Tuesday, August 11, 2015

Digital Heritage: Semantic Challenges of Long-term Preservation

Digital Heritage: Semantic Challenges of Long-term Preservation. Christoph Schlieder. Semantic Web – Interoperability, Usability, Applicability. 05/06/2010.
     This is an excellent article on long term preservation. It is argued that a period of 100 years constitutes an appropriate temporal frame of reference for addressing the problem of semantic aging. Ongoing format migration constitutes currently the best option for temporal scaling at the semantic level.

Digital preservation focuses on finding solutions that scale well along the temporal dimension. In the pre-digital world, the preservation of written records over long periods of time depended on several factors:
  1. The record needs to be preserved physically
  2. The ability to read the record and language need to exist
  3. There must be a community that still shows interest in the record
There are a number of reasons why digital preservation may fail.
  •  Media aging: Any medium that carries a digital encoding will physically deteriorate until it is no longer possible to recover the original bit stream.
  • Semantic aging: The evolution of data formats and the fact that knowledge about data semantics quickly disappears if not specified explicitly.
  • Cultural aging: The community loses interest in some content; the documents are no longer retrieved, and is not maintained and transmitted any more, its loss is almost unavoidable
It is important to identify the time scale with preserving digital items. In the short term, 10 years, there are any number of solutions. So that the short term problem "may be considered solved". The most  ambitious time frame for digital preservation is the one promoted by the Long Now Foundation, which is a formidable period of 10,000 years. This article looks at a more modest period, 100 years. The agenda for this period is:
  1. find strategies to access digital contents from the past 50 years in spite of aging factors
  2. plan the preservation of currently accessible digital content for future use the next 50 years
This problem cannot be solved just by "simply agreeing on a standard format for digital archiving". There are two digital preservation initiatives, migration and emulation. Migration seems to work best for a document-centered workflow, and emulation "constitutes the best solution for archives of highly interactive media". Digital preservation initiatives conceived of preservation as an ongoing process around a digital curation lifecyle model and incorporate migration strategies to reduce risks. This is the basic link between digital preservation and the semantic web. How well this works depends on how solutions scale over the 100 years. "Only by looking at periods that are significantly longer than the 10 years" can we tell how the solutions will work. The aging processes have always been at work. "The best way to overcome the effects of semantic ageing is by migrating digital records into new formats." Long-term preservation forces the research community to adopt a much longer temporal frame of reference."

"Taking cultural ageing seriously means to abandon the idea that digital preservation operates like a time capsule. The picture of content that is enclosed in a digital capsule to be opened at some moment in the future is misleading because it is not the past that sends messages to the future. Rather, it is the present that makes choices, selecting content from the past and linking to it. This ongoing process of linking from the present into the past makes up digital heritage." 

Related posts:

No comments: