Edmund Barton digitisation project

The Edmund Barton digitisation project was made possible through the kind support of Mr Dick Smith and Mrs Pip Smith.

This document descibes the methods and technologies used to capture, store and deliver access to items from the Barton collection.

Capture and derivative creation

The Barton papers were digitised using a UMAX Powerlook 2100XL scanner and Sinar Macroscan 23 and PhaseOne Power FX digital cameras for the oversize items. Over 5,000 single page images were captured of items from the Barton papers. The master image files are held in 24 bit RGB TIFF revision 6.0 format. The following metadata fields are stored in the TIFF file header:

  1. Digital Image ID (unique persistent identifier)
  2. Colour Management System used
  3. Hardware used to capture master image
  4. Software name and version used to create master image
  5. Field to indicate if the image has been stitched
  6. Image description

For items captured using a digital camera, the master images include a colour patch to make it easier for imaging and pre-press professionals to adjust images for specific output. Colour patches are not being used for images captured with help of scanners due to difficulties acommodating them on a relatively small scanning bed, together with the originals. For items captured using a digital camera and for poorly preserved items, a co- master was also created, which was cropped or enhanced, and used as the source for derivatives.

Derivatives of the master image files were created for delivery over the Internet. For each image, three derivative images were created, a thumbnail (150 pixels), a view copy (760 pixels) and an examination copy (1000 pixels). View copies from Series 12 (photographs) are 600 pixels to conform to Library standards for pictorial derivatives. For some types of material, eg newspaper cuttings with type of a small point size, the view and examination copies were too small and were re-created at 1000 pixels (view copy) and 1600 pixels (examination copy) to increase legibility. All pixel dimensions are the length of the maximum side of the image.

Persistent Identifiers

The identifiers used for items from the collection that the Library is making available online are persistent identifiers (PIs). This means that the Library has undertaken to continue to use the identifier unchanged in future to describe the item from its collection, including both the physical item and its digital surrogates. For example, the identifier for the Papers of Sir Edmund Barton is nla.ms-ms51. Persistent identifiers provide a reliable and stable way of naming items from the collection and these identifiers can be used in citations both online and in a print publication. More information on this issue can be found in the document Managing Web Resources for Persistent Access.

A persistent identifier scheme has been developed for all digital copies of collection items and for web sites saved in the PANDORA archive. The persistent identifier scheme varies depending on the type of collection material, eg manuscript, pictorial, maps, music etc, and generally is based on existing numbers or identifiers used to manage the physical collection.

The PI scheme adopted for manuscript material is based on a hierarchy of collection, series, item/folder, and part that reflects the way the physical collection is stored and managed. An example of a persistent identifier for the first part of item 1 of series 1 of the Papers of Sir Edmund Barton (MS51) is: nla.ms-ms51-1-1-s1

nla.ms PI prefix for manuscripts
ms51 Unique number for the manuscript collection
1 First level of the hierarchy within the collection - generally the series within the collection
1 Second level of the hierarchy within the collection - generally the item or folder within the series
s1 The first part of the item or folder (this is a sequence number)
-v or -e If the prefix also has -v or -e following at the end of the persistent identifier, then the persistent identifier is pointing to the view copy of the part (-v) or to the examination copy of the part (-e).

The Library has also developed an online resolver service that will redirect a request for an item to the current storage location of the item. For example, the web address for the Papers of Sir Edmund Barton is http://nla.gov.au/nla.ms-ms51 and if this identifier is used in the browser address or location box, the digital copy of the Papers of Sir Edmund Barton will be found no matter where on the Library's website these are currently located. Similarly, the web address for the first page of item 1 of series 1 for the Barton papers is http://nla.gov.au/nla.ms-ms51-1-1-s1.

Users of this web site are encouraged to use the persistent identifier and the Library's online resolver service when citing parts of the collection. Citations can be made at any level, that is at the collection, series, item or part levels.

Storage

Image files are stored on a dedicated file server. Directory locations and filenames of images are based on the persistent identifier. Directory structures are automatically created to limit the number of files per directory to less than 2,000. A directory scheme has been created for each type of material (eg pictures, maps, music etc) so it is possible to automatically derive the location and filename of a digital object based on its persistent identifier.

Access

The delivery system is controlled by a finding aid for the Barton collection which conforms to the EAD (Encoding Archival Description) XML DTD maintained by the Library of Congress in association with the Society of American Archivists which is beginning to be used extensively in Australia for the description of manuscript collections.

A delivery system has been created to provide access to the Barton collection over the Internet. The delivery system makes use of technology developed by the University of California, Berkeley Library as part of the Making of America II project.

The web pages that make up the delivery system are automatically generated from information held in the EAD finding aid. Pages are either static or dynamically created. The dynamic pages come from an MOA2 Java servlet. The servlet reads XML files on start up. The XML files conform to the MOA2 XML DTD and contain both descriptive metadata and a structural map which relates image files together that make up each item in the Barton collection. The servlet uses these XML files to return appropriate metadata to the web user and also to locate the image files on the filesystem.

The following diagram shows the architecture of the Barton delivery system.

Architcture of Barton Delivery System