A career in (the) ruins

Digital preservation was an interesting vocational choice for me. After an early career in archaeology, it seemed I was unable to escape from what is jokingly described as a ‘career in ruins’. Even as I swapped my kneeling mat and trowel for a comfy office chair and a floppy disk drive the concepts of brokenness and uncertainty have been a constant companion.

Archaeology is all about piecing together and translating partial information from the past, whilst digital preservation is all about finding a way to transmit information to the future. Both disciplines are tinged with an element of failure. As archaeologists, how do we know if we’ve interpreted those messages from the past correctly? As digital archivists, how do we know whether people in the future will find, read and understand the digital messages we are preserving for them today? When you work in a career full of uncertainties, unknowns and speculation, some level of failure is to be expected, and even celebrated.

These two disciplines do collide and interface when we consider the term ‘digital archaeology’, which can be defined as ‘rescuing neglected and damaged data resources’ (Ross and Gow 1999), and the process of reclaiming digital information that has been damaged or become unusable due to technological obsolescence of formats and media.¹

Of course, ‘digital archaeology’ is not the only thing that digital archivists do on a day-to-day basis. It forms a very small part of the work that we undertake. When digital preservation works well, the interventions and preservation activities happen early on. We ‘fix’ things before they become broken. We take actions to collect, describe, manage and maintain the content whilst it is still current and usable. We carry out further activities to assess risks and plan preservation actions before they become an urgent necessity.

Change is inevitable and none of us can foresee the future. We make decisions about the preservation of digital content without the benefit of a crystal ball and with no way of being sure that our decisions were the best ones. As a result, uncertainty is our constant companion and the answer to nearly every question about digital preservation is ‘it depends’. There is no one-size-fits-all solution to preserve our digital archives and, invariably, the details and specifics of each scenario matter (Owens 2018).

Good, better, best …

It surprises me when I see the term ‘best practice’ coming up so often in our line of work. Given the uncertainty that is typically present around what we do and how we do it, the trend to strive for ‘best practice’ feels neither healthy nor realistic. Do we know what ‘best practice’ is in the context of our work? Are we setting ourselves up for greater failure by aiming for this? It is particularly hard to claim that anything we do in this line of work could be considered ‘best practice’ when our workflows and strategies are ever evolving, and we have limited options to evidence our success. How can we know if preservation strategies and actions implemented today have been successful? For many, including those who are just getting started, those with limited resources and support, and those who are trying to find innovative solutions in new areas, ‘good enough’ is a far more achievable aim. Finding a workable solution and documenting the imperfections and opportunities for improvement doesn’t necessarily align with the ‘best practice’ rhetoric but is often a more realistic way of working.²

Accepting imperfection

Consider a ‘digital archaeology’ scenario I faced several years ago in a previous role at the Borthwick Institute for Archives. I was given the task of preserving a series of screenplays that had been written using a legacy word processing package from the early 1990s and stored for many years on 5¼ inch floppy disks. Aside from the issue of recovering the data from an obsolete media format, and establishing which software the files had been created in, the challenge remained to preserve the information in a format that could be read and understood by users who wanted to access those screenplays both now and into the future.

This task brought with it many layers of imperfection and brokenness. Firstly, some of those original floppy disks could no longer be read. Secondly, experiments with file format conversion were not able to preserve every significant feature of the original WordStar files – a compromise ultimately had to be reached. Thirdly, the scenario became more challenging on discovering that some of the files in this collection were already damaged, having suffered from localised corruption at some point in their lifecycle. Having no route to go back and ‘fix’ this corruption, attention turned to how to record and document the issue. As I worked through this set of challenges I blogged about my progress, noting these observations and flagging up the many imperfections of this work (Mitcham 2018).

A contrasting digital preservation case study concerned more modern documents in native Google Document format (Micham 2017). Though exporting these ‘documents’ for archive tended to create an accurate copy of the words on the page, other elements were compromised. Collaborative features of the document such as comments (often a key feature of records created in this platform) were often not replicated well, and key document properties visible within Google Drive (for example, owner, creation date, last modified date) were not consistently retained. Other features of the original, such as the full revision history of the document, are also lost on export. Given there was no perfect preservation solution for these types of document, a compromise had to be reached, and the methodology and its limitations documented.

