On Fri, Nov 8, 2013 at 12:46 PM, Shlomo Sanders <[log in to unmask]> wrote:
I would love to hear people's thoughts on a few questions:
 1. How mature do you think the  BIBFRAME model will be 1 year from today?

Bibframe at the moment lacks certain areas (such as serials, articles, research data sets, continuing resources, holdings, acquisitions, services) and it does not offer many new opportunities yet (for example, integrating e-journals, e-books into electronic supply chains and library online services). I hope this will change in the next years when Bibframe will be adopted by the national libraries and the available Bibframe data collections will grow. The current focus is on "we know how we did it in MARC, now can we do the same thing in Bibframe?". There are several trends Bibframe may follow: tighter integration with existing ILS technologies, or substantial contributions to new Linked Data platforms.
 2. Do you think it is practice to use BIBFRAME as RDF as the interchange format (instead of MARC)?

Bibframe was never intended as an interchange format and I do not think we need another interchange format for doing the same over and over again. With Linked Data, the real question is: what data can be linked online over the web, and what has still to be moved over the wire from library to library? How can we build new forms of union catalogs? How can libraries cooperate with interested communities (publishers, authors, museums, archives, research organizations) to exchange and reuse data sets? How can we establish improved library services by using Bibframe?
 3. How do you feel about migration of all your existing BIBs or at least migration on the fly of
    MARC records being updated and of MARC records ingested from external sources?

Bibframe's Linked Data model is more powerful than MARC. For traditional applications, MARC may be wrapped into Bibframe in the near future. But nobody wants to wrap old data in new formats just for fun. It depends on decisions, what new services must be implemented and what has to be changed. If catalogers still want to view MARC record field codes for efficient cataloging, there is not much justification for implementing Bibframe in a new cataloging module. But if publishers or authors will offer more Linked Data sets for their publications getting integrated into library catalogs, there is no good answer by using MARC. So, if new services with Linked Data have to be built, Bibframe may be an option, and it's not always a question of migration.
4. How do you feel about doing original cataloging in the BIBFRAME model?

I think RDA is for cataloging, not Bibframe? It's possible that librarians want to understand how to decipher various Bibframe data sets, dereference URIs to URLs, and follow links on the web. Most Bibframe modeling details will be hidden behind web UIs and mouse clicks.