All,

I, too, was impressed with the BIBFRAME draft model.  It approaches the 
transition in a practical way in the best sense of the word.  Rather 
than think about what I wish were different (at least at first), I have 
some questions based on the model as it is currently defined.

1.) BIBFRAME Instances:  The document says that Instances include 
relationships to appropriate BIBFRAME Authorities related to 
publication, production, and distribution of the material resource.  My 
question (like many other of mine) comes from musical sound recordings.  
Will relationships to Authorities for names (performers, etc.) be made 
at this level (the description of BIBFRAME Instance doesn't forbid this 
although it doesn't explicitly mention them either) or will each 
realization of a work be considered a new BIBFRAME Creative Work and 
these associations made at that level?  Authorities for subjects could 
be particularly confusing as a performance of a Bach harpsichord piece 
on the piano could have different subjects in regards to the Work and 
the Instance.

2.) Annotations: Can annotations themselves be annotated?  I'm assuming 
that if Stanford held 6 copies of an Instance that we would add 6 
annotations, one for each instance.  If we needed to note that copy 6 
had had preservation treatment, or was missing pages, would we add an 
annotation to the annotation for copy 6?

In general, BIBFRAME seems very centered around resources that have been 
published in some fixed form (perhaps the term "Instance" unfairly makes 
me lean towards that conclusion).  I wonder how BIBFRAME will handle 
continuing resources, those things like websites that evolve with time.  
Will there be multiple Instances of a work reflecting an evolving 
resource at a particular time?  Will annotations need to include 
concepts of time?  I also wonder about implied relationships between 
Instances of the same Work.  I some cases, as in a textual work made 
available in different formats, the content may be identical.  In 
others, like recordings of an aleatory piece for music, the Instances 
may vary wildly.  Does the BIBFRAME model make any assumptions about the 
relationship between Instances of the same Work?  Will there be a place 
for us to explicitly define these varying degree of sameness?

Philip

-- 
Philip E. Schreur
Head, Metadata Department
Stanford University
650-723-2454
650-725-1120 (fax)