Issue 6465: UML 2 Issue: Include(s) and Extend(s) (uml2-superstructure-ftf) Source: Simula Research Laboratory (Mr. Bran Selic, selic(at)acm.org) Nature: Uncategorized Issue Severity: Summary: PROBLEM STATEMENT The notation for the Extend and Include relationships is a dashed arrow with open arrow and the keyword <<extend>> or <<include>> (Superstructure, pp. 516, 518). Nevertheless, the notation examples given in pages 521, 523 and 524 write "extends" and "includes", with an final "s". The other examples are allright. PROPOSED SOLUTION Fix the notation examples. Resolution: see above Revised Text: Actions taken: November 7, 2003: received issue March 5, 2004: moved to the Superstructure FTF from Infrastructure March 8, 2005: closed issue Discussion: ?? Replace the diagram in figure 406 on page 521 with the following diagram: ?? package TransactionUseCases package ATM Services package Administration Card Identification Perform ATM Transaction On-Line Help «extend» condition: {customer selected HELP} extension point: Selection Withdraw Transfer Funds Deposit Funds «includes» «include» Read Log Register ATM at Bank ?? Replace the Notation entry for “Extend” in Table 22 on pg. 523 with the following diagram Perform ATM Transaction «extend» extension points Selection ?? Replace the Notation entry for “Extend (with Condition)” in Table 22 on pg. 523 with the following diagram «extend» Condition: {customer selected HELP} extension point: Selection ?? Replace the Notation entry for “Include” in Table 22 on pg. 524, with the following diagram: Withdraw Card Identification End of Annotations:===== ubject: Fw: UML 2 Issue: Include(s) and Extend(s) X-Mailer: Lotus Notes Release 6.0.2CF1 June 9, 2003 From: Branislav Selic Date: Fri, 7 Nov 2003 08:31:24 -0500 X-MIMETrack: Serialize by Router on D25ML05/25/M/IBM(Release 6.0.2CF1|June 9, 2003) at 11/07/2003 08:31:42, Serialize complete at 11/07/2003 08:31:42 Issue raised by Gonzalo Genova. Bran Selic IBM Software Group -- Rational Software 770 Palladium Drive Kanata, Ontario, Canada K2V 1C8 ph. (613) 591-7915 fax (613) 599-3912 e-mail: bselic@ca.ibm.com ----- Forwarded by Branislav Selic/Ottawa/IBM on 11/07/2003 07:49 AM ----- Gonzalo Genova 11/06/2003 01:34 PM To: "'issues@omg.org'" , Branislav Selic/Ottawa/IBM@IBMCA cc: "Juan Llorens (inf)" , "J.Miguel Fuentes Torres" Subject: UML 2 Issue: Include(s) and Extend(s) This issue refers to UML 2.0 Infrastructure Specification (ptc/03-09-15) and UML 2.0 Superstructure Specification (ptc/03-08-02). PROBLEM STATEMENT The notation for the Extend and Include relationships is a dashed arrow with open arrow and the keyword <> or <> (Superstructure, pp. 516, 518). Nevertheless, the notation examples given in pages 521, 523 and 524 write "extends" and "includes", with an final "s". The other examples are allright. PROPOSED SOLUTION Fix the notation examples. OMG Issue No: 6465 Title: UML 2 Issue: Include(s) and Extend(s) Source: International Business Machines (Mr. Bran Selic, bselic@ca.ibm.com) Summary: PROBLEM STATEMENT The notation for the Extend and Include relationships is a dashed arrow with open arrow and the keyword <> or <> (Superstructure, pp. 516, 518). Nevertheless, the notation examples given in pages 521, 523 and 524 write "extends" and "includes", with an final "s". The other examples are allright. PROPOSED SOLUTION Fix the notation examples. Discussion: · Replace the diagram in figure 406 on page 521 with the following diagram: · · Replace the Notation entry for .Extend. in Table 22 on pg. 523 with the following diagram · Replace the Notation entry for .Extend (with Condition). in Table 22 on pg. 523 with the following diagram · Replace the Notation entry for .Include. in Table 22 on pg. 524, with the following diagram: Disposition: Resolved To: "Eran Gery" Cc: uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 X-Mailer: Lotus Notes Release 6.0.2CF1 June 9, 2003 From: Branislav Selic Date: Sun, 7 Mar 2004 21:18:12 -0500 X-MIMETrack: Serialize by Router on D25ML05/25/M/IBM(Release 6.0.2CF1|June 9, 2003) at 03/07/2004 21:18:14, Serialize complete at 03/07/2004 21:18:14 Hi Eran, My answers are embedded below: > Bran - Is the notation used in figured 406 (and 405) defined > somewhere ? I refer to the package "frames" and > notes for defining local attributes - ? what is it based on ? I know > these frames are used in sequence diagrams, > but did we introduce them as a generic notational technique for > namespacing ? Where ? Please look at Appendix B: Diagrams. Using frames is an optional way of denoting a number of different things in UML 2.0. > In case of package nesting, I would use a package symbol instead the > proposed frame notation. For class internal structure, we use a > class symbol as the "frame" - so we need to be consistent with that. You can use either one. Cheers, Bran From: Eran Gery To: Branislav Selic Cc: uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 Date: Mon, 8 Mar 2004 13:05:52 +0200 X-Mailer: Internet Mail Service (5.5.2656.59) Bran First thanks for pointing me there... Apologize for being a pain, but these notational issues are what end users really care about... From what I saw, the frame is only for a diagram. So figure 405 and 406 can't really be rendered in a UML tool as they suggest a diagram having multiple frames. Another important issue is that the appendix says Each diagram has a frame, a contents area, and a heading, see Figure 460. This is clearly an optional notation and should be presented as such I will log an issue on