Issue 6445: Clarification of use case semantics (uml2-rtf) Source: Pivot Point (Mr. Cris Kobryn, ) Nature: Uncategorized Issue Severity: Summary: Use cases and associated sequence and activity diagrams are widely used by systems engineers to specify the functionality of the system, and describe the interaction between the system and the actors. However, there is much confusion regarding use case semantics. Consider the following recommendations to clarify use case semantics: a) Establish an explicit representation to depict the relationship between a use case and its realization as a sequence diagram, activity diagram, etc. In addition, allow for a similar representation to show the relationship between an included/extended use case, and the interaction fragments which realize the included/extended use case. b) Clarify the relationship between a use case (solid oval) and a collaboration (dashed oval), and determine whether they can represent the same or a similar concept. Resolution: Disposition: Deferred to UML 2.4 RTF Revised Text: Actions taken: November 6, 2003: received issue Discussion: UML already supports this capability since it allows a classifier (in this case a use case) to have associated a number of behaviors that are in some ways related to the classifier (e.g., they may represent realizations of the classifier in some way, or illustrations of specific executions related to the classifier, etc.) However, the proposed resolutions are asking for a specialization of these general (meta)associations that would have special realization semantics. Such semantics might prove to be tricky to define in the case of use cases, and require some study (i.e., what does it mean for a behavior to “realize” a use case? Is it just an illustration of one possible realization or is it more general?) These are, in effect, requirements for new modeling features and, therefore, fall outside the scope of an FTF or RTF. This should be the subject of a separate RFP. End of Annotations:===== eference: UML 2 Superstructure, OMG doc# ptc/03-08-02, Chapter 16 Issue: Clarification of use case semantics Description: Use cases and associated sequence and activity diagrams are widely used by systems engineers to specify the functionality of the system, and describe the interaction between the system and the actors. However, there is much confusion regarding use case semantics. Consider the following recommendations to clarify use case semantics: a) Establish an explicit representation to depict the relationship between a use case and its realization as a sequence diagram, activity diagram, etc. In addition, allow for a similar representation to show the relationship between an included/extended use case, and the interaction fragments which realize the included/extended use case. b) Clarify the relationship between a use case (solid oval) and a collaboration (dashed oval), and determine whether they can represent the same or a similar concept. Reply-To: From: "Cris Kobryn" To: "'Branislav Selic'" , Cc: Subject: RE: Ballot 3 Reissue -- discard old version - resolving critical-path issues for SysML Date: Fri, 20 May 2005 21:53:29 -0700 X-Mailer: Microsoft Office Outlook, Build 11.0.6353 Thread-Index: AcVaV0jNHlGPuu8LTqi4zQyS8duBhwDYS9gg > My apologies to all, but due to some belated input (and a screw up on my part), > I have removed 4 proposed resolutions from Ballot 3. They are > > * 6126, 6372, and 8161 (removed until Actions/Activities groups reach consensus on them) > * 6446 (removed because SysML raised objections to the current resolution) There were two issues, not one, on the original Ballot 3 that the SysML Partners submitted and do not want to be deferred: "Issue 6445: Clarification of use case semantics" and "Issue 6446: Clarification of Information Flow semantics". I submitted both of these issues on behalf on the SysML Partners on November 6, 2003, more than 1.5 years ago. These issues are critical path for SysML because we cannot complete the specification of SysML Requirements and ItemFlow constructs without further clarification of the semantics of UseCases and InformationFlows, respectively. Removing Issue 6446 from Ballot 3 may be a start, but raises the questions of when and how this issue will be resolved by the current RTF. Deferring Issue 6445 on the Ballot 3 reissue is also problematic for us, for the reason explained above. Given the importance of these constructs to SysML, if the RTF is unable to provide us with bona fide resolutions (contrast "Closed, No Change", "Defer" resolutions) for these issues we will need to provide recommended changes to the UML2 Superstructure specification with our final SysML submission. Thanks, Cris ________________________________ From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Monday, May 