Print

Print


Hi Ray,

On Thu, Jul 24, 2014 at 1:27 PM, Denenberg, Ray <[log in to unmask]> wrote:

> From: Robert Sanderson
> > <http://linked-data.stanford.edu/titles/books/1234> a bf:Title ;
> >   bf:value "Lord of the Rings" ;
> >   bf:identifier [
> >     a bf:Identifier ;
> >     bf:assigner "Rob" ;
> >     bf:identifierValue "
> http://linked-data.stanford.edu/titles/books/1234"  ] .
> Rob - I'm trying to work my way through all this, and right now trying to
> figure out your  example and if it is legitimate.
> In his example, the identifier is for a person (bf:Person, which is an
> Authority, which is a core class). There is an identifier scheme (viaf), a
> value within that scheme, and the identifierValueURI simply turns that
> value into a URI. (And if you prefer to put the URI in quotes and treat is
> as a literal, fine.)
>

Yup.  The proposal from Simeon was that __for URIs__ there still needed to
be a way to express additional provenance about them, such as who assigned
them.

I personally think that would be out of scope, and as an example
demonstrated that if we started down that path, there would be all sorts of
strange things you might want to give the provenance for ... such as the
URI for a title.

One might want to say that the URI was assigned by some particular piece of
machine learning based software that automatically reconciles resources.
 Or the URI was assigned at a particular point in time, so that you could
then go back and look at the identities assigned within a given time frame.
 Or according to some set of best practice guidelines. Or ...

Basically, assigning metadata about URIs is a can of worms that, again in
my opinion, we do not want to open.


But Rob in your example the identifier is for a Title. (I don't think we've
> thought much about providing identifiers for titles.)  It would help if you
> provided a scheme (whatever it is) and an identifier within that scheme, to
> help make sense of the example.   (And it's quite possible that I'm missing
> the point. Wouldn't be the first time.)
>

The scheme would be "http://linked-data.stanford.edu/titles/books/" and the
identifier would be 1234 ... but that's not the point I was trying to make.
Simeon was wanting to include statements about URIs via reifying the URI as
a string, within bf:Identifier.  I agree that it's possible, I disagree
that we want to do that :)

Rob

-- 
Rob Sanderson
Technology Collaboration Facilitator
Digital Library Systems and Services
Stanford, CA 94305