Re: comments on OCL



Date view Thread view Subject view Author view

Jos Warmer (J.Warmer@klasse.nl)
Wed, 15 Dec 1999 12:49:59 +0100


Dear pUML group members, I have joined this mailing list two days ago and have seen OCL and my name a number of times. I would like to explain the current situation and my role at this moment. This is a rather lengthy message, because I don't know what discussion have been going on in the pUML group before. At the OMG, I am the person responsible for OCL. As someone stated, my time is limited, and I cannot possibly do everything that could or should be done. Apart from my OCL work, I have to make a living by doing paid customer work. (Yes, it's a cold, hard world outside the universities (:-) The situation currently is: - I am responsible within the UML 1.4 RTF for the OCL section of the UML. I am visiting almost no OMG/RTF meetings, because both the time and the traveling expenses will be too high. The RTF works mostly by phone email and web. - I am working on a number of fixes for 1.4, based on several actual implementationsn of OCL. Current implementors that I am aware of are: - Dresden University of Technology - Cybernetic Intelligence GmbH. - Department of computer sciences, Munich University - Royal Dutch Navy The main work items at this moment are very mundane things like getting the OCL grammar correct (it isn't in 1.3) and clarifying / fixing ambiguities that these implementors have found. Another work item is to come up with an OCL metamodel, to ensure better integration of OCL with the UML metamodel. - All the OCL work that I do is unpaid sparetime work, therefore I have to set priorities. I have made time available in the coming two months, to make sure that the above work items are done for UML 1.4. - Although I am the official person responsible for OCL, it is not the intention that OCL is my personal language. Contributions from anyone around the world are much appreciated and an absolute must to further develop OCL. Concrete examples of contributions are e.g.: - The Amsterdam OCL Group (Steve Cook, Alan Wills, Bernard Rumpe, Anneke Kleppe, Richard Mitchell) has done a lot of concrete work. - I am working with Martin Gogolla and Mark Richters to come up with an OCL metamodel. - The implementors above, that have given much feedback on OCL 1.3 - Many persons sending issues and solutions to the OMG/UML RTF. Other areas where contributions are usefull are: - A formal foundation of OCL. This is _very_ welcome, but I won't do this myself. - Usefull extensions to make OCL more useable. - Fixing even more "bugs". - .... etc ... >> The problem the OMG currently has is that nobody feels >> >really responsible for OCL. Jos Warmer is no longer able to spend much >> >time on OCL. I do feel responsible, but the time thing is sometimes a problem. >> On this point, I can be a little clearer. The pUML group have >> been informally offered the role of managing the OCL part of the semantics >> by the OMG. The group has ageed that this is a role it would be prepared to >> play, provided that Jos (and the other members of the Amsterdam group) are >> happy. >> I am aware of this rumor..... >> If this is the case, then we would be keen to organise a workshop to >> discuss these issues in depth. In fact, I believe that Tony Clark and >> Stuart Kent are putting together a proposal for a workshop in Canterbury >> for early next year. >> Please, send me information about this workshop, I might be interested (:-) >> Perhaps such a workshop would be a useful next step towards resolving these >> type of issues? >This sounds like good news. I made a similar proposal to the OMG (having >a group of academic reserachers taking care of OCL and the UML >well-formedness conditions.) I agree that pUML appears to be the right >forum for this. > To be honest, I am not sure wether I support this position. The OMG is an organisation dealing with industry standards and therefore usability within the industry is the main priority. I think that academic researchers have different priorities. The proposal above might move OCL and the UML well-formedness conditions into the research area, where they shouldn't be. UML and OCL must be _practical_, even when formally not completely sound. (I know that this statement will hurt some of you deeply) Having said this, I would love to have OCL and UML formally completely correct. I think that it is vitally important to have support from the academic community to get a decent foundation for OCL and UML. I welcome collaboration with people from the academic community and hope to get contributions from them, which will improve the OCL/UML specification. A workshop, as mentioned above is certainly useful. It will be most effective if there is an agreed upon way in which workshop result can end up in the OCL specofication. In general I am happy to work together with anyone who is willing and able to contribute. The pUML group could play an important role here. >I am willing to help in this effort. We (the UML group at Dresden) can >contribute, for instance, tool support. Our OCL parser and type checker >is already quite stable and conforms to UML/OCL V1.3. We are planning to >make the tool publicly available soon. If you are interested, please >contact me (hussmann@inf.tu-dresden.de). > Yes I am !! If you can spare a copy for me, I would be very happy. >Best regards >Heinrich >-- >Prof. Dr. Heinrich Hussmann >Dresden University of Technology >Department of Computer Science I hope that this email gives some background information. I will be on the pUML mailing list from now on and participate in the discussions. Regards, Jos _____________________________________________________ Klasse Objecten tel : +31 (0)35 6037646 Chalonhof 153 fax : +31 (0)35 6037647 3762 CT Soest email : J.Warmer@klasse.nl The Netherlands internet: http://www.klasse.nl


Date view Thread view Subject view Author view