Print

Print


On Nov 9, 2014 10:45 AM, "[log in to unmask]" <[log in to unmask]> wrote:

>This is especially so if the Bibframe predicate vocabulary remains committed to the particular Bibframe "two-level" model.

The sound you hear is a nail being struck square on the head.

bibframe created a new conceptual model. The design of this model was not driven by theoretical concerns; nor was it designed as part of a broader linked data strategy.

The primary entities within the bibframe universe appear to be [descriptions].

The primary evaluation criteria appear to have been those dealing with ease of generation from MARC-XML data files. 

The primary use case appears to be the transfer of the contents of one or more bibliographic records between systems.

That being the case, then it might be better to replace the entire currently proposed solution with a literal translation of marc into RDF with the most minimal changes possible that comply with appendix M, and publish those results as the final outcome of the exercise, which would have "FOIA"* benefits.

Review of  published draft documents, minutes, and presentations would suffice to allow an external group to prepare a lessons learned .

Simon

* FOIA does not apply to these bits of the Library of Congress; records access is governed by 36 CFR 703.

36 CFR 703.5(b)(10) provides a deliberative process exception that is somewhat similar in effect to 5 U.S.C. § 552(b)(5). The text is different enough that the Library may have different scope for this privilege - Library determination of whether disclosure is "premature" would presumably be upheld unless arbitrary and capricious.
There is a policy of not disclosing records which are part of on-going reviews or other current projects ; it is only after this point that review under 703.5 is relevant.
This policy is possibly not effective against Sen. Roberts :-)

The rule is intended to follow the spirit of the FOIA, so some case law may apply *by analogy* (indeed, a few sections of the rule appear to be written to incorporate *some* of the case law explicitly).

http://www.law.cornell.edu/cfr/text/36/703.5