Print

Print


I think it's great that so many folks are getting enthusiastic about  
moving forward with bibliographic data, in particular with an idea to  
making library data part of the linked data information space.

I do want to caution, however, that RDF is not a serialization. It is  
a data model that is very different to what we are accustomed to. I  
would like for us to set in place some checks to make sure that we do  
not drag our assumptions and our old models into this new environment  
inadvertently. (See slide 5 of  
http://swib.org/swib11/vortraege/swib11-magnus-pfeffer.pdf for a  
perfect image of this).

In this regard I have two suggestions at this point:

1) engage some folks from outside of the library world who are steeped  
in the Semantic Web (Ivan, I think this is where you come in :-)) to  
take an unbiased look at the starting point and results and make  
suggestions

2) provide more than one interpretation of the options for the future  
metadata; it cannot be possible that there is only one way to  
re-interpret our data.

I suspect that as we move along this path we will encounter some  
interesting practical and philosophical issues. One example that I  
have noticed lately is the use of properties vs. vocabularies for  
things like genre and role. If you look at FaBiO's interpretation of  
FRBR (which could be questioned in many ways, but still made sense to  
someone) [1] you see that what in library data has been a controlled  
list of genres, in their interpretation is a set of subclasses. BIBO  
takes a similar approach, so I'm thinking that there is something to  
this. [2] I don't know which view is better, but surely there is  
something to discuss there before we take a decision.

Before we come up with answers, I'd like to see a discussion of the questions.

kc
[1] http://purl.org/spar/fabio
[2] http://bibotools.googlecode.com/svn/bibo-ontology/trunk/doc/index.html

Quoting "Riley, Jenn" <[log in to unmask]>:

> The MODS/MADS Editorial Committee is pleased to see the LC Bibliographic
> Framework Transition Initiative launched, and believe the stated approach
> to make the Transition Initiative a community effort is crucial to its
> success.  We believe the future of metadata standards is flexible and, as
> such, believe standards work will evolve beyond specifications of
> individual element sets or data properties towards combinations of
> properties that meet individual implementation needs and reflect the open
> assumptions about data suggested by the networked environment.
>
> Over the last several months, the MODS/MADS Editorial Committee has been
> working to articulate a formal data model for MODS, by analyzing the
> current XML Schema, User Guidelines, and implementation practices.  We
> plan to publish this model as a textual specification.  An expression of
> the model in RDF will either accompany the publication or follow shortly.
> A formal data model for MODS will enable implementers to create alternate
> In the future we will review the current MODS/XML schema in light of
> therepresentations of MODS, such as one for a relational database schema.
> defined MODS data model, and determine if a revised version of an XML
> Schema that more fully conforms to the data model is useful to the
> community.
>
> We hope this data modeling
> work will serve as one of many influences on the Bibliographic Framework
> Transition Initiative's work.  More information about a formal data model
> for MODS may be found at:
>
> http://mods.pbworks.com/w/page/43886707/Formal%20data%20model%20for%20MODS%
> 20and%20MADS
>
> Alternatively: http://bit.ly/sohX97
>
> We also note that the term "framework" is sufficiently general as to prompt
> questions. We recognize that it implies more than just a new record
> format, but we also look forward to further clarification on what other
> activities might and might not be in scope for this work.
>
> Signed,
> Jenn Riley, University of North Carolina at Chapel Hill, Chair, MODS/MADS
> Editorial Committee
> Jan Ashton, British LibraryAnn Caldwell, Brown University
> John Chapman, OCLC
> Gordon Dunsire, Consultant
> Rebecca Guenther, Library of Congress (retired, consultant)
> Reinhold Heuvelmann, German National Library
> Bill Leonard, Library and Archives Canada
> Sally McCallum, Library of Congress
> Betsy McKelvey, Boston College
> Jon Stroop, Princeton University
> Imma Subirats, Food and Agriculture Organization of the United Nations
> Robin Wendler, Harvard University
>
>
>
>
>
>
> --------------------------------
> Jenn Riley
> Head, Carolina Digital Library and Archives
> The University of North Carolina at Chapel Hill
> http://cdla.unc.edu/
> http://www.lib.unc.edu/users/jlriley
>
> [log in to unmask]
> (919) 843-5910
>



-- 
Karen Coyle
[log in to unmask] http://kcoyle.net
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet