In relation to the German, plan, which I think is "spot on," I want to
point out that the W3C Library Linked Data group's report  contained
a number of recommendations, and key of those was the creation of
identifiers for the "things" in library data. The emphasis on authority
data in the DNB's plan fits this quite well.
In addition to actual authority files, we should also be looking at
other lists of terms (geographic codes, organization codes, all of those
extensive lists of musical instruments and musical composition types).
To a large extent, these terms have been given identifiers for RDA. 
The MARC 00x codes have also been defined in RDF. 
Assuming our goal is to be able to use RDF as our data model, a
bibliographic description should be mostly made up of identified things
that together describe the resource. We might be further along than we
think due to efforts like DNB's, the RDA in RDF work, and LC's
authorities work. 
I think that the biggest gap right now is to formulate best practices
for the creation of URIs for the resources being described. That was
also a recommendation in the W3C LLD report. We have to assume that
bibliographic data will often be shared but could also be created
independently in different libraries. It would probably be a good idea
to have members of the library community using similar patterns for
resource identifiers even though they may mint those identifiers
independently. I also assume that most folks who will be creating
identifiers would like to have a set of best practices to lean on rather
than creating their own.
On 1/30/12 12:33 PM, Amanda Xu wrote:
> It sounds like a great plan. It definitely offered some insight on what to expect initially in the version of unified bibliographic framework in web-scale management that you've proposed.
> One of my concerns, which is also an urgent issue based on my observations at ALA Midwinter 2012, is how to justify ROIs for internal cataloging record maintenance and storage.
> Many of us working in the field for years understand the importance to separate the concerns:
> 1) indexing, search and retrieval;
> 2) maintenance and storage; and
> 3) data interchange and resource sharing;
> Yet many of us are also being challenged to tie everything we do in metadata services with the mission of the institution that we serve, e.g. user-centered teaching, learning, and research services for academic institutions.
> I'm just playing the devil's advocates here. How can we convince the library admin and leadership of the institution that extra work that we've contributed to ensure metadata standardization for data interchange and resource sharing across the boundaries of institutions, regions, and nations are absolutely essential to the mission of an academic institution?
> Having been an active participant in CONSER Core and DC, I understand the extra mile that we had to take if we were going to make authority and bibliographic records sharable in LC and OCLC WorldCat.
> In addition, one of the benefit assumptions that we can make today is that record hierarchical linking and display is going to be performed by automated means (e.g. best practice in cool URI design, linked data rating and validation schemes, pattern recognitions, and intelligent text analysis of bibs and authority) with some human revision. This might be time-saving effort for us all.
> However, such assumption can't be established and realized unless we implement data quality control and knowledgebase for data profiling, integrating, cleansing and refinement, etc. at universal bibliographic control level in distributed and collaborative fashion, similar to the governance, editorial and publishing structure of wikipedia, which often take more resource, high-level expertise, new tool development and time than before as far as I understand.
> I see this as top-down, bottom-up, and dynamic approaches for data quality control. Within the technical constraints of owl:the same as and union join of SPARQL query, and economic constraint of ROIs, how can we 1) justify the cost and demonstrate ROIs and impact factors; 2) share best practice in working around the limits of technologies and standards; 3) empower existing workforce in bibliographic control; and 4) work together globally as library and museum communities towards the common goal, that is to transition toward the support of universal Bibliographic framework in a much more robust fashion than before, particularly in the areas of version and quality control, maintainability, etc.?
> If national libraries can take the lead in preparing training and justification of ROIs for local, national, and universal practice for Bibliographic control, that would be helpful in building up the momentum in RDA implementation that I've seen at ALA Midwinter 2012 in Dallas, TX.
> Thank you so much for sharing your proposal. The digital inclusive community definitely started with this group as far as I know.
> Sincerely yours,
> Amanda Xu
> Amanda Xu Sent from my iPhone
> On Jan 30, 2012, at 9:26, "Heuvelmann, Reinhold"<[log in to unmask]> wrote:
>> Dear colleagues,
>> as a response to the "Bibliographic Framework Transition Initiative" General Plan announced by the Library of Congress in October 2011, the German National Library has already expressed its appreciation, and the will to support the initiative and to actively take part in the process.
>> During a meeting in December, we decided to set up a paper and to outline some aspects of the framework that we may specifically be able to contribute to. After positive reactions by experts from the Library of Congress, we would like to make these thoughts and intentions available to the broader community.
>> The paper (PDF, 2 pages, 226 KB in size) is available at
>> Best regards
>> Reinhold Heuvelmann
>> Reinhold Heuvelmann
>> German National Library
>> IT / Office for Data Formats
>> Adickesallee 1
>> D-60322 Frankfurt am Main
>> Telephone: +49-69-1525-1709
>> Telefax: +49-69-1525-1799
>> mailto:[log in to unmask]
[log in to unmask] http://kcoyle.net