[Koha-devel] RDF Dereferencing/tracing/following links

[David Cook]

Hi all,

 

The key to Linked Data is obviously the links, but I don’t think any of us
quite know how the links actually work.

 

In theory, I imagine having a record (ie a set of triples), in a local
triplestore, which links to records available on the Internet. Periodically,
those records are downloaded, and stored in a local triplestore as a cache.
Then we somehow follow the links as they are stored in our local triplestore
and have a set of string literals which we can show to users on a web page.

 

In practice, I don’t know how that would work. I can see us getting a list
of record subjects in our local triplestore, and then iterating through each
one to download the links in its objects. But I don’t know where we’d put
that content. I also don’t know if we’d recurse into that content to follow
more links or not. 

 

Let’s say that we store that content in named graph called “cache”. When it
comes time to re-trace the links for our list of record subjects, how do we
know what local content to delete? In theory, we could use the URI from the
object, but many records contain many subjects, so even if we did delete
that subject from “cache”, we haven’t deleted the other subjects that it
links to internally in the triplestore.

 

I’m going to ask around to see if anyone can enlighten me. I’m hoping Petter
might be able to shed light on how Oslo Public Library makes it work,
especially in the context of ls.ext interfacing with Koha!

 

David Cook

Systems Librarian

Prosentient Systems

72/330 Wattle St

Ultimo, NSW 2007

Australia

 

Office: 02 9212 0899

Direct: 02 8005 0595

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.koha-community.org/pipermail/koha-devel/attachments/20170413/7d83830b/attachment.html>