0..*. Elisa will propose a new figure, and search the text to find any relevant references that require revision for resolution, to be discussed at the next telecon. 5. Issue 11107, chapter 14, section 14.2.3.1, it is not clear on how to set values for annotation property Discussion: we reviewed the representation for owl:AnnotationProperty as well as the representation of rdf:Property in the profile sections, and were not sure what the issue was here. Gordon will provide some examples so that we can work through this one at the next telecon. - Roy Bell mentioned that he would begin working through some of the issues in chapter 16; Bob Colomb has also offered to assist via email 4. New Actions: ----------- - Elisa and Evan will work out the WebEx link to include in future agenda messages - For issues 11103, 11104, and 11106: Elisa will revise the metamodels, find related text, and propose resolutions based on the results of discussion today - For issues 11105 and 11107, Gordon will provide examples to support additional discussion 5. Administrative Next Wednesday, which would normally be the next scheduled telecon, is a holiday in the US. Thus, we will cancel the telecon on July 4. Also, the following week, July 11, both co-chairs will be attending the 2007 Metadata Open Forum meeting in New York, so not sure whether or not we will both be able to attend the call, depending on the agenda. We will follow up and confirm, likely that week. Revised spreadsheet including updated status of issues discussed in todays telecon is posted to the wiki, at: http://www.omgwiki.org/odmftf/doku.php - Next Scheduled Telecon: Telecon will be held (tentatively), Wednesday 11 July. Subject: RE: Minutes from ODM-FTF telecon of 27 June Date: Wed, 11 Jul 2007 04:02:32 -0700 X-MS-Has-Attach: yes X-MS-TNEF-Correlator: Thread-Topic: Minutes from ODM-FTF telecon of 27 June Thread-Index: AcfCoeaLSsxJzTliRAqwb7pZDA4GEQBCIOsw From: "Pete Rivett" To: "Guo Tong Xie" , "Elisa F. Kendall" Cc: "Bob Colomb" , "Juergen Boldt" , For item 1 (11105), I don't see the need for 2 associations (in fact as far as I can see it would be a bad thing since there's only one conceptual relationship and the 2 associations would need to be maintained in sync with each other): why not just change the multiplicity of property URIReference.namespace from 0..1 to 0..*? Pete Pete Rivett (mailto:pete.rivett@adaptive.com) CTO, Adaptive Inc. Hello House, 135 Somerford Road, Christchurch, BH23 3PY, UK Tel: +44 (0)1202 491243 Fax: +44 (0)1202 491241 http://www.adaptive.com -------------------------------------------------------------------------------- From: Guo Tong Xie [mailto:XIEGUOT@cn.ibm.com] Sent: Tuesday, July 10, 2007 4:29 AM To: Elisa F. Kendall Cc: Bob Colomb; Juergen Boldt; odm-ftf@omg.org Subject: Re: Minutes from ODM-FTF telecon of 27 June Please find my followup as below. 1. Issue 11105, chapter 10, section 10.7.3, URIReferenceForNamespace should be changed to two uni-directional associations between URIReference and Namespace. In the current model, if two URIReference have the same Namespace, it could not be represented. Actually, we use two uni-directional associations when implementing EODM (http://www.eclipse.org/modeling/mdt/?project=eodm#eodm). 2. Issue 11107, chapter 14, section 14.2.3.1, it is not clear on how to set values for annotation property. In the following example, an AnnotationProperty, creator, is defined. Then an owl:Class, MusicalWork, is defined and its creator is "N.N.". Actually, this is an exmaple from http://www.w3.org/TR/owl-ref/#Annotations. Musical work N.N. The question is that I have no idea on how to set creator of MusicalWork as N.N. by using the current profile. Regards, Guo Tong (Gordon) Xie (...) Manager, Semantic Integration IBM China Research Lab Email: xieguot@cn.ibm.com Tel: 86-10-58748425 Tieline: 905-8425 Fax: 86-10-58748230 Homepage: http://www.research.ibm.com/people/x/xieguot Address: Building 19 Zhongguancun Software Park, 8 Dongbeiwang WestRoad, Haidian District, Beijing,P.R.C.100094 "Elisa F. Kendall" 2007-06-28 05:15 To odm-ftf@omg.org cc Juergen Boldt , Bob Colomb Subject Minutes from ODM-FTF telecon of 27 June [[Juergen: Would you please put this on the OMG server? Thanks.]] Minutes from ODM-FTF Telecon 27 June 2007 Scribe: Elisa Kendall Participants: Roy Bell Raytheon Elisa Kendall Sandpiper Software Evan Wallace NIST Chris Welty IBM Watson Research Guo Tong Xie IBM China Research Regrets: Saartje Brockmans (FZI/AIFB), Bob Colomb (Universiti Teknologi Malaysia) ***** AGENDA ***** 1. Admin - Approved: minutes of meeting held 20 June 2007 (recorded at http://www.omg.org/cgi-bin/doc?ptc/2007-06-07) 2. Review of outstanding actions Outstanding Action Items