Issue 11080: Support UML 2 Profiles for SPEM 2.0 meta-model instances (spem2-ftf) Source: International Business Machines (Dr. Peter Haumer, phaumer@us.ibm.com) Nature: Enhancement Severity: Significant Summary: Meta-Model: Support UML 2 Profiles for SPEM 2.0 meta-model instances The SPEM 2.0 meta-model provides a light-weight extensibility mechanism via the Extensible Element and Kind classes that allow defining special kinds of meta-model elements such as “artifact” versus “deliverable” work product kinds. However, often when the kinds are defined meta-model users also want to define specific attributes or associations for the stereotyped instances such as in UML 2 Profiles. Provide as an additional alternative to the light-weight extensibility mechanism provided in the current specification also the ability to define and UML Profiles for SPEM 2 meta-model instances. As all classes in the SPEM 2 meta-model are derived from classifier the only model change to realize this capability needed would be to merge Infrastructure::Profiles into the SPEM 2 meta-model. Resolution: - Merged Profiles package from UML2 Infrastructure library in SPEM Complete compliance level. Revised Text: - Added this text to Section 2.3: "It is also the only compliance point that merges the Profiles package from the UML 2 Infrastructure, which provides a more complete extensibility mechanism than the light extensibility mechanisms provided in SPEM 2.0 itself. Whereas SPEM 2.0 provides the ability to define instances of a Kind class that allow associating special semantics to meta class instances, UML Profiles allow in addition to that creating and managing stereotype application instances that can store user-defined property values defined for the stereotype with the stereotype instance. SPEM Complete implementers shall either provide both extensibility mechanisms or a mapping of UML Profile stereotypes to SPEM 2 Kinds. The other compliance levels defined in the specification do not require realizing Profiles because the light-weight extensibility mechanisms of SPEM should be sufficient for the audiences listed, but implementers have the option to do so for these levels as well." - Updated Figure 2.3 Actions taken: May 31, 2007: received issue January 15, 2008: closed issue Discussion: End of Annotations:===== m: webmaster@omg.org Date: 30 May 2007 22:57:46 -0400 To: Subject: Issue/Bug Report -------------------------------------------------------------------------------- Name: Peter Haumer Company: IBM mailFrom: phaumer@us.ibm.com Notification: Yes Specification: Peter Haumer Section: 2 FormalNumber: ptc/07-03-03 Version: 2.0 RevisionDate: March/12/2007 Page: 1ff Nature: Enhancement Severity: Significant HTTP User Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.3) Gecko/20070309 Firefox/2.0.0.3 Description Meta-Model: Support UML 2 Profiles for SPEM 2.0 meta-model instances The SPEM 2.0 meta-model provides a light-weight extensibility mechanism via the Extensible Element and Kind classes that allow defining special kinds of meta-model elements such as .artifact. versus .deliverable. work product kinds. However, often when the kinds are defined meta-model users also want to define specific attributes or associations for the stereotyped instances such as in UML 2 Profiles. Provide as an additional alternative to the light-weight extensibility mechanism provided in the current specification also the ability to define and UML Profiles for SPEM 2 meta-model instances. As all classes in the SPEM 2 meta-model are derived from classifier the only model change to realize this capability needed would be to merge Infrastructure::Profiles into the SPEM 2 meta-model. Subject: RE: Call for vote: SPEM 2 FTF Issues and Report Date: Mon, 27 Aug 2007 09:21:00 -0700 X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: Call for vote: SPEM 2 FTF Issues and Report Thread-Index: AcflU85JLde6Bk7STJK+WvsqNWLOWwDbpErA From: "Pete Rivett" To: "Peter Haumer" , Cc: , , , , , Adaptive votes YES to all the proposed resolutions with the exception of: Issue 11080 to which it votes NO. This is because: a) I don't think the profile issue has been fully thought through in terms of interoperability. b) one of the assumptions does not seem relevant " As all classes in the SPEM 2 meta-model are derived from classifier" c) the following is IMHO too open-ended " SPEM Complete implementers shall either provide both extensibility mechanisms or a mapping of UML Profile stereotypes to SPEM 2 Kinds" d) it does not address Chapter 18 (the base plugin) - e.g. presumably this should also be framed as a SPEM profile (as well as a UML profile?) Issue 11084 to which it ABSTAINS. I don't see this as a problem worth addressing - especially since it 'breaks' existing implementations. [Appendix B] to which it ABSTAINS. Because there is no statement of the issue/problem being addressed with the new B.5 - hence there is no formal justification for the new material. 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 PS I assume you are goign to replace all the red editorial guidance from the FTF report template. -------------------------------------------------------------------------------- From: Peter Haumer [mailto:phaumer@us.ibm.com] Sent: Thursday, August 23, 2007 8:03 AM To: spem2-ftf@omg.org Cc: Pete Rivett; chris.armstrong@aprocessgroup.com; james.d.baker@baesystems.com; miyazaki.hir-02@jp.fujitsu.com; Xabier.Larrucea@esi.es; reda.bendraou@lip6.fr; Philippe.Desfray@softeam.fr Subject: Call for vote: SPEM 2 FTF Issues and Report Importance: High Dear SPEM 2.0 FTF member. Find attached my proposal for the SPEM 2.0 FTF report to be submitted to the OMG next week. The updated specification document with change bars seemed to be too large for an attachment, so you can download it from here instead: http://haumer.net/paper/SPEM2-FTF.zip I would like you to review the report and vote on each issue using the following ballot. Please place a YES or NO before each issue number to vote. For the last issue I have not received a number from the OMG, yet; so I called it "Appendix B" in the interim. See the list of FTF members who are allowed to vote below the ballot. The vote ends Saturday 9/25 at Noon GMT. Please, reply to this email list by this time with your votes. My apologies about the short notice. Was a bit surprised about the deadline change and the amount of work this actually was. >>>>> SPEM 2.0 FTF Ballot Issue 11077: Wrong Specification Name Issue 11078: Fix stereotype extensions for components and update examples in spec Issue 11079: UML 2 Profile: Missing Profile meta-model diagrams Issue 11080: Support UML 2 Profiles for SPEM 2.0 meta-model instances Issue 11081: Meta-Model: Missing relationships in Process Behavior Issue 11082: Responsibility Assignment Maps shall be specific for the Work Product State Issue 11083: Meta-Model: Missing association of Task Definition to Qualification Issue 11084: Remove the word .Map. from various classes in the meta-model Issue 11266: Page: 159ff Issue 11284: Section: 14.11 and B.4 Issue Appendix B: Appendix B is out of date. Stereotypes in all images need updating and a section on extends-replace is missing. <<<<< Voting members: Adaptive, Pete Rivett Armstrong Process Group, Chris Armstrong BAE SYSTEMS, J.D. Baker Fujitsu, Hiroshi Miyazaki Fundacion European Software Institute, Xabier Larrucea International Business Machines, Peter Haumer (CHAIR) Laboratoire Informatique de Paris 6, Reda Bendraou Softeam, Philippe Desfray Because one of you reported problems with the spem2-ftf email list I will also cc this email to all the email addresses I have from you. I apologize for the duplication, but I want to make sure that everyone gets this. As you see in the report I was able to address all issues reported. However, I faced some challenges with Framemaker. It would be nice if someone could give me pointers with these issues to prepare the correct document for Juergen next week. - For Issue 11077 I need to update all the running footers with the new specification name "Software & Systems Process Engineering Meta-Model". I tried to do this in Chapter 8 Core, but did not succeed as the version number is cut off now. I would also assume that one can do this change in one place, but did not find out how. - We sometimes have tables nested inside tables. Changes I make in these nested tables (e.g. Section 14.11) do not create a change bar. How do I create one? - No idea how to insert a caption for an image frame. Was able to add a line below the frame that shows the Figure label and the right numbers, but this line seems different than the Figure capture created before. Need help to make them consistent. - I inserted a new figure in Chapter 13, but the figure caption is numbered as 12.1. All following figures are numbered correctly (13.2, 13.3 etc.). Have no idea how to change it to 13. Thanks and best regards, Peter Haumer. ______________________________________________________________ PETER HAUMER, Dr. rer. nat. Rational Method Composer | Eclipse Process Framework Rational Software | IBM Software Group Tel.: +1 (408) 463-5096