NARA’s Conceptual Framework for Preserving Electronic Records
There was a time when NARA’s former conceptual framework for preserving electronic records worked well. Electronic records sent to NARA were products of mainframe databases in ASCII. This situation changed with the developments and use of information and communication technologies, volume of digital information and diversity of documents in multiple formats. In order to address these challenges, NARA created a new framework. The Electronic Records Archives Project (ERA) is “NARA’s strategic initiative to preserve and provide long-term access to uniquely valuable electronic records of the US Government, and to position government-wide management of the lifecycle of all records into the framework of e-government” (Winstead, ERA Project Manager, 2010).
The conceptual framework for preserving electronic records adheres to the concepts of the OAIS Reference Model (Open Archival Information System) and employs the PREMIS model to structure its metadata. However, how does NARA plan to implement the digital preservation strategy? The first question that arises is “What are we trying to manage?” David Kepley starts with “records” and ends with “intellectual entities”. As David underlines, these intellectual entities must be understood by humans and gives the example of an e-mail with attachments with associated metadata that reflect its context. Moreover, the intellectual entities should be produced and maintained authentic and reliable. However, several issues arise at ingestion, storage and transformation stages. Authenticity tests are required for original and derived assets. Another important question that needs to be answered is “How to ensure access?”. David speaks about the provision of original records, the provision of viewers and transformation of records. The latter is an expensive strategy and can be done for preservation and reference purposes. It implies testing, tools, etc. and can be a very complex task. In this case, a cost-benefit analysis should be performed. Preservation is associated with vulnerability; therefore, two measures need to be considered: ubiquity and affordability. Regarding transformation planning, a collaborative work with archivists is recommended. The transformation plan is driven by the needs of research communities. The selection of formats will depend on users. If a format does not work for a research community it may have to be transformed. Records are transformed to suitable formats that should meet the following characteristics: be generic or open formats, ubiquitous in the market place, and self describing. However, one of the consequences of transformation is the impact on records. Will changes be important for certain records? Significant properties of records have to be assessed in close collaboration with archivists for the digital object to remain accessible and meaningful over time.
In summary, this framework ensures that ERA helps fulfill NARA’s mission of keeping records accessible into the future. A preservation prototype has been developed in the past six months on the basis of this framework.
Daniel Oliveira