prior to telecon: Continues Elisa Kendall to talk with Chris Welty, Pete Rivett, and Bob Colomb to determine their plans wrt the ODM-FTF participation. - Chris Welty & Bob Colomb contacted; Pete Rivett remaining Continues Bobbin Teegarden to talk with Pete Rivett to determine what type of file is needed for the UML profiles in ODM and submit issues requesting these be included for the finalized specification. Continues Saartje to look at issue 10843 (issue regarding whether subclassing of OWL Class takes one out of OWL DL) Continues Chris will post the issues he and other IBM colleagues noted in reviewing the RDF and OWL metamodel chapters in October. Continues Elisa will contact Sreedhar Reddy regarding availability to assist with the QVT vetting for finalization purposes. Complete Elisa will revise the spreadsheet per discussion and upload it to the wiki, add URI to the agenda Complete Elisa will add content for issues outside of chapter 16 (by Roy Bell) Complete Roy will add issues related to chapter 16 Continues Evan will add a page to the wiki for managing issues 3. Discussion of outstanding issues / work items - Discussed WebEx access issues -- Elisa and Evan will attempt to determine the right link to include in future agenda email - Reviewed several issues submitted by Gordon, including 11103 through 11107, as follows 1. Issue 11103, chapter 10, section 10.4.1: the role name of superClass in ClassGeneralization is confusing. Discussion: agreed that superClass seems synonymous with RDFSsubClassOf, and that it should be changed, potentially to superClassOf. Elisa will propose a new figure and search the text to find any relevant references that require revision for resolution, to be discussed at the next telecon. 2. Issue 11104, chapter 10, section 10.5.1: the role name of superProperty in PropertyGeneralization is confusing. Discussion: agreed that superProperty seems synonymous with RDFSsubPropertyOf, and that it should be changed, potentially to superPropertyOf, similarly to 11103, above. Elisa will propose a new figure and search the text to find any relevant references that require revision for resolution, to be discussed at the next telecon. 3. Issue 11105, chapter 10, section 10.7.3, URIReferenceForNamespace should be changed to two uni-directional associations between URIReference and Namespace. In the current model, if two URIReference have the same Namespace, it could not be represented. Discussion: participants thought that we understood the issue, but for clarification and to provide more detail for discussion, Gordon will send examples, including the work-around adopted by IBM to use two uni-directional associations. 4. Issue 11106, chapter 11, section 11.4.5, the cardinality of OWLInverseOf in InverseProperty association should be changed from 0..1 to 0..*. One property can have multiple inverse properties Discussion: we used the example of manages and is managed by, as two inverse properties, with equivalent properties bosses and is bossed by to try to tease out this issue. What Gordon was saying is that is managed by should be capable of having both manages and bosses as inverses, and the current metamodel only allows for a single inverse. After discussion, we agreed that the cardinality should be changed from 0..1 to 0..*. Elisa will propose a new figure, and search the text to find any relevant references that require revision for resolution, to be discussed at the next telecon. 5. Issue 11107, chapter 14, section 14.2.3.1, it is not clear on how to set values for annotation property Discussion: we reviewed the representation for owl:AnnotationProperty as well as the representation of rdf:Property in the profile sections, and were not sure what the issue was here. Gordon will provide some examples so that we can work through this one at the next telecon. - Roy Bell mentioned that he would begin working through some of the issues in chapter 16; Bob Colomb has also offered to assist via email 4. New Actions: ----------- - Elisa and Evan will work out the WebEx link to include in future agenda messages - For issues 11103, 11104, and 11106: Elisa will revise the metamodels, find related text, and propose resolutions based on the results of discussion today - For issues 11105 and 11107, Gordon will provide examples to support additional discussion 5. Administrative Next Wednesday, which would normally be the next scheduled telecon, is a holiday in the US. Thus, we will cancel the telecon on July 4. Also, the following week, July 11, both co-chairs will be attending the 2007 Metadata Open Forum meeting in New York, so not sure whether or not we will both be able to attend the call, depending on the agenda. We will follow up and confirm, likely