that. Eran -----Original Message----- From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Mon, March 08, 2004 4:18 AM To: Eran Gery Cc: uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 Hi Eran, My answers are embedded below: > Bran - Is the notation used in figured 406 (and 405) defined > somewhere ? I refer to the package "frames" and > notes for defining local attributes - ? what is it based on ? I know > these frames are used in sequence diagrams, > but did we introduce them as a generic notational technique for > namespacing ? Where ? Please look at Appendix B: Diagrams. Using frames is an optional way of denoting a number of different things in UML 2.0. > In case of package nesting, I would use a package symbol instead the > proposed frame notation. For class internal structure, we use a > class symbol as the "frame" - so we need to be consistent with that. You can use either one. Cheers, Bran To: Eran Gery Cc: uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 X-Mailer: Lotus Notes Release 6.0.2CF1 June 9, 2003 From: Branislav Selic Date: Mon, 8 Mar 2004 08:19:29 -0500 X-MIMETrack: Serialize by Router on D25ML05/25/M/IBM(Release 6.0.2CF1|June 9, 2003) at 03/08/2004 08:19:35, Serialize complete at 03/08/2004 08:19:35 Eran, I am not sure which issue you want to log, but you seem to be making some assumptions about diagrams that may not be shared. Namely, for some reason, you are assuming that you cannot have nested diagrams. This is a possibility that Birger and I discussed and, as far as I am concerned, it is still an option. In fact, I think that it is absolutely necessary when one needs to show things such as relationships between two or more models. Our problem is that there is no formal syntax defined for the notation of UML, merely a set of examples. This is still an outstanding item. If you raise an issue against figures 406, I can tell you that I will propose either a defer or a close, no change resolution for it (since I own that part of the text). I don't mean to be contrarian but I really believe that diagrams such as that are necessary in some cases. Regards, Bran Eran Gery wrote on 03/08/2004 06:05:52 AM: > Bran > > First thanks for pointing me there... > Apologize for being a pain, but these notational issues are what end > users really care about... > > From what I saw, the frame is only for a diagram. So figure 405 and > 406 can't really be rendered in > a UML tool as they suggest a diagram having multiple frames. > > Another important issue is that the appendix says > Each diagram has a frame, a contents area, and a heading, see Figure 460. > This is clearly an optional notation and should be presented as such > I will log an issue on that. > > Eran > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Mon, March 08, 2004 4:18 AM > To: Eran Gery > Cc: uml2-superstructure-ftf@omg.org > Subject: RE: Ballot 9 issue 6465 > > Hi Eran, > > My answers are embedded below: > > > > Bran - Is the notation used in figured 406 (and 405) defined > > somewhere ? I refer to the package "frames" and > > notes for defining local attributes - ? what is it based on ? I know > > these frames are used in sequence diagrams, > > but did we introduce them as a generic notational technique for > > namespacing ? Where ? > > Please look at Appendix B: Diagrams. Using frames is an optional way > of denoting a number of different things in UML 2.0. > > > In case of package nesting, I would use a package symbol instead the > > proposed frame notation. For class internal structure, we use a > > class symbol as the "frame" - so we need to be consistent with that. > > You can use either one. > > Cheers, > > Bran Date: Mon, 08 Mar 2004 15:03:38 +0100 From: Birger Møller-Pedersen User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en To: Branislav Selic CC: Eran Gery , uml2-superstructure-ftf@omg.org Subject: Re: Ballot 9 issue 6465 X-MailScanner-Information: This message has been scanned for viruses/spam. Contact postmaster@uio.no if you have questions about this scanning X-UiO-MailScanner: No virus found X-UiO-Spam-info: not spam, SpamAssassin (score=1.019, required 12, HTML_20_30 0.47, HTML_MESSAGE 0.00, HTML_TITLE_EMPTY 0.54) X-UiO-Spam-score: s Figure 405 is certainly an examle of nested diagrams. Figure 406 is not obviously. One may of course put a diagram around them all (or claim that there is an implicit frame - which is by the way also allowed according to the spec), but the the question is then if it is possible to have relations between diagrams. /birger Branislav Selic wrote: Eran, I am not sure which issue you want to log, but you seem to be making some assumptions about diagrams that may not be shared. Namely, for some reason, you are assuming that you cannot have nested diagrams. This is a possibility that Birger and I discussed and, as far as I am concerned, it is still an option. In fact, I think that it is absolutely necessary when one needs to show things such as relationships between two or more models. Our problem is that there is no formal syntax defined for the notation of UML, merely a set of examples. This is still an outstanding item. If you raise an issue against figures 406, I can tell you that I will propose either a defer or a close, no change resolution for it (since I own that part of the text). I don't mean to be contrarian but I really believe that diagrams such as that are necessary in some cases. Regards, Bran Eran Gery wrote on 03/08/2004 06:05:52 AM: > Bran > > First thanks for pointing me there... > Apologize for being