Staff Papers

Edmund Barton Digitisation Project

A paper by Paula Waring, Manuscript Section, National Library of Australia


The National Library’s Edmund Barton Digitisation Project commenced in early 2000 with a plan to select a manuscript collection for digitisation. As 2001 marked the centenary of Federation, it seemed appropriate to digitise the collection of our first prime minister, Sir Edmund Barton. The Federation relevance of the project attracted a $50,000 donation by Dick and Pip Smith, which provided partial funding.

The National Library’s digitisation program to date has largely concentrated on pictorial material, now mounted on the PictureAustralia database (www.pictureaustralia.org). In 1999 a digitised version of James Cook’s Endeavour journal (MS 1) was released on a CD ROM together with J.C. Beaglehole’s transcript. In 1999 William Wills’ diary (MS 30) of his ill-fated expedition with Robert O’Hara Bourke was also digitised for preservation and exhibition purposes (although it is yet to be made publicly available) and in 2000 the small but significant George Raper collection (MS 9433) of First Fleet manuscripts was mounted for online exhibition (http://www.nla.gov.au/exhibitions/georgeraper).

The Barton Papers (MS 51) are the first comprehensive archive of personal papers to be digitised at the National Library. The papers comprise 1.12 shelf metres of material and include correspondence, diaries, personal documents, speeches, articles, newspaper cuttings, programs, menus, photographs, printed material and several objects including a Royal Doulton plate, a gold medallion and a ceramic vase. The most heavily researched part of the collection is series 6, which contains papers relating to the Federation Campaign, including drafts of the Constitution annotated by Barton. There is also material relating to Barton’s business and family affairs and his position as a Justice of the High Court. Barton’s correspondence also attracts considerable research attention and includes letters by Alfred Deakin, Sir John Forrest, Lord Hopetoun, Joseph Chamberlain, Sir Samuel Griffith, Sir William Lyne, Sir Henry Parkes and Lord Tennyson.

The Papers were donated to the National Library by the Barton family in six instalments between 1929 and 1996. Family papers and papers relating to New South Wales politics from the 1929-55 consignments were separated and sent to the Mitchell Library. When the material was transferred to the present National Library building in 1968, a guide to the papers was created. This finding aid is typical of many created at the time. Individual items (such as letters, diaries etc) were item numbered (albeit in a very idiosyncratic manner) and each piece of correspondence was indexed by author and recipient. The guide was scant on series-level description, but within the more important series, most of the items were described individually. This contrasts with our current practice of describing modern manuscript collections largely at the folder level without item numbering or indexing.

When the decision was taken to digitise the Barton Papers, no full-time project staff was engaged. Any work on the project within the Manuscript Section was to be undertaken within its current staffing levels. For this reason, we wanted to make minimal changes to the finding aid or the collection. Initially some work was undertaken to expand further series descriptions to item level. This process revealed many curious features of the arrangement of the collection that had hitherto been hidden by the more generalised description. After many meetings we were prevailed upon to expand the entire collection to item level description in order to accommodate permanent identifiers. Like most manuscript collections, the Barton Papers range from the sublime to the ridiculous, and we soon found ourselves writing captions for illegible scrawl on torn pieces of paper.

The most ambitious aspect of the project has been the effort expended on providing navigation tools for the online collection, and in particular, the relationship between the images and metadata. The completed finding aid was converted to Encoded Archival Description (EAD) and provides the structure of the site, with hypertext links from the item numbers to the digital objects. We have been unable to find another example of access to a digitised manuscript collection through a finding aid. If anyone knows of a site where this is being done, we would like to hear from you. If we have been reinventing the wheel, we would very much like to compare wheels.

As the project gathered momentum it required input from not only Manuscripts, but also the Preservation Services, Information Technology, Web Services, Multimedia and Imaging Services sections in the Library. This subjected the finding aid and our finding aid practices to greater outside scrutiny than they had received before. The multimedia people did not know what to make of the term ‘series’ and were in favour of devising another term that would mean more to the general public. We held firm. Concerns were raised that separate item numbering was not given to each side of the paper and to the unfortunate existence of a’s, b’s and c’s inserted into the numbered sequence. Further anomalies were uncovered when the correspondence index was manipulated to create an item list of correspondence. Days were spent editing indexing, filing and name errors that had now become glaring after the rearrangement of the data.

One of the risks involved with choosing a twentieth century collection was copyright. While copies of the Barton Papers could be made available to researchers for reference purposes, digitisation of the papers is a form of publication, and under the terms of the Copyright Act we were bound to seek permission from all the copyright owners. Permission was sought from the Barton family at the outset of the project, but contacting heirs for each of the 400 or so correspondents was impossible. A regulation under the Copyright Act states that where the copyright owner is not known, a notice of intention to publish can be placed in the Government gazette. We published all 400 names in the Gazette, but (not surprisingly) have not received any protests from indignant descendents claiming copyright.

The papers were scanned by a contractor based in the Library using a flatbed A3 scanner and an overhead digital camera for larger items. The images were reproduced at a resolution of 300 dpi (at original size), 8 bit colour (24bit) RGB. Longer items, such as thin strips of newspaper cuttings larger than an A3 page, were scanned in segments and stitched together. The scanning process in most cases was successful, although some items, particularly newspaper cuttings with small print, were difficult to read on even the greatest enlargement and needed to be rescanned. Some items were missed during scanning and the checking process was labour intensive and tedious.

The digitised Barton Papers are a showcase collection that enabled the Library to find solutions to problems with storing and presenting scanned images from manuscript collections. The project resulted in the Library acquiring new technical skills and equipment. It has also been extremely expensive. While no firm figures for the cost have been calculated, it involved input from a wide range of senior staff over more than 18 months.

The Barton Digitisation Project raises many issues for future manuscript digitisation projects. Compared to the majority of our holdings, the Barton Papers are a small collection controlled to a detailed level. When digitised, it resulted in a mere 5500 images (compared to the 40 million items estimated to comprise our total holdings). Much of the Library’s psychology on digitisation has been shaped by its history of providing access to individually catalogued pictorial images. This model does not transfer happily to the large unwieldy nature of most modern manuscript collections. Describing large manuscript collections at item level is simply not feasible. If we wish to have any serious impact on the delivery of archival collections digitally, we will need to reduce significantly both the time spent on the creation of navigation tools and our standards for the level of description.

Despite the difficulties, digitising the Barton Papers has been a successful project and results from trials with a select user group have been encouraging. The National Library has an ambitious digitisation project in place for this year which includes digitising more manuscript collections. Our next task is to review the valuable lessons learned from the Barton Project to ensure that future projects are less costly and protracted.

Thanks to a recent television advertising campaign, most Australians now know who Sir Edmund Barton is. His papers will now be coming soon to a computer terminal near you.

At the time of writing, the Barton Papers web site was not available publicly. An announcement will be made in the aus-archivists listserv when the URL is known.

Back to the top