The Schroder Collection has just been added to the Cambridge Digital Library, in a new section for King’s College. This had been the largest private collection of Rupert Brooke papers, until it was acquired by King’s College on the centenary of Brooke’s death (23 April 2015). Details of the acquisition can be read on the King’s Treasures blog, or in the Cambridge University website. The King’s Treasures blog also has a new post on the Schroder papers online resource.
Following on from previous posts about the Oliver Rackham Project and the Introduction to the Digital Content Unit and Cambridge Digital Library at the UL, and remembering this blog’s audience, I will use this post to comment on the creation of the resource. I hope other archivists will find it useful.
The purchase of the Schroder Collection, which cost £500,000, had only been possible due to generous grants from the National Heritage Memorial Fund and the Friends of the National Libraries, along with other private donations. Similarly, the creation of this online resource was only possible as a result of a further grant from the Friends of the National Libraries, who can sometimes fund the digitisation of papers they have helped institutions to purchase.
When preparing the grant application, I had to consider various options. These included different ways of digitising the archives (Cambridge University Library’s Digital Content Unit or a private company such as Townsweb – we quickly discarded the idea of doing it in-house, as we didn’t have the resources) and different platforms for the resource (Cambridge Digital Library or something created in-house, possibly using commercially available templates). Having explored all of our options, we decided to opt for both the Cambridge University Library’s Digital Content Unit and the Cambridge Digital Library. This was not the cheapest option available but we felt that for this particular project the benefits of their proximity and levels of insurance, as well as the sustainability of such a large platform, justified the extra expense. We also felt that the continuity of keeping both elements of the project within the University Library would prove more efficient and shorten the project. Another major consideration in our decision-making process was that the papers of Siegfried Sassoon were already on the Cambridge Digital Library, meaning those with an interest in war poets could find both on the same platform. We also had to select how much of the collection to digitise and what our priorities were. We selected approximately half of the collection, focussing primarily on guard books containing letters by Rupert Brooke and his friends Edward Marsh and William Denis Browne , because that would be best ‘value for money’ in terms of interest-payoff for the work required clearing copyrights.
Once the grant was received, there were four phases:
Phase 1 – Metadata creation
Prior to the project, the Schroder papers were catalogued to guard book level, so we quickly recruited four volunteers to create metadata at the level of individual letters. They followed guidelines I had created to ensure consistency, then checked each other’s work (I checked it too). I am very grateful to those volunteers: Mandy Marvin, Harriet Alder, Maddie McDonagh, and Thelma May.
Phase 2 – Digitisation
After I liaised with Maciej Pawlikowski (Digital Content Unit) to arrange this, his colleague Błażej Mikuła created approximately 1402 high quality tiff image files. The images were faithful digital reproductions in 600 ppi in 1:1 of the original in very consistent manner in colour managed environment using high end / bespoke photographic equipment. Błażej has 10 years of experience in digitisation of the most fragile collections and worked on projects such as Parker on the Web and Darwin Manuscripts. He also digitised the entire collection of Newton’s papers held on Cambridge Digital Library. His experience and the facilities available to him gave us much confidence from the outset that the records were in safe hands and the images would be of the best possible standard.
The images were passed on to their colleagues in the Cambridge Digital Library, in addition to me receiving copies.
Phase 3 – Creation of the resource
Huw Jones (Cambridge Digital Library) then matched the images to the metadata I supplied (in EAD format) and created a test site. We had to carefully plan things such as how to ensure there was the correct hierarchy (we decided on two ‘collection pages’ – King’s College and Schroder Collection). Each guard book was considered an ‘item’ (there was some discussion of terminology like this – the word meant something subtly different in this context compared to an archival catalogue) and the matching of lower-level metadata would have been made more complicated by the fact that each letter could vary from one image to around a dozen.
Phase 4 – Quality assurance
I was given login details to the test site and was asked to check it, proof reading the entries again and ensuring that the images and metadata were correct. There were a surprisingly small number of changes required.
Concurrent with all of these phases, I had been clearing copyright on any items not created by Brooke, Marsh or Denis Browne (all of their copyright had been sought at the planning stage). This included applying for orphan works licences. I created a spreadsheet to track this and during phase 4 I supplied Huw with notes on copyright exemptions, with certain items requiring specific notices or place-holders.
From my perspective, the process was relatively easy, due to good communication and planning. It is easy to see potential pitfalls with digitisation projects though, if one doesn’t spend enough time preparing. While there are other ways of achieving a project like this, I can recommend the Cambridge University Library’s Digital Content Unit and Cambridge Digital Library. I am very grateful to them for all of their hard work.
Assistant Archivist, King’s College