a pain, but these notational issues are what end > users really care about... > > From what I saw, the frame is only for a diagram. So figure 405 and > 406 can't really be rendered in > a UML tool as they suggest a diagram having multiple frames. > > Another important issue is that the appendix says > Each diagram has a frame, a contents area, and a heading, see Figure 460. > This is clearly an optional notation and should be presented as such > I will log an issue on that. > > Eran > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Mon, March 08, 2004 4:18 AM > To: Eran Gery > Cc: uml2-superstructure-ftf@omg.org > Subject: RE: Ballot 9 issue 6465 > > Hi Eran, > > My answers are embedded below: > > > > Bran - Is the notation used in figured 406 (and 405) defined > > somewhere ? I refer to the package "frames" and > > notes for defining local attributes - ? what is it based on ? I know > > these frames are used in sequence diagrams, > > but did we introduce them as a generic notational technique for > > namespacing ? Where ? > > Please look at Appendix B: Diagrams. Using frames is an optional way > of denoting a number of different things in UML 2.0. > > > In case of package nesting, I would use a package symbol instead the > > proposed frame notation. For class internal structure, we use a > > class symbol as the "frame" - so we need to be consistent with that. > > You can use either one. > > Cheers, > > Bran -- Birger Møller-Pedersen Professor Department of Informatics, University of Oslo P.O. Box 1080 Blindern N-0316 Oslo Norway Tel: +47 22 85 24 37 (office) Tel: +47 918 27 27 9 (mobile) http://folk.uio.no/birger To: Birger Møller-Pedersen Cc: Eran Gery , uml2-superstructure-ftf@omg.org Subject: Re: Ballot 9 issue 6465 X-Mailer: Lotus Notes Release 6.0.2CF1 June 9, 2003 From: Branislav Selic Date: Mon, 8 Mar 2004 09:35:25 -0500 X-MIMETrack: Serialize by Router on D25ML05/25/M/IBM(Release 6.0.2CF1|June 9, 2003) at 03/08/2004 09:35:26, Serialize complete at 03/08/2004 09:35:26 Do we use the old Soviet model: "only things that are explicitly allowed are legal" or the free-enterprise model: "only things that are expressly forbidden are illegal"? Since we don't have a formal syntax definition, it seems that the Soviet model cannot be used, because then only the example diagrams that are in the spec would be legal. So, until we define a formal notation syntax for UML, it is not really meaningful to say that some notation is improper just because there are no examples of it in the spec. Bran Birger Møller-Pedersen 03/08/2004 09:03 AM To Branislav Selic/Ottawa/IBM@IBMCA cc Eran Gery , uml2-superstructure-ftf@omg.org Subject Re: Ballot 9 issue 6465 Figure 405 is certainly an examle of nested diagrams. Figure 406 is not obviously. One may of course put a diagram around them all (or claim that there is an implicit frame - which is by the way also allowed according to the spec), but the the question is then if it is possible to have relations between diagrams. /birger Branislav Selic wrote: Eran, I am not sure which issue you want to log, but you seem to be making some assumptions about diagrams that may not be shared. Namely, for some reason, you are assuming that you cannot have nested diagrams. This is a possibility that Birger and I discussed and, as far as I am concerned, it is still an option. In fact, I think that it is absolutely necessary when one needs to show things such as relationships between two or more models. Our problem is that there is no formal syntax defined for the notation of UML, merely a set of examples. This is still an outstanding item. If you raise an issue against figures 406, I can tell you that I will propose either a defer or a close, no change resolution for it (since I own that part of the text). I don't mean to be contrarian but I really believe that diagrams such as that are necessary in some cases. Regards, Bran Eran Gery wrote on 03/08/2004 06:05:52 AM: > Bran > > First thanks for pointing me there... > Apologize for being a pain, but these notational issues are what end > users really care about... > > From what I saw, the frame is only for a diagram. So figure 405 and > 406 can't really be rendered in > a UML tool as they suggest a diagram having multiple frames. > > Another important issue is that the appendix says > Each diagram has a frame, a contents area, and a heading, see Figure 460. > This is clearly an optional notation and should be presented as such > I will log an issue on that. > > Eran > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Mon, March 08, 2004 4:18 AM > To: Eran Gery > Cc: uml2-superstructure-ftf@omg.org > Subject: RE: Ballot 9 issue 6465 > > Hi Eran, > > My answers are embedded below: > > > > Bran - Is the notation used in figured 406 (and 405) defined > > somewhere ? I refer to the package "frames" and > > notes for defining local attributes - ? what is it based on ? I know > > these frames are used in sequence diagrams, > > but did we introduce them as a generic notational technique for > > namespacing ? Where ? > > Please look at Appendix B: Diagrams. Using frames is an optional way > of denoting a number of different things in UML 2.0. > > > In case of package nesting, I would use a package symbol instead the > > proposed frame notation. For class internal structure, we use a > > class symbol as the "frame" - so we need to be consistent with that. > > You can use either one. > > Cheers, > > Bran -- Birger Møller-Pedersen Professor Department of Informatics, University of Oslo P.O. Box 1080 