HathiTrust Digital Library was launched in October 2008 as a joint undertaking by 25 research libraries to preserve and provide access to millions of volumes of their digitized holdings. Drawing on the collective experience of the founding members, which include the schools of the Committee on Institutional Cooperation (the Big Ten and the University of Chicago), the University of California system, and the University of Virginia, a robust and scalable infrastructure was created to house, manage, and provide access to the collections. This paper traces the development of the initiative from its origins, describing the challenges it has faced, its strategies for addressing them, and its vision and hopes for the future.

The National Geospatial Digital Archive, funded by the Library of Congress, is a partnership between the University of California, Santa Barbara and Stanford University with the purpose of providing long-term preservation of and access to geospatial data and imagery. Geospatial content in the public domain as well as that which is not is being accessioned into our collections. While public domain data may often be accessioned without having to seek explicit permission from the agency or group creating and/or hosting the materials, private data should be captured with a clear understanding of the rights and responsibilities of the stakeholders, be they the content owner or the university. In addition, the two collecting nodes (and any future nodes) must agree upon standards of conduct in order to trust that any content brought into the network will be managed well now and into the future.Joint agreements have been written by UCSB and Stanford that codify all of these relationships. This paper will lay out those agreements, which include the Content Provider Agreement, the NGDA Content Collection Node Agreement, and the NGDA Content Collection Node Procedure Manual. These three agreements provide the legal and structural framework of the NGDA allowing the stewardship network to grow over time with the understanding that each new partner will meet the standards necessary to provide secure long-term preservation of the content.

The Chronopolis Project Partners are building a digital preservation environment to manage and store data as a service to Data Providers. The system does not hold data from the partners. A subset of the project team, the Metadata Working Group made an analysis of the system by tracing a digital object through the system. System activities were divided into Event Types and metadata gathered by the digital object on its way through the system were documented.

The U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center archives electronic science collections that total over two petabytes in size and over 100,000 rolls of aerial and satellite film. Limited resources, the evolution of missions, and recommendations from advisory committees have led to the development of a scientific records appraisal process as a means for determining long-term archiving priorities. The process was formed through extensive literature searches describing approaches used to appraise administrative, physical artifacts, and science records. Less information was available that specifically addressed science records; therefore, relevant portions from each records appraisal process was assembled. In addition, involvement with the appropriate stakeholders was deemed critical and led to the active participation of scientists, records managers, and senior managers in the process. As part of the documentation portion of the process, an extensive online tool was developed to capture information describing each collection and detail preservation or access challenges that may be part of a collection. The U.S. National Archives and Records Administration recommends the tool as a best practice for U.S. federal agencies. To date, over 30 science collections have been appraised. This paper will detail the process used to appraise science collections for long-term archiving, the composition and rationale for the tool elements, and the results the USGS has attained.

The National Geospatial Digital Archive (NGDA) is one of eight initial projects funded by the Library of Congress's National Digital Information Infrastructure and Preservation Program (NDIIPP). The project's overarching goal is to answer the question: How can we preserve geospatial data on a national scale and make it available to future generations? This paper summarizes the project's work in four areas: analysis of the characteristics of geospatial data relevant to preservation; elucidation of the “relay” principles of long-term preservation; development of an OAIS-compliant archive system; and development of a wiki- and repository-based format registry.

Implicit within the metadata strategy of most archiving or preservation institutions is the use of format registries to contain important information, usually technical in nature, that is common among like formats or data types.At present, there are several approaches to data models, policies, and implementation models for format registries that are in various stages of implementation and/or conception including The National Archives' (UK) PRONOM Technical Registry, the Global Digital Format Registry (GDFR), funded by the Mellon Foundation and led by Harvard University Library, and the Library of Congress (USA).The National Geospatial Digital Archive (NGDA) project funded by the Library of Congress' National Digital Information Infrastructure and Preservation Program (NDIIPP) has been investigating whether existing or planned format registry efforts do or would support the often quite complex geospatial data formats which NGDA and other institutions are collecting for long term preservation.This paper discusses results of a comparative study of the data models of pertinent format registries in which instances of over 20 proprietary and open geospatial formats were examined to assess whether the elements within the data models could adequately describe a given format and its relatives, and if not, what other kinds of information would be important to include.The paper places the findings and recommendations into the context of previous work done by the NGDA team and others about what preservation metadata would be appropriate for geospatial resources. In addition, the paper discusses differences in definitions for key format registry concepts describing relationships among formats.Finally, we identify related research questions yet to be answered regarding the usability of existing and potential format registry efforts for geospatial resources, and the broader questions about the practicability of gathering this and other pertinent preservation metadata for geospatial resources.

This paper describes (1) a lessening of restrictions on the ingest acceptance criteria, and (2) a growing reliance on repository-supplied Submission Information Package (SIP) creation tools planned for the Harvard University Library's digital preservation repository (the DRS). Reasons for these changes include producer demand for the repository to accept new formats and genres, growing quantities of digital content producers want to preserve but can not adequately process, and the increase in the amount of content that is acquired by instead of created by DRS producers. In response to these needs the DRS will greatly change its SIP requirements and ingest process. Most notably files in any format will be accepted by the DRS. This paper describes steps taken to mitigate the effect of these changes so that the DRS can still meet its preservation commitments while broadening the range of acceptable content. One of these mitigating steps is the development of a new tool, FITS, which will be used to automate format identification and metadata extraction during SIP creation.

The Smithsonian Institution Archives (SIA) and the Rockefeller Archive Center (RAC) conducted a three-year pilot that explored preservation challenges with email collections. This paper reviews the acquisition model and workflow used based on the OAIS Reference Model. Rather than focusing on individual messages, the Collaborative Electronic Records Project (CERP) settled on preserving an account as a whole, maintaining the structure and relationships within a collection as well as simplifying metadata management. This paper also reviews some of the challenges with the email collections, including lack of organization and inclusion of non-record/sensitive material. Both archives also addressed the importance of sound recordkeeping practices and retention schedules and issued various guidance documents for depositors.CERP also collaborated with another research team (the EMail Collaborative Initiative (EMCAP)) to develop an XML schema capable of encompassing a complete email account and its content. The E-Mail Account XML schema defines a standard XML structure for preserving an email account along with its internal organization, its messages and attachments, and the interrelationships of the messages without sacrificing granular email message data. This paper describes the schema, its unique characteristics, and its value to the archival and digital preservation communities in the context of, and comparison to, other efforts to digitally preserve email.The schema structure positions preserved email accounts for multiple levels of searching strategies including: individual messages, account-wide, and cross-account search and retrieval. This helps to expose social networks and message interrelationships present in, and across, accounts.The E-Mail Account schema has made possible the preservation of large bodies of related e-mail in a single XML file, as demonstrated in the recent EMCAP and CERP projects. Unlike other work in the area of e-mail preservation, this XML schema is distinct in: 1) its account-based paradigm; 2) the granularity of data captured; 3) its alignment with the email message standard RFC 2822; 4) the support of a single XML file representation of the account; and 5) its incorporation into two separately developed e-mail preservation software applications.

Since 2006, the Contemporary Art Group at the Research Department of French Museum Conservation and Research Center (C2RMF - Ministry of Culture) has been working on technological obsolescence phenomena. The purpose of this program is to study the impact of technological evolution on contemporary art works and, more precisely, on their preservation. Among contemporary art collections, our group focuses essentially on technological installations, with audiovisual or digital elements, and is mainly concerned with obsolescence phenomena.Since the late 1960s, an increasing number of art works kept in contemporary art collections consist in still or moving images. Many of these images are now conserved on obsolete devices and formats which essentially has as consequences that it becomes difficult to produce new copies when operating copies are damaged; it becomes difficult when the state of conservation of originals is alterated, to remedy (i.e. to treat or replace); and it becomes difficult to find the necessary equipment to read (or diffuse) these images. It becomes simply impossible to present certain parts of museum art collections: their accessibility is threatened and hence their own conservation and preservation.Faced with this combined obsolescence of media, devices and formats, those responsible for collections, the curators or conservators, frequently carry out the digitization of repositories for purposes of diffusion or preservation. However, a digitization campaign requires specialized and technological skills which are currently not available in museums.Because arts works are unique and non standard objects, methods of mass digitization used in other domains, such as television or libraries archives can not be directly imported. We actually focus on avant-garde or experimental cinema digitization, because film in 16 or 35mm will probably soon no longer be available. We made a study on the digitization of such movies, with focused on conservation of colors and on time based phenomena like flicker. Avant-garde and experimental films often consist in abstracts images, based on colors and flicker. We wanted to evaluate the modifications leaded by the change of technology.We detail here an experiment focusing on color. We first defined the framework of reproduction adopted by the two main French repositories of art movies. Then, we worked with a postproduction company and analyzed digitization, based on the colorimetric characterization of input and output digitization devices, inspired by color management in photography.Three colorists did a digitization with the aim of preserving the original colors. The results show a huge difference between film colors and digital colors. An analyze of such differences found correlations between colorists and connects these results with photography, color management and psychophysics. Limits of color conservations in digitalization of experimental movies are exposed here as well as advices in order to minimize then.