In conserving the analogue and digital ‘mass media’ of the 20th century, the Federal Office of Culture recognises the current challenge associated with preserving Switzerland's cultural heritage.Greater emphasis is currently placed on the cataloguing and conservation of audiovisual documents and electronic publications than on the digitalisation of printed works.The information society attaches importance to sound, image and film documents, to radio and television recordings, digital media carriers and multimedia processes as contemporary documents. Audiovisual documents increasingly influence our view of the past and offer new points of reference in the process of how memory develops. They belong in the memory of the multimedia society and have to be preserved before they disintegrate.

The report starts with an introduction to the project e-Helvetica at the Swiss National Library. The goal consists in undertaking organizational measures to collect, catalog, prepare and preserve electronic Helvetica over the long-term, while establishing an archival system for these publications. After that it describes the Ingest system that has been implemented to automate the process of receiving data and metadata, preparing and transforming it for long-term preservation and saving it in archival storage. Special attention is paid to the aspects of harvesting websites.

This article explores the idea of extending the concept of a managed environment, beyond the digital repository and forward in time to include the entire digitization process, for an extremely large and complex institution. Rather than managing digitization independently of the digital repository, this article suggests that it is advantageous to work from the beginning of the digitization effort towards the end goal of long-term preservation of digital objects and related metadata in the digital repository. In particular, this article briefly examines the U.S. National Archives and Records Administration's developing approach to managing digitization projects, including: characteristics of digital objects produced during the digitization process; working data models and approaches for collecting all types of metadata to be done locally in the digitization labs and in centralized environments; where this information is held; digitization activities and influence/impact across functional areas; and discussion of how tools and processes that have been implemented at the local lab level can inform the definition and development of a centralized, institution-wide information technology infrastructure to support digitization at NARA.

Digital repositories are a manifestation of complex organizational, financial, legal, technological, procedural, and political interrelationships. Accompanying each of these are innate uncertainties, exacerbated by the relative immaturity of understanding prevalent within the digital preservation domain. Recent efforts have sought to identify core characteristics that must be demonstrable by successful digital repositories, expressed in the form of check-list documents, intended to support the processes of repository accreditation and certification. In isolation though, the available guidelines lack practical applicability; confusion over evidential requirements and difficulties associated with the diversity that exists among repositories (in terms of mandate, available resources, supported content and legal context) are particularly problematic. A gap exists between the available criteria and the ways and extent to which conformity can be demonstrated. The Digital Repository Audit Method Based on Risk Assessment (DRAMBORA) is a methodology for undertaking repository self assessment, developed jointly by the Digital Curation Centre (DCC) and DigitalPreservationEurope (DPE). DRAMBORA requires repositories to expose their organization, policies and infrastructures to rigorous scrutiny through a series of highly structured exercises, enabling them to build a comprehensive registry of their most pertinent risks, arranged into a structure that facilitates effective management. It draws on experiences accumulated throughout 18 evaluative pilot assessments undertaken in an internationally diverse selection of repositories, digital libraries and data centres (including institutions and services such as the UK National Digital Archive of Datasets, the National Archives of Scotland, Gallica at the National Library of France and the CERN Document Server). Other organizations, such as the British Library, have been using sections of DRAMBORA within their own risk assessment procedures.Despite the attractive benefits of a bottom up approach, there are implicit challenges posed by neglecting a more objective perspective. Following a sustained period of pilot audits undertaken by DPE, DCC and the DELOS Digital Preservation Cluster aimed at evaluating DRAMBORA, it was stated that had respective project members not been present to facilitate each assessment, and contribute their objective, external perspectives, the results may have been less useful. Consequently, DRAMBORA has developed in a number of ways, to enable knowledge transfer from the responses of comparable repositories, and incorporate more opportunities for structured question sets, or key lines of enquiry, that provoke more comprehensive awareness of the applicability of particular threats and opportunities.

This contribution focuses on core structures and performing aspects of digital media archives. It takes up a practical point of view and focuses on open source solutions. By means of recently developed database systems four subjects will be faced: the presentation of different metadata models and their practical implementation, ingesting techniques for metadata and media files (workflow), storage systems and last but not least the appearance of so called meta archives. They give considerations to the consolidation of a distributed networked search among heterogeneous metadata systems.

This paper discusses the Planets approach to migration tool development. The approach consists of enhancing existing migration tools rather than developing tools from scratch. This pragmatic approach is based on the Planets view of the current situation for migration tools and two claims. The first claim is that the market will cover the required tools for commonly used formats. The second claim is that in the long term less tools will be required due to growing use of archiving standard formats. The Planets view on the current situation, the scope of tool development and the claims stated are, however, open for discussion and re-evaluation.

The TIFF uncompressed file format is a widely accepted standard for storing master images of digitisation projects. As in the Koninklijke Bibliotheek, The National Library of the Netherlands (KB) the scale of these projects is rising rapidly, the need for an alternative, compressed file format is felt. This paper contains a summary of a research project in which four alternative file formats - JPEG 2000 Part 1 lossless and lossy, PNG, JPEG and TIFF LZW - have been described and tested. Four consequences of a choice for either of the formats have been described: consequences for storage, image quality, long term durability and functionality. In the final recommendations these consequences were weighed against three reasons the KB distinguishes for wanting to store master images:1. Substitution (JPEG 2000 Part 1 lossless or PNG)2. Redigitalisation is no option (Visual lossless JPEG 2000 Part 1 lossy or JPEG)3. Master file is the basis for access (JPEG and JPEG 2000 lossy with higher degrees of compression)

Preservation of databases is a complex task that has no evident and straightforward solutions available. At Mikkeli UAS we are researching and evaluating a preservation model called “Normalized Object Model”.We create the Object Model by extracting to an object the structure and metadata of the structure as well as the content and content metadata from databases at preservation. The Normalized Object Model reduces the complexity of the original database by preserving the structure as metadata. The Normalized Object Model relieves maintenance of preserved databases from knowledge on the original structure of them and resists technical quality problems at ingest.We achieve Normalization by creating one single object structure to contain all database structure elements and one single property structure to contain database content elements. Normalization makes it possible to query several preserved databases simultaneously regardless of structure.A library of object and property types shall be established to create rules for and assist at planning and designing future, standardized database structures.

A continuing question that plagues the Archiving community is the choice of file format to use for archiving electronic documents. It is complicated by many factors including: the wide spectrum of archiving situations and specialized requirements, continuing evolution of existing document file formats, the invention of new kinds of documents and electronic content which must be archived, and the difficulty of planning to have all the appropriate hardware, software and media available in the future to process these documents. In this paper we focus primarily on the choice among existing document file formats and discuss their current suitability for various anticipated tasks.

On request of nestor, the German network of expertise in digital long-term preservation and long-term availability, the German Institute for Standardisation (DIN) has set up a standards committee on long-term preservation standards in the beginning of 2008. This article will introduce the initial work program of the DIN standards committee.