Blindern N-0316 Oslo Norway Tel: +47 22 85 24 37 (office) Tel: +47 918 27 27 9 (mobile) http://folk.uio.no/birger From: Eran Gery To: Branislav Selic Cc: uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 Date: Mon, 8 Mar 2004 16:50:21 +0200 X-Mailer: Internet Mail Service (5.5.2656.59) Bran - the issue is that the appendix implies that frames are mandatory. I want the text toclarify they are optional. -----Original Message----- From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Mon, March 08, 2004 3:19 PM To: Eran Gery Cc: uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 Eran, I am not sure which issue you want to log, but you seem to be making some assumptions about diagrams that may not be shared. Namely, for some reason, you are assuming that you cannot have nested diagrams. This is a possibility that Birger and I discussed and, as far as I am concerned, it is still an option. In fact, I think that it is absolutely necessary when one needs to show things such as relationships between two or more models. Our problem is that there is no formal syntax defined for the notation of UML, merely a set of examples. This is still an outstanding item. If you raise an issue against figures 406, I can tell you that I will propose either a defer or a close, no change resolution for it (since I own that part of the text). I don't mean to be contrarian but I really believe that diagrams such as that are necessary in some cases. Regards, Bran Eran Gery wrote on 03/08/2004 06:05:52 AM: > Bran > > First thanks for pointing me there... > Apologize for being a pain, but these notational issues are what end > users really care about... > > From what I saw, the frame is only for a diagram. So figure 405 and > 406 can't really be rendered in > a UML tool as they suggest a diagram having multiple frames. > > Another important issue is that the appendix says > Each diagram has a frame, a contents area, and a heading, see Figure 460. > This is clearly an optional notation and should be presented as such > I will log an issue on that. > > Eran > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Mon, March 08, 2004 4:18 AM > To: Eran Gery > Cc: uml2-superstructure-ftf@omg.org > Subject: RE: Ballot 9 issue 6465 > > Hi Eran, > > My answers are embedded below: > > > > Bran - Is the notation used in figured 406 (and 405) defined > > somewhere ? I refer to the package "frames" and > > notes for defining local attributes - ? what is it based on ? I know > > these frames are used in sequence diagrams, > > but did we introduce them as a generic notational technique for > > namespacing ? Where ? > > Please look at Appendix B: Diagrams. Using frames is an optional way > of denoting a number of different things in UML 2.0. > > > In case of package nesting, I would use a package symbol instead the > > proposed frame notation. For class internal structure, we use a > > class symbol as the "frame" - so we need to be consistent with that. > > You can use either one. > > Cheers, > > Bran Date: Mon, 08 Mar 2004 18:37:01 +0100 From: Birger Møller-Pedersen User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en To: Branislav Selic CC: Eran Gery , uml2-superstructure-ftf@omg.org Subject: Re: Ballot 9 issue 6465 X-MailScanner-Information: This message has been scanned for viruses/spam. Contact postmaster@uio.no if you have questions about this scanning X-UiO-MailScanner: No virus found X-UiO-Spam-info: not spam, SpamAssassin (score=1.219, required 12, HTML_20_30 0.47, HTML_FONTCOLOR_UNKNOWN 0.10, HTML_MESSAGE 0.00, HTML_TAG_EXISTS_TBODY 0.10, HTML_TITLE_EMPTY 0.54) X-UiO-Spam-score: s There is also the "vendor model" (or call it a "market model"): tool vendors will implement the part of the notation they think their customers want (and will buy). My only observation was that 406 is not obviously an example of nested diagrams, and gave an interpretation that made it into this (implicit enclsong diagram). What is obvious, however, is that unless these are nested diagrams, no tool will implement relations between diagrams, simply because tools usually have one diagram visible at a time. What if the three package diagrams were simply package symbols? Does the spec say something about the legality of relations across package borders? Must there be an import? This is not just a notation issue, but an issue of visibility acrss package boundaries. /birger Branislav Selic wrote: Do we use the old Soviet model: "only things that are explicitly allowed are legal" or the free-enterprise model: "only things that are expressly forbidden are illegal"? Since we don't have a formal syntax definition, it seems that the Soviet model cannot be used, because then only the example diagrams that are in the spec would be legal. So, until we define a formal notation syntax for UML, it is not really meaningful to say that some notation is improper just because there are no examples of it in the spec. Bran Birger Møller-Pedersen 03/08/2004 09:03 AM To Branislav Selic/Ottawa/IBM@IBMCA cc Eran Gery , uml2-superstructure-ftf@omg.org Subject Re: Ballot 9 issue 6465 Figure 405 is certainly an examle of nested diagrams. Figure 406 is not obviously. One may of course put a diagram around them all (or claim that there is an implicit frame - which is by the way also allowed according to the spec), but the the question is then if it is possible to have relations between diagrams. /birger Branislav Selic wrote: Eran, I am not sure which issue you want to log, but you seem to be making some assumptions about diagrams that