Print

Print


This is a "what's in the FRBR(IFLA) model often does not see the light of
day" situation.

The FRBR model makes a provision for "aggregate and component entities"
(1998, section 1-3.3), the use of which can resolve a number of problems
related to modeling multiple WEMI descriptions and multiple resources.

If you tell yourself that an aggregate entity acts as a "container" for
one or more component entities, you can then imagine separate and
overlapping metadata requirements for each type of entity.

While some of you are evidently good at holding many (many) triples in
your head and reasoning from them, I found it more useful to steal from
the physicists and use diagrams* to represent assemblages of component and
aggregate entities. First, see:

http://www.slideshare.net/RonMurray/from-mobydick-to-mashups slide#
138-141 for the FRBR-diagram connection.


Now say to yourself:

1. Each hollow box contains the relevant WEMI "aggregate" entity metadata
and the solid boxes contain component, or the "regular" WEMI metadata.

2. Then say that each box blooms out to a WEMI graph with a UID that acts
as a target for "attribute" nodes.


3. Finally, each link drawn from each box to another box resolves into to
a link between the WEMI UIDs in each graph that bloomed out from a box.

So if you have an aggregate Item-Level description that draws together
multiple component Item-level descriptions, the aggregate's attributes
(and linkages) applicable for all items can trickle down, while component
attributes and UIDs can link to item-specific information.


Ron Murray

-------------

* That slideshow is rather extensive, but can clarify a lot of modeling
issues, should you choose to delve.

The above aggregate/component strategy comes from faithful use of the LISP
computing language - so the BF folks should not have much trouble
translating it to BF.



On 1/14/15 10:30 AM, "Karen Coyle" <[log in to unmask]> wrote:

>I don't think we've helped Rob, so I'm going to hazard a suggestion, for
>discussion.
>
>On 1/13/15 3:35 PM, Robert Sanderson wrote:
>> My objections are:
>>
>> 1. Annotation is not an appropriate base class for this, regardless of
>> whether it's a record or a real world object.
>> This is my primary concern. The thing with a barcode is not "about"
>> the Instance. There's no comment or body of the thing with the barcode.
>
>Create a third "circle" in the bibliographic area for curated "Item".
>This could be physical or digital. Every "thing" will be represented by
>an Item. This will essentially be the MARC Holdings Record. For each
>item curated by the library there will be:
>
>Item-level bibliographic information - bindings, provenance, special
>characteristics
>Item-level preservation information
>Serial holdings
>Identification of the holding library
>Additional location information ("Alcove B")
>Item location number (call number)
>Item inventory identifier
>[add what I left out]
>
>If you have 2 copies, you have two of these Item graphs.
>
>Complication I see: (I'm sure there are more)
>- It would be convenient to say that operationally every inventory
>identifier = new Item graph. However, in some libraries, parts get their
>own inventory identifiers (e.g. multiple DVDs in a case).That would
>split up sets. Also, some libraries barcode each issue of a serial.
>"Holdings" gets complicated pretty quick.
>- For a given library, this may not be the most efficient way to manage
>holdings. You could imagine a Library or Location graph that links to
>Item graphs. I suppose we could presume that internally a system can
>"re-graph" this based on needed functionality.
>- Item vs. annotation gets especially complicated for collections of
>unique items. In a sense, an artwork could be an item, and will have no
>manifestation. How far should we push the physicality division between
>Instance/Manifestation and "thing-i-ness"? (There may be no answer to
>this question.)
>- Nothing in BF helps us understand how this data interacts with
>purchasing, circulation, or other library/archive management functions.
>So it's hard to know if we've got what we need for all of those. It's
>the Item level that interacts heavily with those functions, so perhaps
>this needs to be thought through.
>
>OK, there's a straw person. Go at it.
>kc
>
>> otion of authority records, we should equally drop the notion of
>> holdings/item records and simply identify and describe what we're
>> concerned with.
>>
>> I want to be able to make assertions about books in the same way that
>> I want to make assertions about people, locations, events, and all of
>> the other things we hold dear.  Say I take a photograph of the book. I
>> want an identifier in order to say <photograph> <depicts> <book> ...
>> not <photograph> <depicts> <item-record> ... that would be a very
>> different photograph, likely of a computer screen :)
>
>-- 
>Karen Coyle
>[log in to unmask] http://kcoyle.net
>m: +1-510-435-8234
>skype: kcoylenet/+1-510-984-3600