Another issue is the cataloging of sub-assemblies and wholes within wholes.
Material Culture collections and Archaeological collections may have
especial need of such a utility.  Objects such as an automobile in some
circumstances may have to be described as a set of units and each unit as a
set of components.  Archaeological collections may define accession groups
that record the items delivered from a single excavation.  Within that
group there may be logical groups, such as those items that derive from a
single dwelling and/or items from a single archaeological level.  On some
occasions, as for an historical house, or for entire buildings such as
cathedrals, it may be important to identify each part as a unit in a
cascading set of ever-inclusive wholes.  Obviously, to meet the
requirements of this kind of cataloguing, some type of hierarchical object
description scheme must be defined.  And this scheme must be able to define
groups that do not necessarily correspond to the accession history or to
the logic of assigning accession numbers.

To the problem of counting objects that exist as group entities and as
unique items, I'd suggest requiring the count at group level to contain a
"calculated" value based on a dynamic assessment of the number of items
catalogued under it.  Such calculated values should exclude any entity
whose count itself is calculated, and should display identifiably as a
calculated item.

Finally, object group administration for inventory purposes should allow
collection managers to compare the number of counted records within a group
to the number of records that are expected to be in a group.  In addition,
when reporting on an item that is a part of a group or subgroup, the report
should be able to extract identifying information from each level of object
description superior to the item cited.  When an object has been separated
from the normal storage location of the group, whether on display, in
conservation or on loan, the counting system should recognize that the
group has been physically split up.  Similarly, when each object is
recorded as in its permanent storage area (be that on display or in
storage), the database should be able to indicate immediately that all the
parts are in their permanent locations.

One last point.  Museums do not always hold all object that make up a
logical group.  One museum may hold one panel of a triptych, and other
museums the remainder.  Object management systems should be built so that
records for associated objects can be entered into the database, even
though they are not holdings of the institution in question.



===========================
Robert A. Baron
Museum Computer Consultant
P.O. Box 93
Larchmont, NY 10538 (USA)
mailto:[log in to unmask]