Andy Jackson is a Digital Preservation Specialist at the DPC.
I started to learn about digital preservation when I joined the British Library in 2007, during the early stages of the PLANETS Project. At the time, it seemed that taking care of digital collections was all about two big problems: keeping the bytes safe, and fighting off format obsolescence. Our preservation plans were built on the assumption that the data was safe for now, and we could keep our digital stories safe by migrating and emulating our way into the future.
We were wrong.
Yes, it's a problem if you have files that depend on software that is no longer profitable. And yes, managing big piles of digital files is harder than it looks. But projects like PLANETS showed that these technical challenges can be overcome as long as there is someone around who has the time and resources to care for the collection. That's the biggest risk. That there is no one left to care.
From the slow decay of austerity to the deliberate destruction of public resources to outright war, the biggest losses have come from the 'resources' and 'organisation' legs of the three-legged stool that digital preservation depends on.
In some cases, the sheer size and impact of these actions have led to a robust response. The Saving Ukrainian Cultural Heritage Online (SUCHO) Project proved that enough people outside Ukraine care about Ukrainian culture to crystallise an international response. Similarly, the Data Rescue Project and Safeguarding Research & Culture are actively responding to the large-scale loss of public data and the current risks to our cultural, intellectual, and scientific heritage.
But the smaller, less visible collections are harder to track.
The consequences of the large-scale cuts to USAID, repeated cuts to the UK’s aid budget, and the wider defunding of the aid sector are distributed across many organisations. Most aid organisations operate locally, at smaller scales, while managing delicate datasets and sensitive records that cannot be made public, and so lack the visibility of scientific datasets and cultural works. These digital collections document the lives of people in crisis, along with the actions of the organisations that come to their aid. If we cannot save the archives of these people and organisations, those individuals lose their history, and those organisations (and indeed all of us!) lose the ability to learn from their past actions and evolve.
These small organizations are being forced to make difficult decisions. Delivering aid must come first, but without adequate data collection and recordkeeping it will be impossible to work out how effective the different interventions are. This is all exacerbated by the need for speed and convenience forcing organisations to adopt dominant digital platforms that actively prevent data export and archiving.
"Never have the memories of so many been in the hands of so few"
- Vendor lock-in and the hijacking of history - Patricia Sleeman
The Humanitarian Archive Emergency (HAE) project has been working to understand and document this crisis. As well as detailed reports on the state of play, they have started to build a registry of humanitarian records, archives, and databases, and a Digital Humanitarian Archives Toolkit aiming to help those working in the field.
Having read the report, one thing I wonder is how much technology we might be able to re-use across different disasters. For example, I was very excited to see the SHIELD / SciOp – TIB – Leibniz Information Centre for Science and Technology project, building distributed infrastructure to secure the future of public research. Could we do the same for humanitarian archives? Is it possible to build the need for individual privacy and protection into a similar system without re-inventing all the wheels?
For me, World Digital Preservation Day is not just about advocacy. It's also about building connections. So if you're reading this and either need help with your humanitarian data, or can offer your skills or other resources to attack this problem, please let us know.