Posting to RDA-L concerning the BIBFRAME model document
On 11/26/2012 09:53 AM, Bernhard Eversberg wrote:
A lot of speculation. We have to simply ask for the reasoning that resulted in the draft as it is. Maybe – another speculation – they came to the conclusion (after studying Jim Weinheimer’s musings, for instance) that WEMI is impracticable or not desirable or not worth the time and effort.
Is there none of the insiders here to provide some background on all this?
I would personally like the BIBFRAME suggestions much more if all
references to FRBR were taken out. As Eric Miller mentioned in his talk
at LC about the new bibliographic framework,
http://www.loc.gov/today/cyberlc/feature_wdesc.php?rec=5605 (and the
ensuing discussion on Autocat), the framework must be simple. If it is
not simple, nobody will even think about using it. Very wise words.
Let’s face it: the FRBR structure is bizarre and difficult even for
trained catalogers to grasp. In addition, it is based on some unproven
ideological concepts. Can we really expect non-librarian web creators to
understand it so that they can implement it in some kind of coherent way?
While the new FRBR relationships may sound good at first, the reality is
rather different. Introducing those relationships will seriously devalue
the records in the catalog and degrade search results if we are not to
undertake massive retrospective work. That is a simple fact
This is certainly not the underlying idea of linked data. The idea of
linked data is to take the data that you currently have, share it
openly in certain ways, and link information where it can be linked.
Just by doing that, it is assumed that you have increased its value. We
should find out if it is true. We don’t have to devalue the millions of
records we have now.
The FRBR user tasks are from an earlier time, and in any case, the
public hasn’t been able to do them since keyword searching was
introduced–even in our library catalogs. That has been quite awhile now
and I have never seen or heard of anyone complaining. Those original
tasks have been long forgotten and have now been superceded in a
multitude of ways. Besides, if somebody wants to navigate WEMI, it can
be done now with the right catalog software.
For navigation and discovery, let advances in software deal with all of
that, just as modern software can now extract all the headings and turn
them into facets for further searching. What more will advances in
software be able to do as full-text becomes increasingly available and
library metadata can interact with it more and more? And as image
searching (http://www.tineye.com) and sound searching
The new bibliographic format should be aimed at the public. They will
be the consumers–not catalogers. We should not be trying to make
ideological statements that others will not understand or care about,
such as trying to teach WEMI or to convince them that people really
want the FRBR user tasks. People want resources. Create records that are
reliable, based on what we have now. If other projects harvest our
metadata, assume that they will not retain our format, but will
transform it into something else that serves their needs–just as
libraries do with the formats they get.
The first steps in the new format should be to make it in the simplest
ways possible so that web creators can use our records as soon as
possible. The catalog world is starving for feedback! Once feedback
starts coming in, the future should become clearer.