Print

Print


On Thu, Jul 31, 2014 at 3:06 PM, Ford, Kevin <[log in to unmask]> wrote:

>
> > I'm less in favor of publisherStatement to transcribe and then repeat the
> > same information in somewhat of a jumbled fashion with repeating
> > publishedAt/By/On.  If there's the possibility of multiples, as
> demonstrated,
> > then the information shouldn't get lost as to which place is associated
> with
> > which organization, IMO.
>
> I'm not a fan of the repetition either, but RDA often requires
> /transcription/


Then I retract my event based structure :)


I also have fantasies that - down the road apiece - a cataloger would be
> able to type in a publicationStatement into a text field, at which point
> background programming would perform some kind of entity recognition and
> populate the proposed fields without the cataloger having to do double the
> work.  That doesn't get around the inherent duplication of data, but it
> mitigates the effort that produced it.
>

+1


> Here the structure isn't imposed just for the sake of having structure,
> it's to
> > model the publication event and its participants.
>
> This seems complicated (more so, in fact) and returns us, more or less,
> back to where it is now, which is to say a mostly non-reusable resource.
>  Also, those two wasAssociatedWiths would have to remain in the order in
> which they appeared on the source.  You can appreciate the headache that
> introduces in RDF-land.
>

:S  Indeed, if that's important, then fooStatement for all foo properties
that require such is (sad face) certainly an easy route to take.


> I liked your earlier question to Karen about what it all meant.  Are the
> publishers (in two different locations) working together to produce the
> /same thing/ or are we looking at two manifestations, each published by one
> of the indicated publishers in that particular year.  The latter would make
> things a lot easier, as you noted, and it is how we've interpreted that
> construct, but the documentation is vague on this point.
>

The http://lccn.loc.gov/99040733/marcxml case (or any repeating subfields,
really) is also tricky to get right automatically.  The second $b would
need to become
  _:instance1 bf:distributedBy [ a bf:Organization ; bf:label "Random
House" ]

Rob