- From: Victor Porton <porton@narod.ru>
- Date: Wed, 09 Jul 2014 23:17:15 +0300
- To: Ruben Verborgh <ruben.verborgh@ugent.be>
- Cc: SW-forum Web <semantic-web@w3.org>
09.07.2014, 23:13, "Ruben Verborgh" <ruben.verborgh@ugent.be>: >>> So we have the situation (I presume) where a.ttl and b.ttl are valid, >>> but where the combination of both is not? >> Yes, see the above example. >>> Are a.ttl and b.ttl then universally valid, or just in a certain context? >> I don't understand this question. >> >> I am writing a specification to which they should conform. > > What I meant is: does a.ttl contain triples that are > valid at any time and any given context, > or is the information place- or time-bound? They are meant to be valid at any time and any given context. But nothing prevents users of my system to create or modify the files dependently on date. > For instance, the triples: > - Barack_Obama is CurrentPresidentOfUSA > - Bill_Clinton is CurrentPresidentOfUSA > clearly conflict if you put them together; > that's because the first one is true in 2014, > the second one is true in another context. No, not related to change of things and moreover not related to temporal changes. > But I understand correctly what you mean, the problem might be in the modeling. > > So given the following: >> <http://example.org/the-transformation> >> a :transformer ; >> dc:description <http://...> ; >> # Other Dublin Core metadata. >> :source-namespace <...> ; >> :target-namespace <...> ; >> :precedence <...> ; > > Do you mean to say that: > a) "X is a transformation"? > b) "you should apply the transformation X with those and those parameters"? With those and those parameters and in those and those situations (such as if the source namespace happen to be the same as the value of :source-namespace predicates). -- Victor Porton - http://portonvictor.org
Received on Wednesday, 9 July 2014 20:17:45 UTC