may not be shared. Namely, for some reason, you are assuming that you cannot have nested diagrams. This is a possibility that Birger and I discussed and, as far as I am concerned, it is still an option. In fact, I think that it is absolutely necessary when one needs to show things such as relationships between two or more models. Our problem is that there is no formal syntax defined for the notation of UML, merely a set of examples. This is still an outstanding item. If you raise an issue against figures 406, I can tell you that I will propose either a defer or a close, no change resolution for it (since I own that part of the text). I don't mean to be contrarian but I really believe that diagrams such as that are necessary in some cases. Regards, Bran Eran Gery wrote on 03/08/2004 06:05:52 AM: > Bran > > First thanks for pointing me there... > Apologize for being a pain, but these notational issues are what end > users really care about... > > From what I saw, the frame is only for a diagram. So figure 405 and > 406 can't really be rendered in > a UML tool as they suggest a diagram having multiple frames. > > Another important issue is that the appendix says > Each diagram has a frame, a contents area, and a heading, see Figure 460. > This is clearly an optional notation and should be presented as such > I will log an issue on that. > > Eran > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Mon, March 08, 2004 4:18 AM > To: Eran Gery > Cc: uml2-superstructure-ftf@omg.org > Subject: RE: Ballot 9 issue 6465 > > Hi Eran, > > My answers are embedded below: > > > > Bran - Is the notation used in figured 406 (and 405) defined > > somewhere ? I refer to the package "frames" and > > notes for defining local attributes - ? what is it based on ? I know > > these frames are used in sequence diagrams, > > but did we introduce them as a generic notational technique for > > namespacing ? Where ? > > Please look at Appendix B: Diagrams. Using frames is an optional way > of denoting a number of different things in UML 2.0. > > > In case of package nesting, I would use a package symbol instead the > > proposed frame notation. For class internal structure, we use a > > class symbol as the "frame" - so we need to be consistent with that. > > You can use either one. > > Cheers, > > Bran From: Eran Gery To: Branislav Selic Cc: Birger Møller-Pedersen , uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 Date: Tue, 9 Mar 2004 11:22:52 +0200 X-Mailer: Internet Mail Service (5.5.2656.59) Bran - Why not replace 406 with package symbols and avoid the confusion ? Why pulling diagram nesting into this if not necessary ? I know someone needs to do this and its painful but if you have the original frame picture the transformation should not be too bad. -----Original Message----- From: Birger Møller-Pedersen [mailto:birger@ifi.uio.no] Sent: Mon, March 08, 2004 7:37 PM To: Branislav Selic Cc: Eran Gery; uml2-superstructure-ftf@omg.org Subject: Re: Ballot 9 issue 6465 There is also the "vendor model" (or call it a "market model"): tool vendors will implement the part of the notation they think their customers want (and will buy). My only observation was that 406 is not obviously an example of nested diagrams, and gave an interpretation that made it into this (implicit enclsong diagram). What is obvious, however, is that unless these are nested diagrams, no tool will implement relations between diagrams, simply because tools usually have one diagram visible at a time. What if the three package diagrams were simply package symbols? Does the spec say something about the legality of relations across package borders? Must there be an import? This is not just a notation issue, but an issue of visibility acrss package boundaries. /birger Branislav Selic wrote: Do we use the old Soviet model: "only things that are explicitly allowed are legal" or the free-enterprise model: "only things that are expressly forbidden are illegal"? Since we don't have a formal syntax definition, it seems that the Soviet model cannot be used, because then only the example diagrams that are in the spec would be legal. So, until we define a formal notation syntax for UML, it is not really meaningful to say that some notation is improper just because there are no examples of it in the spec. Bran Birger Møller-Pedersen 03/08/2004 09:03 AM To Branislav Selic/Ottawa/IBM@IBMCA cc Eran Gery , uml2-superstructure-ftf@omg.org Subject Re: Ballot 9 issue 6465 Figure 405 is certainly an examle of nested diagrams. Figure 406 is not obviously. One may of course put a diagram around them all (or claim that there is an implicit frame - which is by the way also allowed according to the spec), but the the question is then if it is possible to have relations between diagrams. /birger Branislav Selic wrote: Eran, I am not sure which issue you want to log, but you seem to be making some assumptions about diagrams that may not be shared. Namely, for some reason, you are assuming that you cannot have nested diagrams. This is a possibility that Birger and I discussed and, as far as I am concerned, it is still an option. In fact, I think that it is absolutely necessary when one needs to show things such as relationships between two or more models. Our problem is that there is no formal syntax defined for the notation of UML, merely a set of examples. This is still an outstanding item. If you raise an issue against figures 406, I can tell you that I will propose either a defer or a close, no change resolution for it (since I own that part