that week. Revised spreadsheet including updated status of issues discussed in todays telecon is posted to the wiki, at: http://www.omgwiki.org/odmftf/doku.php - Next Scheduled Telecon: Telecon will be held (tentatively), Wednesday 11 July. From: ewallace@cme.nist.gov Date: Tue, 21 Aug 2007 12:53:48 -0400 (EDT) To: odm-ftf@omg.org Subject: Issue 11105 X-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-MailScanner: Found to be clean X-MailScanner-SpamCheck: X-MailScanner-From: ewallace@mailhost.mel.nist.gov X-Spam-Status: No I have been going through the proposed resolutions posted in the ODM-FTF wiki [1] for Gordon's issues 11103-11105, and I don't think I agree with the resolution for 11105. The issue reads: Bi-directional URIReferenceForNamespace association In section 10.7.3, URIReferenceForNamespace should be changed to two uni-directional associations between URIReference and Namespace. In the current model, if two URIReference have the same Namespace, it could not be represented. But the proposed resolution states: Given that there is really one conceptual relationship, per discussion ... we propose chaing the multiplicity of property URIReferenceForNamespace from 0..0 to 0..* to address this. As I understand the issue, we are talking about two relationships between these classes and not one. Of course, this all depends on what Gordon meant by "have the same Namespace". I think what he is looking for is a role that would yield links to all the URIReferences "contained" in a Namespace. The current association between these two classes is meant to represent URIReferences which identify namespaces, hence the association name URIReferenceForNamespace, while the URIReferences in question would identify elements in an owl ontology or rdfs vocabulary and not Namespaces. As the model stands, there is no easy way to derive this information, so I think there is motivation for adding a new association. That said, the model fragment in [2] wouldn't be my solution, and doesn't appear to solve the problem identified since it still has a multiplicity of 1 URIReference for the namespaceURIRef association. If Gordon wants the model to support finding the URIRefs contained in a Namespace then a second bidirectional relationship should be added for this purpose. If, however, Gordon merely wants support for additional URIRefs identifying a single namespace then I agree we should just weaken the multiplicity constraint on URIReferenceForNamespace as proposed in the posted resolution. If the latter is what is wanted, what are the use-cases for having multiply identifying URIRefs for a namespace? -Evan [1] http://www.omgwiki.org/odmftf/doku.php [2] http://www.omg.org/archives/odm-ftf/msg00142.html X-YMail-OSG: SJP5.BMVM1kHTwJZIU8AaoAANll_Og31R8TosF79ITDcOwR5L9JIPI8gHiYck7sXMC._H3KGXi3sXicLrBq8Qb_qfNDq35myd._bgI8Sh40Bg9t9PUGny.AJswqof1q_oXAB54TvmbyQJsU- Date: Tue, 21 Aug 2007 10:43:59 -0700 From: "Elisa F. Kendall" Organization: Sandpiper Software, Inc. User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.2) Gecko/20040804 Netscape/7.2 (ax;nscd1) X-Accept-Language: en-us, en To: ewallace@cme.nist.gov Cc: odm-ftf@omg.org Subject: Re: Issue 11105 See some additional comments, below. Elisa ewallace@cme.nist.gov wrote: I have been going through the proposed resolutions posted in the ODM-FTF wiki [1] for Gordon's issues 11103-11105, and I don't think I agree with the resolution for 11105. The issue reads: Bi-directional URIReferenceForNamespace association In section 10.7.3, URIReferenceForNamespace should be changed to two uni-directional associations between URIReference and Namespace. In the current model, if two URIReference have the same Namespace, it could not be represented. But the proposed resolution states: Given that there is really one conceptual relationship, per discussion ... we propose chaing the multiplicity of property URIReferenceForNamespace from 0..0 to 0..* to address this. As I understand the issue, we are talking about two relationships between these classes and not one. Of course, this all depends on what Gordon meant by "have the same Namespace". I think what he is looking for is a role that would yield links to all the URIReferences "contained" in a Namespace. The current association between these two classes is meant to represent URIReferences which identify namespaces, hence the association name URIReferenceForNamespace, while the URIReferences in question would identify elements in an owl ontology or rdfs vocabulary and not Namespaces. As the model stands, there is no easy way to derive this information, so I think there is motivation for adding a new association. That said, the model fragment in [2] wouldn't be my solution, and doesn't