One of the nice things about RDF is that it isn't record based. That means you can map any source of information into a common RDF model and then merge the results into a coherent dataset. This admittedly get harder when you have incompatible RDF models competing with each other. If "the shape of the data" is the same, it's not too much of a problem because OWL can pull the entities back together. If the shapes are too different, though, it ends up being a mess. That's one of the reasons some of us are pushing for vocabulary reuse and why it is important to model vocabularies as intuitively and realistically as possible.
Jeff
Sent from my iPad
On May 26, 2013, at 2:07 AM, "J. McRee Elrod" <[log in to unmask]> wrote:
> Jeff Young said:
>
>> Almost all crosswalks are a dumb down.
>
> We have sworn off both Dublin Core and ONIX. MARC records so produced
> are really poor. But the AACR2 compatible records we produce from
> RDA, and the UKMARC records produce from MARC21, are both quite good,
> due to the high quality and consistency of the source records.
>
> Both RDA and Bibframe over estimate the ability of many libraries to
> afford the system changes they require. Libraries which must have
> UKMARC are growing fewer.
>
>
> __ __ J. McRee (Mac) Elrod ([log in to unmask])
> {__ | / Special Libraries Cataloguing HTTP://www.slc.bc.ca/
> ___} |__ \__________________________________________________________
>
|