16, 2005 1:33 PM To: uml2-rtf@omg.org Subject: Ballot 3 Reissue -- discard old version My apologies to all, but due to some belated input (and a screw up on my part), I have removed 4 proposed resolutions from Ballot 3. They are * 6126, 6372, and 8161 (removed until Actions/Activities groups reach consensus on them) * 6446 (removed because SysML raised objections to the current resolution) The new revised ballot is attached. Please discard the previous version. Regards, Bran To: Cc: SysMLpartners@googlegroups.com, uml2-rtf@omg.org, sanford.friedenthal@lmco.com Subject: RE: Ballot 3 Reissue -- discard old version - resolving critical-path issues for SysML X-Mailer: Lotus Notes Release 6.0.1CF1 March 04, 2003 From: Branislav Selic Date: Sat, 21 May 2005 10:39:13 -0400 X-MIMETrack: Serialize by Router on D25ML01/25/M/IBM(Release 6.53HF247 | January 6, 2005) at 05/21/2005 10:39:31, Serialize complete at 05/21/2005 10:39:31 Cris, The reason that 6445 was not withdrawn was because it was concluded that that the change requested in that issue was unnecessary since there was already a solution that did not require a change. This conclusion was reached after discussions with members of the SysML team. Specifically, since a UseCase is a behaviored classifier, it is possible to associate any number of behaviors (e.g., interactions) with a use case. The interpretation of those behaviors is up to the modeler. In the case of SysML, the profile can associate a specific semantics to one or more of those behaviors. The most obvious thing would be to make these behaviors represent the "realization" of a use case. This is precisely the kind of semantic "tightening" that profiles can provide without forcing those semantics on domains where they may not be appropriate. I would like to note that we are and have been working very actively and effectively with the SysML team particularly in these closing phases of your submission. For example, we had a very successful telecon yesterday with Alan and Sandy, where we all agreed on some additions to the profiles mechanism to ensure that SysML needs are met. We are also collaborating on resolutions to structural modeling issues raised by SysML. I think this cooperation is actually working out quite well. Cheers, Bran Selic IBM Distinguished Engineer IBM Rational Software 770 Palladium Drive Kanata, Ontario, Canada K2V 1C8 ph.: (613) 591-7915 fax: (613) 599-3912 e-mail: bselic@ca.ibm.com "Cris Kobryn" 05/21/2005 12:54 AM Please respond to Cris.Kobryn To Branislav Selic/Ottawa/IBM@IBMCA, cc Subject RE: Ballot 3 Reissue -- discard old version - resolving critical-path issues for SysML > My apologies to all, but due to some belated input (and a screw up on my part), > I have removed 4 proposed resolutions from Ballot 3. They are > > * 6126, 6372, and 8161 (removed until Actions/Activities groups reach consensus on them) > * 6446 (removed because SysML raised objections to the current resolution) There were two issues, not one, on the original Ballot 3 that the SysML Partners submitted and do not want to be deferred: "Issue 6445: Clarification of use case semantics" and "Issue 6446: Clarification of Information Flow semantics". I submitted both of these issues on behalf on the SysML Partners on November 6, 2003, more than 1.5 years ago. These issues are critical path for SysML because we cannot complete the specification of SysML Requirements and ItemFlow constructs without further clarification of the semantics of UseCases and InformationFlows, respectively. Removing Issue 6446 from Ballot 3 may be a start, but begs the questions of when and how this issue will be resolved by the current RTF. Keeping Issue 6445 on the Ballot 3 reissue is problematic for us, since this is also critical to finalize our specification. Given the importance of these constructs to SysML, if the RTF is unable to provide us with bona fide resolutions (contrast "Closed, No Change", "Defer" resolutions) for these issues we will need to provide recommended changes to the UML2 Superstructure specification with our final SysML submission. Thanks, Cris SysML Partners ________________________________ From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Monday, May 16, 2005 1:33 PM To: uml2-rtf@omg.org Subject: Ballot 3 Reissue -- discard old version My apologies to all, but due to some belated input (and a screw up on my part), I have removed 4 proposed resolutions from Ballot 3. They are * 6126, 6372, and 8161 (removed until Actions/Activities groups reach consensus on them) * 6446 (removed because SysML raised objections to the current resolution) The new revised ballot is attached. Please discard the previous version. Regards, Bran Reply-To: From: "Cris Kobryn" To: "'Branislav Selic'" , Cc: , Subject: RE: Ballot 3 Reissue -- discard old version - resolving critical-path issues for SysML Date: Sat, 21 May 2005 09:29:30 -0700 X-Mailer: Microsoft Office Outlook, Build 11.0.6353 Thread-Index: AcVeEygYn3Zrj3XfRmGA+LFwEa7crQABNMEA > Specifically, since a UseCase is a behaviored classifier, it is possible to associate any number of behaviors (e.g., interactions) with a > use case. The interpretation of those behaviors is up to the modeler. In the case of SysML, the profile can associate a specific > semantics to one or more of those behaviors. The most obvious thing would be to make these behaviors represent the "realization" > of a use case. This is precisely the kind of semantic "tightening" that profiles can provide without forcing those semantics > on domains where they may not be appropriate. I agree that defining specific behavioral realizations of Use Cases is the most obvious solution, but disagree that this should be the work of profiles or a separate RFP. The issue raised is general and critical; it is not specific to Systems Engineering applications. The ambiguous semantics of Use Case realization has been problematic since UML1, and will continue to be so until more precise semantics are specified and standardized. Given the numerous limitations and problems that we have encountered workining intensively with UML2 Profiles over the last 2 years, I am skeptical that we will be able to "tighten" the semantics of Use Cases without recommending changes to UML2 Superstructure in our final submission. We will have a better idea of what is needed after we sort out the final details of our numerous recommended changes to Profiles (see below). > I would like to note that we are and have been working very actively and effectively with the SysML team particularly in > these closing phases of your submission. For example, we had a very successful telecon yesterday with Alan and Sandy, > where we all agreed on some additions to the profiles mechanism to ensure that SysML needs are met. We are also > collaborating on resolutions to structural modeling issues raised by SysML. I think this cooperation is actually working > out quite well. While I agree that some good technical work has been done regarding fixing Profiles, the RTF process speed to fix critical problems has been less than impressive. For example, 9 months and still counting to fix obvious deficiencies with Profiles is less than quick, as is 1.5 years to resolve two critical path issues by a recommendation to "Defer". Hopefully, we can collaborate to address Issue 6446 in a more expeditious manner. Thanks, Cris -------------------------------------------------------------------------------- From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Saturday, May 21, 2005 7:39 AM To: Cris.Kobryn@SysML.org Cc: SysMLpartners@googlegroups.com; uml2-rtf@omg.org; sanford.friedenthal@lmco.com Subject: RE: Ballot 3 Reissue -- discard old version - resolving critical-path issues for SysML Cris, The reason that 6445 was not withdrawn was because it was concluded that that the change requested in that issue was unnecessary since there was already a solution that did not require a change. This conclusion was reached after discussions with members of the SysML team. Specifically, since a UseCase is a behaviored classifier, it is possible to associate any number of behaviors (e.g., interactions) with a use case. The interpretation of those behaviors is up to the modeler. In the case of SysML, the profile can associate a specific semantics to one or more of those behaviors. The most obvious thing would be to make these behaviors represent the "realization" of a use case. This is precisely the kind of semantic "tightening" that profiles can provide without forcing those semantics on domains where they may not be appropriate. I would like to note that we are and have been working very actively and effectively with the SysML team particularly in these closing phases of your submission. For example, we had a very successful telecon yesterday with Alan and Sandy, where we all agreed on some additions to the profiles mechanism to ensure that SysML needs are met. We are also collaborating on resolutions to structural modeling issues raised by SysML. I think this cooperation is actually working out quite well. Cheers, Bran Selic IBM Distinguished Engineer IBM Rational Software 770 Palladium Drive Kanata, Ontario, Canada K2V 1C8 ph.: (613) 591-7915 fax: (613) 599-3912 e-mail: bselic@ca.ibm.com "Cris Kobryn" 05/21/2005 12:54 AM Please respond to Cris.Kobryn To Branislav Selic/Ottawa/IBM@IBMCA, cc Subject RE: Ballot 3 Reissue -- discard old version - resolving critical-path issues for SysML > My apologies to all, but due to some belated input (and a screw up on my