of the text). I don't mean to be contrarian but I really believe that diagrams such as that are necessary in some cases. Regards, Bran Eran Gery wrote on 03/08/2004 06:05:52 AM: > Bran > > First thanks for pointing me there... > Apologize for being a pain, but these notational issues are what end > users really care about... > > From what I saw, the frame is only for a diagram. So figure 405 and > 406 can't really be rendered in > a UML tool as they suggest a diagram having multiple frames. > > Another important issue is that the appendix says > Each diagram has a frame, a contents area, and a heading, see Figure 460. > This is clearly an optional notation and should be presented as such > I will log an issue on that. > > Eran > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Mon, March 08, 2004 4:18 AM > To: Eran Gery > Cc: uml2-superstructure-ftf@omg.org > Subject: RE: Ballot 9 issue 6465 > > Hi Eran, > > My answers are embedded below: > > > > Bran - Is the notation used in figured 406 (and 405) defined > > somewhere ? I refer to the package "frames" and > > notes for defining local attributes - ? what is it based on ? I know > > these frames are used in sequence diagrams, > > but did we introduce them as a generic notational technique for > > namespacing ? Where ? > > Please look at Appendix B: Diagrams. Using frames is an optional way > of denoting a number of different things in UML 2.0. > > > In case of package nesting, I would use a package symbol instead the > > proposed frame notation. For class internal structure, we use a > > class symbol as the "frame" - so we need to be consistent with that. > > You can use either one. > > Cheers, > > Bran To: Eran Gery Cc: Birger Møller-Pedersen , uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 X-Mailer: Lotus Notes Release 6.0.2CF1 June 9, 2003 From: Branislav Selic Date: Tue, 9 Mar 2004 09:22:04 -0500 X-MIMETrack: Serialize by Router on D25ML05/25/M/IBM(Release 6.0.2CF1|June 9, 2003) at 03/09/2004 09:22:10, Serialize complete at 03/09/2004 09:22:10 Eran, I think that the original figure actually came from Gunnar and Karin a while ago. I am not at all worried about the effort involved in redrawing the figure. But, I actually like the fact that the figure illustrates the possibility that diagrams can be nested. The way to clarify the confusion is not to eliminate it from this place -- because it gives at least one example of this capability in the spec -- but to document this feature in the Diagrams appendix. (BTW, note that the diagram interchange spec does allow diagrams to be nested.) Can you please be more specific about your concerns? Do you think that it is not a good idea to nest diagrams? If so, why? Thanks, Bran Eran Gery 03/09/2004 04:22 AM To Branislav Selic/Ottawa/IBM@IBMCA cc Birger Møller-Pedersen , uml2-superstructure-ftf@omg.org Subject RE: Ballot 9 issue 6465 Bran - Why not replace 406 with package symbols and avoid the confusion ? Why pulling diagram nesting into this if not necessary ? I know someone needs to do this and its painful but if you have the original frame picture the transformation should not be too bad. -----Original Message----- From: Birger Møller-Pedersen [mailto:birger@ifi.uio.no] Sent: Mon, March 08, 2004 7:37 PM To: Branislav Selic Cc: Eran Gery; uml2-superstructure-ftf@omg.org Subject: Re: Ballot 9 issue 6465 There is also the "vendor model" (or call it a "market model"): tool vendors will implement the part of the notation they think their customers want (and will buy). My only observation was that 406 is not obviously an example of nested diagrams, and gave an interpretation that made it into this (implicit enclsong diagram). What is obvious, however, is that unless these are nested diagrams, no tool will implement relations between diagrams, simply because tools usually have one diagram visible at a time. What if the three package diagrams were simply package symbols? Does the spec say something about the legality of relations across package borders? Must there be an import? This is not just a notation issue, but an issue of visibility acrss package boundaries. /birger Branislav Selic wrote: Do we use the old Soviet model: "only things that are explicitly allowed are legal" or the free-enterprise model: "only things that are expressly forbidden are illegal"? Since we don't have a formal syntax definition, it seems that the Soviet model cannot be used, because then only the example diagrams that are in the spec would be legal. So, until we define a formal notation syntax for UML, it is not really meaningful to say that some notation is improper just because there are no examples of it in the spec. Bran Birger Møller-Pedersen 03/08/2004 09:03 AM To Branislav Selic/Ottawa/IBM@IBMCA cc Eran Gery , uml2-superstructure-ftf@omg.org Subject Re: Ballot 9 issue 6465 Figure 405 is certainly an examle of nested diagrams. Figure 406 is not obviously. One may of course put a diagram around them all (or claim that there is an implicit frame - which is by the way also allowed according to the spec), but the the question is then if it is possible to have relations between diagrams. /birger Branislav Selic wrote: Eran, I am not sure which issue you want to log, but you seem to be making some assumptions about diagrams that may not be shared. Namely, for some reason, you are assuming that you cannot have nested diagrams. This is a possibility that Birger and I discussed and, as far as I am concerned, it is still an option. In fact, I think that it is absolutely necessary when one