What is ‘active’ digital preservation?
In my second digital preservation post I discussed “bitstream” preservation – the minimum level of activity necessary to ensure survival of the bits (or bytes) of a digital object. Maintaining digital collections this way may help ensure the assets survive unchanged but does not guarantee long-term access and use.
Storage media may become corrupted, and some formats or software may become unreadable as tech moves on. Archivists need to take an interventionist approach and adopt technological approaches to ensure information can be accessed in future. The two main approaches, emulation and migration, are discussed below.
Emulation
Emulation involves mimicking an obsolete computer environment to read software or applications that cannot be read on newer hardware. It first became popular within the video gaming community during the 1990s, nostalgic enthusiasts developing emulators to resurrect classic games.
A cursory YouTube search will bring up links where you can watch emulator demos or walk-throughs of a title, and websites exist where you can play these online. From the early 2000s, emulation became increasingly popular in the digital preservation world, particularly amongst practitioners dealing with complex multi-part resources.
There are drawbacks. Creating emulators is a technologically complex and expensive undertaking. There are significant intellectual property issues. Video games, for example, are inherently complex assets, created collaboratively between software companies, programmers, sound and graphic designers and others.
Emulators themselves are at risk of obsolescence. Most are open source, their existence depending on the support of a community of enthusiasts – if this is withdrawn the technology disappears, the assets they support becoming inaccessible. Emulation is not always necessary, in the case of most standard file formats migration provides a preferable alternative.
Migration
File format migration (or conversion) involves copying data from one obsolete/at-risk format to another so it can be read on newer computer applications. It is arguably the most common form of active digital preservation activity.
Migration has many advantages. It is relatively cheap, and most software applications support it to some degree. If you don’t have a digital preservation repository system or proprietary software packages, numerous open-source applications exist that can do this. A list of these can be found on the Community Owned digital Preservation Tool Registry (COPTR) Wiki.
Most integrated digital preservation systems have such non-proprietary tools built in. The image below shows a range of migration options presented during a preservation action in our Preservica system.
Preservation should ultimately be reversible – we would always bit preserve an original file and should be able to revert to it should a migration pathway prove unsatisfactory in future. Preservation masters should exist in formats which are robust, widely used, and well-documented, and preferred migration formats should follow this standard.
For example, it would not be good practice to migrate from TIFF file to a “lossy” JPEG format, compared to JPEG 2000 or Portable Network Graphic [PNG]. This only assumes TIFF is at risk of obsolescence – currently it remains a well-supported format and a robust standard for the purposes of digitisation.
Preservica can perform a range of migrations, not only to preservation but also access versions. An example of how this looks – where a TIFF has been migrated to a JP2 preservation version and PDF access surrogate – is below.
Migration does have drawbacks. It is not always preferable, or indeed possible to migrate in the case of interactive multi-part assets. Whenever you migrate, you risk losing a degree of functionality and context integral to the object’s creation and previous use. Critically, pressure is exerted on storage as successive migration actions lead to multiple renditions of master files building over time. Knowing exactly when to migrate is tricky to get right.
Any organisation taking digital preservation seriously needs a strategy in place for undertaking migration and (possibly) more complex preservation actions. I am working on a migration strategy for Archives & Collections, although I’m confident the bulk of the mainly static formats we have are robust, well-supported and can be migrated easily if required.
The digital world is ever shifting. In the future our service may be offered increasing quantities of more complex resources, possibly augmented reality or artificial intelligence. To enable long-term preservation and access, migration alone will not be an option. A hybrid strategy would likely be required, incorporating emulation and other activity, requiring significant investment in skills and resourcing.
Michael Hunkin, Digital Preservation Officer
The Iron Room 