Though these early experiments could not be described as a complete success, the very process of working through and highlighting the problem areas was helpful. Indeed, subsequent work has built on these initial experiments and the conversation has moved forward in a productive way.³

No one likes to fail, but I see neither of these examples as a low point in my career. When working with constant change, complex problems and the lack of a single, perfect solution, we need to take a pragmatic and realistic approach, acknowledging and documenting the imperfections that come alongside this. ‘Best practice’ this surely isn’t, but the ability to experiment and to share points of failure with the wider community can be incredibly valuable in moving our collective thinking forward.

Admitting failure

The Digital Preservation Coalition (DPC) provides a supportive network for a global community of organisations who have a need to preserve digital content for the long term. Many of our events facilitate the sharing of expert knowledge and provide opportunities for members to talk about the innovative work they are involved in – whether it was ultimately successful or not. It is so much harder to talk about failure, but many would agree that it is equally valuable.

With this in mind, we established a series of networking events with a specific goal to discuss and share failures. ‘Digital Preservationists Anonymous’ (also known as ‘Fail Club’) was set up as a support group to ‘reflect on things that have not gone as well as we liked’ with a goal to be ‘liberating, cathartic and therapeutic’. The first meeting was held in 2017, and it has been a repeating event on our programme ever since (Middleton 2017).

I attended my first ‘Fail Club’ in 2018 before I joined the DPC staff team. A small group of us, led by a DPC facilitator, huddled in a top room of a pub in York, and with the door firmly shut, and in lowered tones, I was able to talk to a small group of peers about a digital preservation advocacy failure that I was experiencing. It was a cathartic experience and one that I was able to move forward from in a positive way (having realised that most of the people round the table related to the issue I was having and were able to share their own strategies to move forward).

I’ve since been involved in several of these events in the role of facilitator and it always turns out to be a rewarding experience. The conversation naturally flows once we provide a designated space to open up about failure. I can’t say any more because (of course) ‘what happens in Fail Club, stays in Fail Club’, but comments received from attendees refer to it as a ‘safe and enriching opportunity to share professional frustrations, mistakes and fears’ and ‘a great forum to chat about difficulties I’ve faced at work that I’ve not found solutions for’.⁴

Encouraging and supporting conversations around failure should happen by default at any professional forum or event. As well as ‘Fail Club’, the DPC facilitate member-only events where speakers and attendees are given a safe space to talk openly about the challenges. Though these are more often online and without a physical door to push shut, we build an atmosphere of safety, ensuring our inclusion and diversity policy is referenced and understood, starting with a friendly round of introductions, reassuring attendees that the meeting isn’t being recorded and applying Chatham House Rules where appropriate.⁵

Success from failure

We learn a lot from failure and it is often a stepping stone to success. Even our ‘Fail Club’ events end up with a focus on positive steps to move forward (so much so that we recently debated rebranding them with a more positive name). Skills that are required to be successful in digital preservation include an ability to experiment, to try things out, to record and analyse results, to be resilient in the face of failure. Curiosity is a key attribute for success – the mindset that thinks ‘I wonder what would happen if I tried this …’ is encouraged. Working in this way undoubtedly leads to an element of failure, but if you don’t try, you will never have a chance of succeeding.

The recent focus within the library and archives sector on a cyber-attack at the British Library has been on the negative effects of the incident and the impact on the organisation’s ability to continue to run key services, but there remain glimmers of hope for the digital preservation community at least (Keating 2024). What would have happened if robust digital preservation processes were not in place at the British Library? A blog post from Andy Jackson (who was working at the British Library at the time of the attack) describes how a project to replicate the UK Web Archive ‘went from being a cumbersome mitigation of a theoretical risk, to an absolutely crucial part of an active disaster recovery plan’ (Jackson 2024). Is this an extreme example of what a digital preservation success might look like in the presence of catastrophe? Perhaps. But simply by talking about what went wrong and what could have been done differently, the lessons learned can be shared widely. As noted by Roly Keating (CEO of the British Library) ‘some good will have come from this dreadful incident’.⁶