needs to show things such as relationships between two or more models. Our problem is that there is no formal syntax defined for the notation of UML, merely a set of examples. This is still an outstanding item. If you raise an issue against figures 406, I can tell you that I will propose either a defer or a close, no change resolution for it (since I own that part of the text). I don't mean to be contrarian but I really believe that diagrams such as that are necessary in some cases. Regards, Bran Eran Gery wrote on 03/08/2004 06:05:52 AM: > Bran > > First thanks for pointing me there... > Apologize for being a pain, but these notational issues are what end > users really care about... > > From what I saw, the frame is only for a diagram. So figure 405 and > 406 can't really be rendered in > a UML tool as they suggest a diagram having multiple frames. > > Another important issue is that the appendix says > Each diagram has a frame, a contents area, and a heading, see Figure 460. > This is clearly an optional notation and should be presented as such > I will log an issue on that. > > Eran > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Mon, March 08, 2004 4:18 AM > To: Eran Gery > Cc: uml2-superstructure-ftf@omg.org > Subject: RE: Ballot 9 issue 6465 > > Hi Eran, > > My answers are embedded below: > > > > Bran - Is the notation used in figured 406 (and 405) defined > > somewhere ? I refer to the package "frames" and > > notes for defining local attributes - ? what is it based on ? I know > > these frames are used in sequence diagrams, > > but did we introduce them as a generic notational technique for > > namespacing ? Where ? > > Please look at Appendix B: Diagrams. Using frames is an optional way > of denoting a number of different things in UML 2.0. > > > In case of package nesting, I would use a package symbol instead the > > proposed frame notation. For class internal structure, we use a > > class symbol as the "frame" - so we need to be consistent with that. > > You can use either one. > > Cheers, > > Bran From: Eran Gery To: Branislav Selic Cc: Birger Møller-Pedersen , uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 Date: Wed, 10 Mar 2004 14:06:34 +0200 X-Mailer: Internet Mail Service (5.5.2656.59) Bran - I am not strongly against here, but since the focus of this section is not to deal with "advanced" notations, apprently it may only confise the readers. This issue of nested diagrams better be clarified in the appendix, IMO. But again, I am not hung up on this... -----Original Message----- From: Branislav Selic [mailto:bselic@ca.ibm.com] Sent: Tue, March 09, 2004 4:22 PM To: Eran Gery Cc: Birger Møller-Pedersen; uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 issue 6465 Eran, I think that the original figure actually came from Gunnar and Karin a while ago. I am not at all worried about the effort involved in redrawing the figure. But, I actually like the fact that the figure illustrates the possibility that diagrams can be nested. The way to clarify the confusion is not to eliminate it from this place -- because it gives at least one example of this capability in the spec -- but to document this feature in the Diagrams appendix. (BTW, note that the diagram interchange spec does allow diagrams to be nested.) Can you please be more specific about your concerns? Do you think that it is not a good idea to nest diagrams? If so, why? Thanks, Bran Eran Gery 03/09/2004 04:22 AM To Branislav Selic/Ottawa/IBM@IBMCA cc Birger Møller-Pedersen , uml2-superstructure-ftf@omg.org Subject RE: Ballot 9 issue 6465 Bran - Why not replace 406 with package symbols and avoid the confusion ? Why pulling diagram nesting into this if not necessary ? I know someone needs to do this and its painful but if you have the original frame picture the transformation should not be too bad. -----Original Message----- From: Birger Møller-Pedersen [mailto:birger@ifi.uio.no] Sent: Mon, March 08, 2004 7:37 PM To: Branislav Selic Cc: Eran Gery; uml2-superstructure-ftf@omg.org Subject: Re: Ballot 9 issue 6465 There is also the "vendor model" (or call it a "market model"): tool vendors will implement the part of the notation they think their customers want (and will buy). My only observation was that 406 is not obviously an example of nested diagrams, and gave an interpretation that made it into this (implicit enclsong diagram). What is obvious, however, is that unless these are nested diagrams, no tool will implement relations between diagrams, simply because tools usually have one diagram visible at a time. What if the three package diagrams were simply package symbols? Does the spec say something about the legality of relations across package borders? Must there be an import? This is not just a notation issue, but an issue of visibility acrss package boundaries. /birger Branislav Selic wrote: Do we use the old Soviet model: "only things that are explicitly allowed are legal" or the free-enterprise model: "only things that are expressly forbidden are illegal"? Since we don't have a formal syntax definition, it seems that the Soviet model cannot be used, because then only the example diagrams that are in the spec would be legal. So, until we define a formal notation syntax for UML, it is not really meaningful to say that some notation is improper just because there are no examples of it in the spec. Bran Birger Møller-Pedersen 03/08/2004 09:03 AM To Branislav Selic/Ottawa/IBM@IBMCA cc Eran Gery , uml2-superstructure-ftf@omg.org Subject Re: Ballot 9 issue 6465 Figure 405 is