661 2004-04-13 09:41:53 +0000 Adding partnerType and roleType to the language model 2004-12-01 17:16:02 +0000 1 1 1 Unclassified WS Choreography Spec: Naming and terminology unspecified Other other RESOLVED WONTFIX http://lists.w3.org/Archives/Public/public-ws-chor/2004Apr/0005.html P2 normal -- 1 gritzinger gritzinger oldest_to_newest 1674 0 gritzinger 2004-04-13 09:41:53 +0000 Back in February (16 Feb 04) I made a contribution to the group that had two main parts (http://lists.w3.org/Archives/Public/public-ws- chor/2004Feb/0024.html). The first part suggested adding partnerType and roleType to the language model and hence to the language. Maybe this has been discussed at a meeting when I was not present, but I have not been notified of any discussion or proposed resolution. I would hope to see these two elements added to the language or else good reason given why this is not appropriate (which may well be the case). So I would like this to be added to the list of issues, or a pointer provided to the resolution and rationale 2163 1 martin.chapman 2004-08-24 16:18:29 +0000 resolved at the July 04 f2f - partner becomes participanttype and role becomes roletype. 2575 2 gritzinger 2004-10-28 15:06:44 +0000 We currently seem to have 'fallen between the stools' with regard to participantType and roleType with a participantType being defined as consisting of one or more roles (not roleTypes) - section 2.3.2. My original idea was to define participantType(s) as consisting of one or more roleTypes, and to define roleTypes as they currently are, but then to instantiate / specialise participants from the participantTypes. Participants would consist of specific roles (which would be of a roleType). Thus you could specify a generic distributor participantType (perhaps having 3 'standardised' roleTypes of 'talkToBuyer', 'talkToSupplier' and 'talkToShipper') and then make specific participants DistributorA, Walmartdistribution, or, if we support instances distributor[N]. If the idea of having participantTypes and then having 'instances' of those types is now considered a step to far for version 1 CDL then I suggest that we go with participant and role and be very careful with the use of the word "type" in the sentences that contain these words. When the group decides on the principles here I will offer to propose precise text changes. Best Regards Tony A M Fletcher