In its report on lessons from the First World War commemorations in 2014-18, the House of Commons’ Digital, Culture, Media and Sport Committee was pleased to hear that DCMS would fund the preservation of the resulting digital legacy. ‘However,’ the report notes dryly, ‘it is unfortunate that the need for this was not foreseen at the start of the commemorations. Given that the DCMS leads on digital policy, a strategic approach to preserving digital assets should form part of initial planning of any future government-funded arts or heritage programmes.’
The need for a better approach goes well beyond big-ticket programmes like the WWI commemorations. In our recent report Getting it together, supported by the Open Data Institute’s Stimulus Fund, we propose a digital preservation strategy for the whole UK museum sector. Its scope would extend beyond the obvious assets resulting from digitisation projects to also cover a wide range of outputs (such as exhibition text) that too often gather digital dust in long-forgotten spreadsheets and documents.
We see scope for multiple providers of digital preservation solutions, each meeting recognised standards such as CoreSealTrust, an external validation process that demonstrates compliance with requirements drawn up by a working group of the Research Data Alliance. Some of these providers might be standalone services, others based within existing cultural heritage or academic institutions with spare capacity to sell.
The question is how to pay for the level of digital preservation the museum sector needs. For once, the problem is not really a financial one: museums around the country already spend money each year on ad hoc digital storage arrangements, while many millions in grant-funding continues to be available for projects that create new digital assets of various kinds.
Economies of scale
For legacy assets, our proposal is that museums should combine the purchasing power that is currently spread between countless small-scale solutions. Museums are already spending money on digital storage; by coordinating their procurement, it seems highly likely they could get better, proactively managed digital preservation for the same or less.
Digital collections can deteriorate just as much as physical ones, and need the same monitoring and prompt attention at the first sign that something is wrong. A certified digital repository provides this level of service, just like keeping your physical collections in a carefully monitored environment where they are regularly checked by a conservator. (For a good, illustrated introduction to one such check, ‘file fixity’, see this piece by Jefferson Bailey on the Library of Congress blog.)
There is an environmental dimension to the problem too. As a recent Julie’s Bicycle webinar reminds us, if the internet were a country, it would be the sixth largest polluter in terms of greenhouse gas emissions. The huge data centres that lie behind all online activity vary enormously in their energy efficiency and their dependence on power from different sources. By pooling the needs of the whole museum sector, we could procure greener digital storage too. And if the chosen service providers supported the International Image Interoperability Framework, we wouldn’t need all the working copies of hi-res images that currently proliferate across our systems.
For newly created assets resulting from grant-funded projects, we propose micro-endowment, a tried-and-tested model for buying long-term digital preservation. The main funders of museum digital projects could easily require grant recipients to allocate enough budget to endow the long-term preservation of the outputs in a trustworthy digital repository.
The University of York’s Archaeology Data Service (ADS) provides a compelling case study. ADS preserves the digital archives created during archaeological fieldwork and charges a one-off fee up front that goes into an endowment fund, income from which is part of a mixed business model that also includes participating in a range of research projects. It has been certified as a trustworthy data repository; the most recent assessment of its service is published on the CoreSealTrust website and includes a great deal of detailed information about how ADS meets the sixteen certification requirements.
The US-based Permanent Legacy Foundation, also has an endowment-based model, influenced by the work of David Rosenthal. In short, the model estimates the future annual costs of storing a of digital archive of a given size, and also estimates the amount of money that would be needed to generate enough annual interest to meet those costs. The current calculation is around $10 per GB (gigabyte) of storage. Users are charged a one-off fee depending on the size of the archive they wish to upload, and the Foundation adds this lump sum to its endowment investments. The annual interest generated covers the cost of the multi-cloud storage negotiated with commercial providers, and also the Foundation’s operating expenses, such as data migration, integrity checking and format conversion. The Foundation is working towards certification as a trusted digital repository as defined by the international standard ISO 16363:2012.