- From: Christian de Sainte Marie <csma@ilog.fr>
- Date: Mon, 25 Sep 2006 18:02:37 +0200
- To: RIF WG <public-rif-wg@w3.org>
All, (I hope this message completes action 96, that is, makes it clear why I find that some of the UC are not clear enough :-) Some of the UC are not clear enough in that they do not explain clearly: 1. why RIF would be useful and what the benefits would be, and/or 2. how RIF would be used and with what requirements. A touchstone for (1) is looking at what would happen without RIF: if the UC works fine without interchanging rules in another form, the benefits of interchange, and RIF, should be stated explicitely. An example were they are not so clear is UC 2 (negotiating eCommerce transactions through disclosure of buyer and seller policies and preferences; although it has been improved, the case for policy interchange remains rather weak, in my opinion). An example where it is not clear (to me) what rules are interchanged, between which parties and for what purpose is UC 9 (BPEL orchestration of rule-based Web services). UC 7 (interchanging rule extensions to OWL) does not really provide a scenario for rule interchange, either, but I understand that some UCs can focus on one single aspect (here, the requirement that RIF must be able to extend OWL; the case for interchange thus depends on the case for interchanging ontologies). Regarding (2), one issue on which many of the UC could be more helpful is the protocol of rule interchange: who initiate the interchange, what happens on the other side, e.g. if the party that receives/retrieves the rules cannot understand them, etc: again, not all the UCs need to be specific about such issues, but, particularly regarding error handling and/or "dialect" mismatch, it would be useful to have at least one UC be specific. Another aspect that could be usefully clarified in some of the use cases is the relationship between interchanged rules, vocabularies and datasets. As I proposed in a previous email, I wonder if we should not include a discussion of each of the UC that would make explicit the benefit of using RIF in the UC, how RIF is used (the protocal of interchange), and the requirements on RIF that the use case illustrate. And maybe a more general discussion of how the different UCs cover the field. Christian
Received on Monday, 25 September 2006 16:02:16 UTC