Dear Policy people,
In case this might prove useful, here's a short summary of the RLG
list of a trusted repository's responsibilities and its recommendations
for establishing a network of trusted repositories.
Yrs,
Sarah
------------------------------------
Trusted Digital Repositories (RLG report, May 2002)
OPERATIONAL RESPONSIBILITIES
I. Negotiate for and accept appropriate information from information
producers and rights holders. Determined in part by how much control the
repository has over resource creation and submission. There must be
well-documented policies about what is selected for deposit and what
format it will be in. Negotiations cover:
* Legal issues, such as copyright, long-term maintenance and
access. May need separate negotiations for current access and
long-term preservation. Must have effective procedures and
workflows for obtaining copyrights for access and preservation.
* Preservation metadata. Repository must have agreeements with
content providers for bibliographic and technical metadata. There
must be comprehensive metadata specification and agreed-on
standards for implementation
* Authenticity checks. Need procedures and systems to confirm
submitted materials are what expected.
* Record keeping, with adequate documentation of all
transactions with content providers.
II. Obtain sufficient control of the information. This covers period
after submission while preparing materials for storage in the repository.
* Analyze digital content. Repository, rights owner, and system
managers must access digital objects to determine which
properties are significant enough to require preservation
(including rights clearance in deliberations) and apply policies
about acceptable formats and do any migrations. Should be as
automated as possible.
The repository needs to decide what level of preservation should
be applied to the material before putting it in the repository.
Simply maintaining a bytestream doesn't ensure that digital
material will be preserved at an acceptable level. The collection
manager judges the appropriate preservation and access levels to
fulfill the repository's responsiblities and the stakeholders'
(e.g., author and publisher) needs.
A digital object's significant properties dictate its underlying
technical form, which must be documented and supported, and the
amount of metadata that must be stored alongside the bytestream
to ensure that the object is accessible at the agreed-upon level.
The more significant properties deemed necessary, the more
associated metadata that will be required. The significant
properties might be an object's textual content, DTD, stylesheets,
software for running video files, etc. They should be determined
on a system-level automated process, by policy decisions
previously made for object classes. Technical metadata should
also be automatically generated.
* Continuing access arrangements. As technology changes, these
may need to be reevaluated.
* Metadata verification/creation and documentation to support
long-term preservation.
* unique and persistent id of materials.
* AIP creation. Find reliable and practical way to store digital
object and its associated metadata. Can store object in
repository and metadata in another system, but not all that good
for long term preservation. Alternative is encapsulating object
w/metadata as single object (perhaps using METS_.
* Authentication and integrity checking. Should also confirm
object's usability and functionality.
* Archival storage. Must be a well-documented policy for storage
and maintenance. If using a 3rd party, there must be
service-level agreements. Policy should include systems for
routine integrity checking of the bytestream,
geographically-distrubted back-up systems (i.e., redundancy of
data storage), security, and disaster preparedness, response, and
recovery.
III. Determining repository's designated community. Preservation occurs
for the community, by how the community values an object's content. The
community's technical capability and knowledge base also determines
technical infrastructure. To measure this, the repository should analyze
and document the current designated community and consider its future
needs and modes of access.
IV. Ensure preserved information is independently understandable to
the designated community. That is, the community can understand it without
expert assistance. This can mean that it the technical metadata for
rendering binary data into meaningful digital objects corresponds to the
lowest common level of technical capacity (e.g., assuming that everyone
can use a web browser). Over time, this will change, and it may be
necessary to change access methods w/o changing the stored object. To
help, the community and its technical level should be clearly defined.
Also, the technical metadata should be well-documented and maintained and
evolving technology should be monitored.
V. Following documented policies and procedures. Necessary for
long-term preservation. In research repositories, digital policies may
need to co-exist with nondigital policies. Should be link between policies
and procedures, since over time will reduce costs if can support
automation and scaling. Need:
* Policies for collections development, access control, storage,
defining designated community & community's knowledge base
* A rigorous system for updating policies & procedures as
technology changes and the community develops
* Explicit links between policies & procedures, allowing easy
application across heterogeneous collections.
VI. Making the preserved information available to designated
community. Access must be clearly defined if the repository is to
understand its implications: different levels of access require different
policies and management. Access arrangements will change as law,
technology, licenses, and local resource constraints change.
Resource discovery, authenticity (during submission, ingestion, storage
and distribution), legal issues, pricing, user support, and record keeping
all need to be considered. Need:
* System for discovering resources
* Appropriate mechanisms for authenticating digital materials
* Appropriate access control mechanisms and an "access rights
watch."
* Mechanism for managing electronic commerce
* User support programs
VII. Advocating good practice in creating digital resources. Need to
involve designated community and software suppliers in developing
standards and best practices. Key players (RLG, OCLC, etc.) should
facilitate this dialogue.
RECOMMENDATIONS
1) Develop a framework and process to support the certification of
digital repositories.
2) Research and create tools to identify the attributes of digital
materials that must be preserved.
3) Research and develop models for cooperative repository networks
and services.
4) Design and develop systems for unique, persistent identification
of digital objects that expressly support long-term preservation.
5) Investigate and disseminate information about the complex
relationship between digital preservation and intellectual
property rights.
6) Investigate and determine which technical strategies best provide
for continuing access to digital resources.
7) Investigate and define the minimal-level metadata required to
manage digital information for the long term. Develop tools to
automatically generate and/or extract as much of the required
metadata as possible.
This archive was generated by hypermail 2b30 : Tue Jun 11 2002 - 10:14:08 EDT