Page 56 - Building Digital Libraries

Acquiring, Processing, Classifying, and Describing Digital Content

headings, one pro vides very specific headings, and the
third applies every heading she can think of, search results
will be very in con sistent. If the people assigning headings
refer to people, or ganizational units, geographical entities,
and concepts in different ways, the results will be inconsis-
tent. Different users have different levels of knowledge
and are willing to invest diff erent amounts of effort, and
some may apply metadata de signed to further personal or
political objectives rather than to help the library.
To mitigate these issues, the library needs to train con-
tribu tors, edit their contributions, identify contributions,
and provide version control. Except for the fact that users
do not need to physically access materials to contribute
metadata and provide descriptions, the challenges of
having users contribute metadata for digital collections are
identical to those for having them contribute metadata for
physical collections.

Resource Identification

Part of the ingestion process needs to involve creating identifiers so the
resources can be retrieved later. Identifiers are a little less straightforward
than may appear on the surface. All repository systems automatically cre-
ate identifiers, and most provide “permanent” Uniform Resource Locators
(URLs), but these often become problematic in future systems after the
resources are migrated. Moreover, the question of what identifiers to create
and what metadata to associate them with arises when an object consists
of many components.
Understanding identifiers requires familiarity with some jargon—you
will hear about URLs, Uniform Resource Names (URNs), and Uniform
Resource Identifiers (URIs). People are most familiar with URLs because
they use them every time they access a resource on the Web.
Here are the important things to know:

• URIs are text identifiers that identify a resource. A URI may
or may not include a network location.
• All URLs are also URIs. A URL must include a network
location.
• All URNs are also URIs. URNs unambiguously identify
resources but do not include a network location.
• When sharing information about how to access resources,
you should use URIs that will be the same after you migrate
rather than URLs that are different for each repository
platform.

Figure 3.1 demonstrates the difference:
41