[Dini-ag-kim-bestandsdaten] distinction between collection and holding

[Voß, Jakob]

Owen Stephens wrote:

> This makes me wonder about the relation between Holdings and 
> Collection Level Descriptions.

The distinction between collection and holding can be quite arbitrary
especially when dealing with digital documents. We already have
physical edge-cases such as loose-leaf editions and serials. In the
end, a collection is both a set of documents and a document in its
own right. Don't let physical traditions of book-binding blind the 
concepts of collections and holdings.

examples of collections include:

* The entire Internet Archive
* All documents in my local town library combined
* A set of 21 rare books that once belonged to a book-lover, now part of our university library
* The bookmarks in my browser
* An incomplete comic series on my shelve at home
* A book that combines articles by several authors
* Any book is a collection of its pages

The distinction between collection and holding is not an either-or, but a relation.
Some collections contain holdings and some collections are holdings. Some
holdings are also collections, some holdings are contained in one or multiple
collections, and some holdings are instances/copies/manifestations/whatever 
of a collection or of a part of a collections.

A holding ontology, as far as I am looking for, should less define the nature
of a holding, but how holdings can be related to other entities, such as collections,
organizations, places etc.

Carsten wrote:

> On the website I found this levels listing of the granularity of library metadata [1].
> And surely we have to deal with level 6 and 7.

[1] http://www.w3.org/2005/Incubator/lld/wiki/Library_standards_and_linked_data#Granularity_of_library_metadata

Sorry, but this classification is as stupid as a printed catalog card. It
is so backward and broken, we should better forget about it.

Jakob