Page 19 - Building Digital Libraries

CHAPTER 1

be done. Search boxes do not appear until code specifies their
size, location, and appearance. Search results do not display
until resources are acquired, encoded, structured, and made
searchable. Patrons cannot even use text, images, or audio
unless they have software that interprets the files presented by
the repository.
Different purposes require different designs and technical
infrastructure. Dissertations and articles may require embargo
support, as well as the ability to provide access to supplemen-
tary data which require anonymization or access controls.
Images for use in marketing and presentations may require the
ability to manage large sequences contracted from individual
photo shoots, access controls, and release forms that limit how
and when they can be used. Datasets often require specialized
metadata and ways for users to interact with resources. Music
may require rights management as well as specialized data stor-
age, display, and user controls. Each type of repository needs to
interact with totally different workflows for acquisition and use.

Is the library the appropriate entity to meet this need?
It’s essential to articulate how the repository fits within the
library’s mission, priorities, and capabilities. Libraries have
expertise in organizing and preserving certain types of infor-
mation, but they lack the professional and technical resources
to manage a wide variety of materials. Just as the post office
isn’t the best entity to provide e-mail service just because they
have extensive experience delivering mail, the library isn’t the
best entity to manage and provide access to electronic resources
simply because it has long been trusted to preserve and provide
access to diverse physical resources.
Librarians are technically sophisticated and have strong
organizational skills. However, their expertise, tools, and meth-
ods are largely oriented around needs surrounding biblio-
graphic materials, and they are not usually equipped to provide
services that require specialized technical knowledge, major
computing facilities, or sophisticated access management.
Many communities already have repositories designed around
the needs of specific types of resources such as genomic or pro-
teomic datasets. Researchers who need to provide supporting
data for articles may be required to deposit their materials in
specific repositories. Even cloud services designed for individ-
ual use are appropriate in many cases.
Even when libraries have the expertise and technical capa-
bility to provide a service, they should still examine whether it
is appropriate to do so. Managing scientific datasets, extensive
high-resolution video, e-mail, resources containing sensi-
tive health or personal data, or information that is intricately
4