Can not use blank nodes at the start as they cause significant problems
later when trying to do deletes and such via update protocols in Fuseki.

Small models are built locally and sent as update requests to the Fuseki
server so methods that require access to do renames will not be efficient
as there is no efficient way to run them on the Fuseki server.

On Tue, May 22, 2018 at 2:55 PM, Martynas Jusevičius <marty...@atomgraph.com
> wrote:

> Why generate URIs at all in the beginning, can't you use blank nodes?
>
> Rewriting URIs is generally a bad idea in a Linked Data setting. Make one
> datasource canonical and let the other one deal with that. Or maybe you can
> configure your proxy in a way that hides the port number and you don't need
> the second version (just a guess).
>
> Also, you generate document URIs, and persons are not documents. Hash URIs
> would probably be best for persons.
>
> If you choose to ignore the above, this method might help you:
> https://jena.apache.org/documentation/javadoc/jena/org/apache/jena/util/
> ResourceUtils.html#renameResource-org.apache.jena.rdf.model.Resource-java.
> lang.String-
>
> On Tue, May 22, 2018 at 1:00 PM, Claude Warren <cla...@xenei.com> wrote:
>
> > I have what I think may  be a common problem and am looking for suggested
> > patterns and anti-patterns for a solution.
> >
> > For the sake of this example let's assume that the system described
> creates
> > FOAF records.
> >
> > == PROBLEM
> >
> > Backend:
> >
> > The backend system generates FOAF records but does not have any
> information
> > about where they will be stored/deployed.  So it generates records like
> >
> > <urn:mysystem:foaf1> a foaf:Person ;
> >     foaf:name "Jimmy Wales" ;
> >     foaf:mbox <mailto:jwa...@bomis.com> ;
> >     foaf:homepage <http://www.jimmywales.com> ;
> >     foaf:nick "Jimbo" ;
> >     foaf:depiction <http://www.jimmywales.com/aus_img_small.jpg> ;
> >     foaf:interest <http://www.wikimedia.org> ;
> >     foaf:knows <urn:mysystem:foaf2> .
> >
> > <urn:mysystem:foaf2> a foaf:Person ;
> >     foaf:name "Angela Beesley" .
> >
> >
> > This data is stored in a Fuseki based server.
> >
> > Frontend 1:
> >
> > The front end should take replace the <urn:mysystem> based URIs with http
> > based URIs that point to the frontend.  So assuming the frontend is at
> > http://frontend:8080 and has a method to return RDF in turtle format the
> > RDF should look like
> >
> >  <http://frontend:8080/foaf1> a foaf:Person ;
> >      foaf:name "Jimmy Wales" ;
> >      foaf:mbox <mailto:jwa...@bomis.com> ;
> >      foaf:homepage <http://www.jimmywales.com> ;
> >      foaf:nick "Jimbo" ;
> >      foaf:depiction <http://www.jimmywales.com/aus_img_small.jpg> ;
> >      foaf:interest <http://www.wikimedia.org> ;
> >      foaf:knows <http://frontend:8080/foaf2> .
> >
> >  <http://frontend:8080/foaf2> a foaf:Person ;
> >      foaf:name "Angela Beesley" .
> >
> > Frontend 2:
> >
> > There is a second frontend with a different URL. http://frontend2:8080,
> > frontend 1 and frontend 2.  Frontend 2 does not have access to frontend 1
> > (assume that there is a firewall that prohibits the access).   Frontend2
> > should produce RDF like:
> >
> >  <http://frontend2:8080/foaf1> a foaf:Person ;
> >      foaf:name "Jimmy Wales" ;
> >      foaf:mbox <mailto:jwa...@bomis.com> ;
> >      foaf:homepage <http://www.jimmywales.com> ;
> >      foaf:nick "Jimbo" ;
> >      foaf:depiction <http://www.jimmywales.com/aus_img_small.jpg> ;
> >      foaf:interest <http://www.wikimedia.org> ;
> >      foaf:knows <http://frontend2:8080/foaf2> .
> >
> >  <http://frontend2:8080/foaf2> a foaf:Person ;
> >      foaf:name "Angela Beesley" .
> >
> > == Question
> >
> > How can I setup a system that will automatically convert one URI to
> another
> > without storing multiple copies of the data (e.g. not multiple datasets).
> > I have thought about using owl:sameAs and am wondering if there is a
> > reasoner that will process it.
> >
> > Anyway, has anyone else come across this problem (I figure so) and does
> > anyone have a possible solution?
> >
> >
> >
> >
> > Thx,
> > Claude
> > --
> > I like: Like Like - The likeliest place on the web
> > <http://like-like.xenei.com>
> > LinkedIn: http://www.linkedin.com/in/claudewarren
> >
>



-- 
I like: Like Like - The likeliest place on the web
<http://like-like.xenei.com>
LinkedIn: http://www.linkedin.com/in/claudewarren

Reply via email to