Last words

Since leaving archaeology my career isn’t exactly in ruins, but failure has been a consistent, and not entirely unwelcome, companion. Working in an evolving field with success measures that are far into the future we need to leverage failure to our best advantage. By failing we learn how to move forward, and in talking about failure we share that learning. Curiosity and resilience in the face of failures are key qualities for practitioners in this field. With a focus on continuous improvement and by striving to meet ‘good practice’ rather than ‘best’ we can set realistic targets and goals. Understanding and documenting the elements of brokenness and imperfection in our work is digital preservation good practice and is something that should be celebrated.

Notes

1. 1 The term ‘digital archaeology’ is also used to refer to the use of digital technologies to support the discipline of archaeology. See for example ‘Digital Archaeology’, Wikipedia, last modified 22 March 2025, https://en.wikipedia.org/wiki/Digital_archaeology (accessed 25 November 2024).

2. 2 A longer discussion on the topic of good practice versus best can be found here: ‘When is “Good” Better than “Best”? In Support of Digital Preservation Good Practice’, Jenny Mitcham, last modified 18 April 2023, https://www.dpconline.org/blog/when-is-good-better-than-best-in-support-of-digital-preservation-good-practice (accessed 25 November 2024).

3. 3 For example: ‘What’s Up, (with Google) Docs? – The Challenge of Native Cloud Formats’, Paul Young, last modified 4 March 2021, https://www.dpconline.org/blog/whats-up-with-google-docs (accessed 25 November 2024).

4. 4 ‘Digital Preservationists Anonymous’, Digital Preservation Coalition, https://www.dpconline.org/events/eventdetail/218/-/digital-preservationists-anonymous (accessed 25 November 2024).

5. 5 ‘Inclusion and Diversity Policy’, Digital Preservation Coalition, August 2021, https://www.dpconline.org/docs/about-1/dpc-policies-and-forms/1807-dpc-inclusion-and-diversity-policy/file (accessed 25 November 2024).

6. 6 Keating quoted in Connor Jones, ‘British Library’s candid ransomware comms driven by “emotional intelligence”’, 20 May 2024. https://www.theregister.com/2024/05/20/the_british_library_owes_lauded/ (accessed 25 November 2024).

References

• Jackson, Andy. ‘Grasping a Petabyte: The Impact of Replicating the UK Web Archive’. 1 July 2024. Accessed 24 November 2024. https://anjackson.net/2024/07/01/grasping-a-petabyte/.
• Keating, Roly. ‘Learning Lessons from the Cyber-attack’. 8 March 2024. Accessed 24 November 2024. https://blogs.bl.uk/living-knowledge/2024/03/learning-lessons-from-the-cyber-attack.html.
• Middleton, Sarah. ‘Losing All Hope to Find Freedom: Fail Club Is Here to Help’. Last modified 3 August 2017. Accessed 24 November 2024. https://www.dpconline.org/blog/fail-club-to-help.
• Mitcham, Jenny. ‘How Can We Preserve Google Documents?’ 28 April 2017. Accessed 24 November 2024. https://digital-archiving.blogspot.com/search?q=how%20can%20we%20preserve%20google%20documents.
• Mitcham, Jenny. ‘An Imperfect Migration Story’. 8 June 2018. Accessed 24 November 2024. https://digital-archiving.blogspot.com/2018/06/an-imperfect-migration-story_34.html.
• Owens, Trevor. The Theory and Craft of Digital Preservation. Johns Hopkins University Press, 2018.
• Ross, Seamus and Ann Gow. Digital Archaeology: Rescuing Neglected and Damaged Data Resources: A JISC/NPO Study within the Electronic Libraries (eLib) Programme on the preservation of Electronic Materials. Library Information Technology Centre, 1999. Accessed 24 November 2024. https://purehost.bath.ac.uk/ws/portalfiles/portal/11350174/p2.pdf.