W3C home > Mailing lists > Public > public-rif-wg@w3.org > March 2006

Re: [UCR] alternative Use case publication

From: Axel Polleres <axel.polleres@uibk.ac.at>
Date: Tue, 21 Mar 2006 15:19:08 +0100
Message-ID: <44200B5C.1030909@uibk.ac.at>
To: Dave Reynolds <der@hplb.hpl.hp.com>
CC: axel@polleres.net, public-rif-wg@w3.org

Dave Reynolds wrote:
> Axel Polleres wrote:
>> I just added an alternative use case concerning "publication" on the 
>> Wiki:
>>  http://www.w3.org/2005/rules/wg/wiki/UCR/PublicationAlternative
>>  It shall empasize the use of RIF for defining implicit metatdata and 
>> linking metadata from differen sources by rules.
> Now that the f2f2 minutes have been published I can see that this [1] is 
> supposed to be a replacement for the vanished use case on Publication.
> On the one hand this does seem to be a perfectly good use case and I'd 
> be happy to see some improved [*] version of it in the second UCR draft.
> On the other hand I don't think it is a complete replacement. The 
> essence of the original publication use case [2] was the additional 
> requirement for a human legible syntax that could be used to enable 
> publication of rules for human developers. That was put up as a strawman 
> to encourage us to explicitly consider whether a legible syntax 
> (alongside the main XML or RDF syntax) should be within scope. Axel's 
> new use case has no such requirement. The alleged requirement for human 
> legibility is not quite the same as the human orientedness called for in 
> UC 2.5.
> Dave
> [*] Some comments towards such a tidy up:
> - if those data examples are to be included they should be RDF
> - the movie/sciFiMovie rule would be better expressed using rdfs:subClassOf
> - either delete or rephrase the value-judgement-full statement:
> [[[
> Recursive dependencies among the interlinked rulesets greatly extend and 
> generalize the simple one-way RDF data access via SPARQL.
> ]]]
> Sparql can explicitly query multiple datasets and can implement the 
> alternativeimdb examples.

I don't see how you can express recursive (closure) queries in  SPARQL. 
That was mainly what I meant to say here. SPARQL is meant to query fact 
bases and not interlinked rule bases with intensional, linked and 
possibly recursive view definitions. I assume this is also what you 
meant to express with [2].

  Do you have a concrete suggestion how to rephrase this or is it a rule 
with such closure definition that you miss here and would you have a 
suggestion for that?


> [1] Which is now at: 
> http://www.w3.org/2005/rules/wg/wiki/UCR/Candidate_Use_Cases_for_2nd_Draft/PublicationAlternative 
> [2] 
> http://www.w3.org/2005/rules/wg/wiki/Publication_of_semantics_%28e.g._SKOS%2C_RDFS%29 

Dr. Axel Polleres
email: axel@polleres.net  url: http://www.polleres.net/
Received on Tuesday, 21 March 2006 14:19:22 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:47:37 UTC