Re: [office-metadata] Question about the IRI of a RDF subject from an ODFdocument

[Elias Torres <eliast@us.ibm.com>]

"Bruce D'Arcus" <bruce.darcus@OpenDocument.us> wrote on 02/06/2007 07:52:48
AM:

> Hi Elias,
>
> Thanks, and good to see you back (even if bruised)!
>
> On Feb 5, 2007, at 11:43 PM, Elias Torres wrote:
>
> > However, we need more than that, because our package doesn't have a
> > stable
> > location. Hence, I would suggest we ground it on some URN, like LSID
> > for
> > example or whatever.
> >
> > urn:lsid:www.oasis-open.org:SOME_PACKAGE:SOME_UUID/eliast.jpg
>
> So you are suggesting that all documents get a global base URI, and
> that we use that to create stable URIs for embedded files (like
> images).
>
> Right?

Kind of. I was more focused on creating stable URIs for embedded files and
just mentioned xml:base as a mechanism to make it simpler, shorthand.

>
> How would you suggest we specify that? Would it be enough to say that
> the document requires a global uri that is stable and unique, and
> suggest your lsid as a good option to achieve that?
>
> Am thinking in some cases, an organization or user might want to
> specify their own http URIs as the base URI.
>
> > Now if we use RDF/XML we can make use of
> > xml:base="urn:lsid:www.oasis-open.org:SOME_PACKAGE:SOME_UUID/".
> >
> > <eliast.jpg> dc:title "Elias' mugshot" .
> >
> > For the case that the same exact picture lived on Flickr, then we can
> > use
> > owl:sameAs:
> >
> > <eliast.jpg> owl:sameAs
> > <http://www.flickr.com/photos/images/holidayimage.png> .
>
> And what do you think of the idea I floated earlier to use the manifest?
>
>   <manifest:file-entry
>       manifest:uri="http://flickr.com/photos/doe/1";
>       manifest:full-path="Pictures/foo.jpg."/>

This is the equivalent N3, right?

[
  rdf:type manifest:file-entry ;
  manifest:uri "http://flickr.com/photos/doe/1"; ;
  manifest:full-path "Pictures/foo.jpg"
]

I think it's ok, but it's quite verbose, 3 triples, to do what 1 triple
using owl:sameAs does.

>
> In that case, if a uri is present, then that is used as the subject.
> Otherwise, one adds the path to the base uri.

I guess what I'm after is trying to solve the problem of "relative URIs"
because the string (not a URI) "Pictures/foo.jpg" is only valid within this
package. If we were able to generate globally unique URIs then we could
make use of the statements (knowledge) made in any ODF also outside of it.
In other words, I'm trying to reduce the amount of special processing
within ODF spec before the metadata model can be useful in third-party RDF
tools.

>
> I have no opinion on which is better; just asking.
>
> Bruce
>