certainly an examle of nested diagrams. Figure 406 is not obviously. One may of course put a diagram around them all (or claim that there is an implicit frame - which is by the way also allowed according to the spec), but the the question is then if it is possible to have relations between diagrams. /birger Branislav Selic wrote: Eran, I am not sure which issue you want to log, but you seem to be making some assumptions about diagrams that may not be shared. Namely, for some reason, you are assuming that you cannot have nested diagrams. This is a possibility that Birger and I discussed and, as far as I am concerned, it is still an option. In fact, I think that it is absolutely necessary when one needs to show things such as relationships between two or more models. Our problem is that there is no formal syntax defined for the notation of UML, merely a set of examples. This is still an outstanding item. If you raise an issue against figures 406, I can tell you that I will propose either a defer or a close, no change resolution for it (since I own that part of the text). I don't mean to be contrarian but I really believe that diagrams such as that are necessary in some cases. Regards, Bran Eran Gery wrote on 03/08/2004 06:05:52 AM: > Bran > > First thanks for pointing me there... > Apologize for being a pain, but these notational issues are what end > users really care about... > > >From what I saw, the frame is only for a diagram. So figure 405 and > 406 can't really be rendered in > a UML tool as they suggest a diagram having multiple frames. > > Another important issue is that the appendix says > Each diagram has a frame, a contents area, and a heading, see Figure 460. > This is clearly an optional notation and should be presented as such > I will log an issue on that. > > Eran > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Mon, March 08, 2004 4:18 AM > To: Eran Gery > Cc: uml2-superstructure-ftf@omg.org > Subject: RE: Ballot 9 issue 6465 > > Hi Eran, > > My answers are embedded below: > > > > Bran - Is the notation used in figured 406 (and 405) defined > > somewhere ? I refer to the package "frames" and > > notes for defining local attributes - ? what is it based on ? I know > > these frames are used in sequence diagrams, > > but did we introduce them as a generic notational technique for > > namespacing ? Where ? > > Please look at Appendix B: Diagrams. Using frames is an optional way > of denoting a number of different things in UML 2.0. > > > In case of package nesting, I would use a package symbol instead the > > proposed frame notation. For class internal structure, we use a > > class symbol as the "frame" - so we need to be consistent with that. > > You can use either one. > > Cheers, > > Bran Reply-To: From: "Cris Kobryn" To: "'Branislav Selic'" , Cc: Subject: RE: Ballot 9 -- vote starts today at 6 pm EST Date: Mon, 15 Mar 2004 15:20:32 -0800 X-Mailer: Microsoft Office Outlook, Build 11.0.5510 Thread-Index: AcQBe5N4eUfxG6IsSKGhVOb6hYsYigJZZKCg With the exception of Issue# 6465, which contains an inconsistent use of keywords in Figure 406, Telelogic votes YES on the block of resolutions for ballot 9. Telelogic votes YES on Issue# 6465 if the inconsistent usage of both "<>" and "<>" keywords is fixed (both should use "<>"); otherwise it votes NO. (We believe this editorial correction can be made without revoting on the issue, since the intent is clear and the mistake is patent.) -- Cris > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Wednesday, March 03, 2004 3:45 PM > To: uml2-superstructure-ftf@omg.org > Cc: mu2i-ftf@omg.org > Subject: Ballot 9 -- vote starts today at 6 pm EST > > > You have two weeks to vote (and re-vote, if you want) on these proposed > resolutions. > > > > Bran To: Cc: mu2i-ftf@omg.org, uml2-superstructure-ftf@omg.org Subject: RE: Ballot 9 -- vote starts today at 6 pm EST X-Mailer: Lotus Notes Release 6.0.2CF1 June 9, 2003 From: Branislav Selic Date: Mon, 15 Mar 2004 18:33:55 -0500 X-MIMETrack: Serialize by Router on D25ML05/25/M/IBM(Release 6.0.2CF1|June 9, 2003) at 03/15/2004 18:33:59, Serialize complete at 03/15/2004 18:33:59 To all: Cris has pointed out that there is still one overlooked case of the use of <> in the proposed resolution for issue 6465 (the one between Withdraw and CardIdentification use cases) -- all the other cases have been corrected. No one spotted this before (thanks, Cris). Unless somone objects, I will assume that everyone agrees with his conclusion and that the change can be made when the text is inserted into the spec without requiring a revote. Regards, Bran "Cris Kobryn" 03/15/2004 06:20 PM Please respond to cris.kobryn To Branislav Selic/Ottawa/IBM@IBMCA, cc Subject RE: Ballot 9 -- vote starts today at 6 pm EST With the exception of Issue# 6465, which contains an inconsistent use of keywords in Figure 406, Telelogic votes YES on the block of resolutions for ballot 9. Telelogic votes YES on Issue# 6465 if the inconsistent usage of both "<>" and "<>" keywords is fixed (both should use "<>"); otherwise it votes NO. (We believe this editorial correction can be made without revoting on the issue, since the intent is clear and the mistake is patent.) -- Cris > -----Original Message----- > From: Branislav Selic [mailto:bselic@ca.ibm.com] > Sent: Wednesday, March 03, 2004 3:45 PM > To: uml2-superstructure-ftf@omg.org > Cc: mu2i-ftf@omg.org > Subject: Ballot 9 -- vote starts today at 6 pm EST > > > You have two weeks to vote (and re-vote, if you want) on these proposed > resolutions. > > > > Bran