Digital Preservation Matters - 12 September 2008

It’s Happening Now: This is the Tera Era of Data Storage. Larry Swezey. Computer Technology Review. 16 September 2008.

New visual and audio drive storage capacities upward but the new digital data explosion is very different. More data is being produced and it becoming a more important part of in all aspects of our lives. But the large files we see now are just beginning. The size of files and the amount of data is increasing dramatically. The sizes are moving into the terabyte range [already there in many cases]. More data is being retained. AV items demand more storage. People expect large amount of information to be available almost immediately.

Using METS, PREMIS and MODS for Archiving eJournals. Angela Dappert, Markus Enders. D-Lib Magazine. September/October 2008.

Many decisions need to be made on metadata, including the structural and preservation metadata. The British Library is developing a system for ingest, storage, and preservation of digital with eJournals as the first content stream and developing a common format for the eJournal OAIS Archival Information Package (AIP). EJournals are complex and outside the outside the control of the digital repository so it does not have the structure for submission packets, format standards and such. This article shows one approach to defining an eJournal Archival Information Package. It has a database that provides an interface for resource discovery and delivery. An archival store is a long-term storage component that supports preservation activities. All archival metadata is linked to the content and placed into the archival store. The archival metadata is represented as a hierarchy of METS files with PREMIS and MODS components that reference all content. Each manifestation of an article is stored in a separate METS file. There is no existing metadata schema that has all the descriptive, preservation and structural metadata, but this is how they use a combination of METS, PREMIS and MODS to create an eJournal Archival Information Package.

Introducing djatoka: A Reuse Friendly, Open Source JPEG 2000 Image Server. Ryan Chute, Herbert Van de Sompel. D-Lib Magazine. September/October 2008.

Support for the JPEG 2000 format is emerging in major consumer applications, many consider it suitable for digital preservation. This introduces djatoka, an open source JPEG 2000 image server with basic features and they urge others to help develop it. Often the tiff format is used for the high resolution and a derivative image is available on the web. JPEG2000 has multiple resolutions, region extraction, lossless and lossy compression, and display can start without waiting for the entire file to be loaded. djatoka improves the performance, supports many formats, manipulation of the image (such as watermarking), and works with Open URL.