part), > I have removed 4 proposed resolutions from Ballot 3. They are > > * 6126, 6372, and 8161 (removed until Actions/Activities groups reach consensus on them) > * 6446 (removed because SysML raised objections to the current resolution) There were two issues, not one, on the original Ballot 3 that the SysML Partners submitted and do not want to be deferred: "Issue 6445: Clarification of use case semantics" and "Issue 6446: Clarification of Information Flow semantics". I submitted both of these issues on behalf on the SysML Partners on November 6, 2003, more than 1.5 years ago. These issues are critical path for SysML because we cannot complete the specification of SysML Requirements and ItemFlow constructs without further clarification of the semantics of UseCases and InformationFlows, respectively. Removing Issue 6446 from Ballot 3 may be a start, but begs the questions of when and how this issue will be resolved by the current RTF. Keeping Issue 6445 on the Ballot 3 reissue is problematic for us, since this is also critical to finalize our specification. Given the importance of these constructs to SysML, if the RTF is unable to provide us with bona fide resolutions (contrast "Closed, No Change", "Defer" resolutions) for these issues we will need to provide recommended changes to the UML2 Superstructure specification with our final SysML submission. Thanks, Cris SysML Partners ________________________________ From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Monday, May 16, 2005 1:33 PM To: uml2-rtf@omg.org Subject: Ballot 3 Reissue -- discard old version My apologies to all, but due to some belated input (and a screw up on my part), I have removed 4 proposed resolutions from Ballot 3. They are * 6126, 6372, and 8161 (removed until Actions/Activities groups reach consensus on them) * 6446 (removed because SysML raised objections to the current resolution) The new revised ballot is attached. Please discard the previous version. Regards, Bran To: Cc: sanford.friedenthal@lmco.com, SysMLpartners@googlegroups.com, uml2-rtf@omg.org Subject: RE: Ballot 3 Reissue -- discard old version - resolving critical-path issues for SysML X-Mailer: Lotus Notes Release 6.0.1CF1 March 04, 2003 From: Branislav Selic Date: Sat, 21 May 2005 22:22:56 -0400 X-MIMETrack: Serialize by Router on D25ML01/25/M/IBM(Release 6.53HF247 | January 6, 2005) at 05/21/2005 22:22:57, Serialize complete at 05/21/2005 22:22:57 Cris In reply: > Given the numerous limitations and problems that we have encountered > workining intensively with UML2 Profiles over the last 2 years, I am > skeptical that we will be able to "tighten" the semantics of Use > Cases without recommending changes to UML2 Superstructure in our > final submission. We will have a better idea of what is needed after > we sort out the final details of our numerous recommended changes to > Profiles (see below). I guess we'll have to wait for your input then. The issue raised in 6445 can be resolved with the following sentence: "Any behaviors owned by a use case specify concrete realizations of the owning use case." For dealing with the need to invoke the behaviors of included use cases: "In cases where a behavior realizing a use case needs to invoke an interaction realizing an included use case, a special modeling approach is required since owned behaviors cannot be exported. To enable this, it is necessary to define a parameterized interaction in a namespace that is accessible by both the including and the included uses cases. The behaviors of each use case can then invoke this parameterized behavior by substituting their appropriate local values for the parameters" Although something similar could also be done for extending use cases that would not be appropriate since one of the key features of this relationship between use cases is that the extended use case MUST NOT have any reference to the extending use case. > While I agree that some good technical work has been done regarding > fixing Profiles, the RTF process speed to fix critical problems has > been less than impressive. For example, 9 months and still counting > to fix obvious deficiencies with Profiles is less than quick, as is > 1.5 years to resolve two critical path issues by a recommendation to > "Defer". Hopefully, we can collaborate to address Issue 6446 in a > more expeditious manner. As you well know from your long-term experience with both UML and SysML, it is not easy to reconcile over a dizen different points of view expeditiously. However, I would like to point out that the UML task forces have handled over 800 issues in the first cycle and have already put away over a 150 additional issues in the past 6 weeks. I think this is exemplary performance and the team should be commended rather reprimanded. Bran