- From: George Stoilos <gstoil@image.ntua.gr>
- Date: Fri, 24 Feb 2006 17:38:59 +0200
- To: "Public-Rif-Wg \(E-mail\)" <public-rif-wg@w3.org>
- Cc: "Giorgos Stamou \(E-mail\)" <gstam@softlab.ntua.gr>
- Message-ID: <002601c63958$6f31d580$760b6693@image.ece.ntua.gr>
Section 2.1(No) This UC needs some effort to make explicit the role of rule interchange (if possible). Section 2.2 (No) This section need revision in order to make the need for interchange or merging rulesets more explicit. The first scenario looks very similar to the one in the Charter, but the points where merging is mentioned (in the Charter) have been removed. Maybe these could strengthen the scenario. Rule interchage should also be mentioned in the second scenarion. For example the Hospital's Brain Reasoning System could integrate rules that come from heterogeneous rule bases of different research labs, to enhance its results or even to aquire knowledge for a domain that it does not have. Section 2.3 (Yes) But need to be extended. Section 2.4 (Yes) I don't see what rules are being exchanged in this use case. I get the feeling that rules are executed localy (e.g. decide if you are going to send your credit card) and then some ground facts are transmitted (address, city, credit card) or a no answer. Section 2.5 (Yes) I agree with Jim that this UC is a bit hard to understand. I think that it should also be reduced in size. Section 2.6 (Yes) Section 2.7 (Yes) Section 2.8 (Yes, but..) I also (like Christian) believe that this use case is no different from UC 6 (Publication). They both speak of an ontology enhanced with rules, which add more semantics on properties. Doesn't the rule for the hasUncle property communicate the semantics of this property, as it is written in UC 6? G. Stoilos
Received on Friday, 24 February 2006 15:39:14 UTC