- From: Changhai Ke <cke@ilog.fr>
- Date: Wed, 12 Nov 2008 19:16:24 +0100
- To: "Christian de Sainte Marie" <csma@ilog.fr>, "Sandro Hawke" <sandro@w3.org>
- Cc: "RIF WG" <public-rif-wg@w3.org>
Christian, I hope we can reserve an opportunity to discuss about the precise names for the tags. Your main point here is that the "so called keyword" is in fact a value for an XML tag, and this is very good. My proposal for the name (for the value of the tag) is to call it "forward chaining" or "inference". (But please avoid "standard", it may give arise to additional thoughts). Changhai -----Original Message----- From: public-rif-wg-request@w3.org [mailto:public-rif-wg-request@w3.org] On Behalf Of Christian de Sainte Marie Sent: Wednesday, November 12, 2008 1:47 PM To: Sandro Hawke Cc: RIF WG Subject: Re: [Admin] Agenda for RIF telecon 11 November All, Sorry, my cellphone was off, as usual, yesterday. Here below, my comments re the two issues raised by Sandro re the proposed resolution (btw, I am fine with the actual resolutions :-). The first point is my use of the term "keyword". Maybe poor choice; blame my poor command of the english language. I mean something like: reserved value or whatever. My understanding is that the indication of the intended conflict resolution strategy cannot be an attribute, for extensibility reasons; so, it has to be an element. My current proposal [1] is to extend the syntax for group as follows: <Group> <behavior> <ConflictResolution> xsd:anyURI </ConflictResolution>? <Priority> -10,000 =< xsd:int =< 10,000 </Priority>? </behavior>? <sentence> [ RULE | Group ] </sentence>* </Group> where a specific value (proposed: rif:standardForward) would be reserved to indicate that standard forward chaining strategy we propose. Hence the term "keyword" (because "rifstandardForward" would be a keyword indicating that strategy, wouldn't it? Or does "keyword" mean something completely different from what I think?). Same for the second point (I mean: same blame): yes, "not being a conformance point" was meant to mean "suggested, not required for conformant implementation". Cheers, Christian
Received on Wednesday, 12 November 2008 18:17:10 UTC