They would be in RDF, as well. A URI has no more or less inherent meaning than an alpha-numeric code. They are both symbols for a more sophisticated concept. In Bibframe, Finnish data would not have triples with Finnish predicates, just objects (no different than MARC).
The argument for using numeric/letter codes to define the data model is a canard, frankly. If, say, the URI ends in an English term, that is no different to a non-English speaker than 260$e is to a cataloger today. A text based URI is no more ambiguous; in either case clarity is the result of how well the rules/properties are defined, not how they're named.
While one could argue that it sets those who speak a particular language at an advantage over others, I think it's safe to say that ship has already sailed: HTML, every XML and RDF schema I've ever used, and programming languages have all been rooted in English. To insist on language neutrality would just be another example of libraries ignoring obvious ways to fit into the larger ecosystems.
On Wednesday, May 29, 2013, Dana Pearson wrote:
still new to this list but do find it fascinating...Mac's point...
>I do like talking in unambiguous MARC field and subfield codes! The
ambiguity of language, and reuse of terms with new meanings, creates
seems well taken...I was working on a project this spring with a number of languages that I'm not familiar with...remember looking at MARC records in Finland and the Netherlands to help understand the MARC records I had created from another metadata format...the focus was 245 titles both parallel and multiple...since I know 245 subfields, my questions were easily answered..