appear to solve the problem identified since it still has a multiplicity of 1 URIReference for the namespaceURIRef association. If Gordon wants the model to support finding the URIRefs contained in a Namespace then a second bidirectional relationship should be added for this purpose. If, however, Gordon merely wants support for additional URIRefs identifying a single namespace then I agree we should just weaken the multiplicity constraint on URIReferenceForNamespace as proposed in the posted resolution. If the latter is what is wanted, what are the use-cases for having multiply identifying URIRefs for a namespace? I understood this to mean support for additional URIRefs, which is why we took the approach given in the proposal, but if this isn't right we could change the model as you suggest. It's possible that there is motivation for both if Gordon can assist with the description of what he really needs. Thanks, Elisa -Evan [1] http://www.omgwiki.org/odmftf/doku.php [2] http://www.omg.org/archives/odm-ftf/msg00142.html To: "Elisa F. Kendall" Cc: Bob Colomb , odm-ftf@omg.org Subject: Re: [ODM-FTF] Proposals for review, issues 11103-11106 X-Mailer: Lotus Notes Release 7.0 HF277 June 21, 2006 From: Guo Tong Xie Date: Fri, 24 Aug 2007 18:19:52 +0800 X-MIMETrack: Serialize by Router on D23M0037/23/M/IBM(Release 7.0.2HF32 | October 17, 2006) at 08/24/2007 18:19:55, Serialize complete at 08/24/2007 18:19:55 Elisa, Let me use a simple example to illustrate issue 11105. Following is a simple RDF document from http://www.w3.org/TR/rdf-primer/. Eric Miller Dr. If we represent above RDF document in RDF/OWL metamodel, there would be: 1. Two namespaces, xmlns:rdf and xmlns:contact 2. Eight URIReferences as below 1) http://www.w3.org/1999/02/22-rdf-syntax-ns# 2) http://www.w3.org/2000/10/swap/pim/contact# 3) http://www.w3.org/People/EM/contact#me 4) http://www.w3.org/2000/10/swap/pim/contact#Person 5) http://www.w3.org/2000/10/swap/pim/contact#fullName 6) http://www.w3.org/2000/10/swap/pim/contact#mailbox 7) http://www.w3.org/2000/10/swap/pim/contact#personalTitle 8) mailto:em@w3.org Then there would be two kinds of relationships between above Namespace and URIReference. Relationship1(http://www.w3.org/1999/02/22-rdf-syntax-ns#, xmlns:rdf) Relationship1(http://www.w3.org/2000/10/swap/pim/contact#, xmlns:contact) Of course, you may want inverse relationship of Relationship1, which is InverseOfRelationship1. Then you would have InverseOfRelationship1(xmlns:rdf, http://www.w3.org/1999/02/22-rdf-syntax-ns#) InverseOfRelationship1(xmlns:contact, http://www.w3.org/2000/10/swap/pim/contact#) At the sametime, there would be Relationship2, Relationship2(http://www.w3.org/2000/10/swap/pim/contact#Person, xmlns:contact) Relationship2(http://www.w3.org/2000/10/swap/pim/contact#fullName, xmlns:contact) Relationship2(http://www.w3.org/2000/10/swap/pim/contact#mailbox, xmlns:contact) Relationship2(http://www.w3.org/2000/10/swap/pim/contact#personalTitle, xmlns:contact) I guess the issue is much clear now. In the orignial deisgn, we have ONE bi-directional association between Namespace and URIReference. I guess its intention is to represent Relationship1 and InverseOfRelationship1 (according to the semantics of bi-directional association). More precisely, Relationship1 is namespace and InverseOfRelationship1 is namespaceURIRef. Then Relationship2 is MISSING. So my proposal is to ADD one uni-directional association to represent Relationship2, which is missing in the current metamodel. Changing cardinality of association end - namespace, does not help to represent Relationship2. I hope I make myself clear. Thanks. Regards, Guo Tong (Gordon) Xie (л¹úÍ®) Manager, Semantic Integration IBM China Research Lab Email: xieguot@cn.ibm.com Tel: 86-10-58748425 Tieline: 905-8425 Fax: 86-10-58748230 Homepage: http://www.research.ibm.com/people/x/xieguot Blog: http://blogs.tap.ibm.com/weblogs/xieguot@cn.ibm.com Address: Building 19 Zhongguancun Software Park, 8 Dongbeiwang WestRoad, Haidian District, Beijing,P.R.C.100094 "Elisa F. Kendall" 2007-08-21 04:21 To odm-ftf@omg.org cc Bob Colomb Subject [ODM-FTF] Proposals for review, issues 11103-11106 Hi all, I've drafted resolutions to issues 11103 thru 11106 based on discussion, Pete's email, document search, etc. Please take a look at these so that we can put them to vote after the call on Wednesday. Thanks, Elisa [attachment "resolution_11103_11105.doc" deleted by Guo Tong Xie/China/IBM] [attachment "resolution_11106.doc" deleted by Guo Tong Xie/China/IBM] In section 10.7.3, URIReferenceForNamespace should be changed to two uni-directional associations between URIReference and Namespace. In the current model, if two URIReference have the same Namespace,