I think you're right on point.  The grant I'm working on now, Linked 
Data for Libraries, starts with use cases to drive the whole process.  
It's been very refreshing and I have to say a more unusual experience, 
lol.  So much of the Bibframe discussion has been focused on MARC 
conversion.  Its evaluation as a tool to solve discovery and access 
issues moving forward is almost never addressed.  I'm hoping LD4L will 
give us the opportunity to give that more thought.


On 6/26/2014 7:26 AM, [log in to unmask] wrote:
> Phil--
> That is a great example. I didn't for a moment intend to imply that moving from models of description based around records to models based around fine-grained assertions isn't going to have really good outcomes for patrons: I sincerely believe that it will and that it's a very important task. My concern was more to the area of process. I was wondering how use cases like the one you outline below have been assembled and examined and used as a basis for the Bibframe work. For example, at "" neither the "Analyses" nor the "Use Cases" bullet is actually linked to anything.
> ---
> A. Soroka
> The University of Virginia Library
> On Jun 25, 2014, at 12:12 PM, Philip Evan Schreur <[log in to unmask]> wrote:
>> I totally agree here!  I'm headed to ALA right now and hope to write more later but let me give one example of where Bibframe makes huge strides for users.  Music has been very I'll served by MARC, the combination of complex data, multiple works contained in a single resource, and a flat file structure makes discovery extremely difficult.
>> Bibframe allows you to include information such as performers, subject headings, timings, etc. in the individual bf:work areas for each individual resource in a multiple-resource collection( I.e. Sound recording).  All those false drops in searching that are pervasive in discovery systems now that are based on MARC can be eliminated.  This is a HUGE win for us music folks.  A brilliant solution and directly user focused.
>> Philip Schreur
>> Stanford University
>> Sent from my iPhone
>>> On Jun 25, 2014, at 8:45 AM, "[log in to unmask]" <[log in to unmask]> wrote:
>>> Amen.
>>> I'll add my own thought: the notions of "use case" and "functional requirement" are most powerful and meaningful when they are centered on _patron_ uses and requirements for functions exercised by _patrons_ (as opposed to _librarians_, using the term broadly: I mean to distinguish between those who are served and those who serve). It is not very clear to me by what process ideas about patron uses and requirements have informed the construction of Bibframe.
>>> ---
>>> A. Soroka
>>> The University of Virginia Library
>>>> On Jun 25, 2014, at 10:34 AM, Karen Coyle <[log in to unmask]> wrote:
>>>> Nowhere, however, do a see a serious discussion in the library data creation community of use cases and functional requirements. (And, believe me, FRBR does not provide this.) Many of the elements that have been added to MARC over the years (after taking many hours of discussion within the MARC committee) rarely appear in actual library data. Yet more continue to be added. Where are we headed? Why? What is the result we seek?

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