We’re often asked at the British Library whether we have preservation plans for our digital collections. Because that’s what digital preservation is all about, right? Preservation plans are a fundamental part of the preservation toolbox that enable us to preserve our digital collections over time and minimise exposure to risks that would otherwise threaten their longevity. It’s obvious therefore that all collections should have preservation plans. Right? Well, yes. And no. Maybe… it depends. In short, it’s complicated. 

The answer, if you’re interested, at least for us at the British Library, is no – we don’t have preservation plans in place for all of our digital collections. But it’s not because we don’t think we need them. It’s more that we don’t think we need them *yet*. We’ve been doing lots of work on preservation planning over the years, developing our underlying knowledge about the collections for example via our digital collection profiles and format sustainability assessments. Most recently, our attention has turned to practical, technical preservation plans, specifically as part of our work on the Integrated Preservation Suite project, IPS. IPS is developing the infrastructure that we believe we will need to deliver preservation planning at scale, for all of our digital collections. We’ve written about IPS elsewhere so I won’t go into that in detail here (besides which, Peter is planning a separate blogpost with an IPS update). But as we realised fairly early on, people have different ideas about what a preservation planning process entails, and when it needs to be done. 

Part of the reason for this, we suspect, comes from the fact that whilst preservation planning is a widely accepted concept in the digital preservation community, it is relatively poorly defined from a process perspective. Preservation planning is one of the six major functions of the Open Archival Information System (OAIS) reference model, otherwise known as ISO 14721. OAIS defines preservation planning as one of six functional entities that ‘provides the services and functions for monitoring the environment of the OAIS and which provides recommendations and preservation plans to ensure that the information stored in the OAIS remains accessible to, and understandable by, and sufficiently usable by, the Designated Community over the Long Term, even if the original computing environment becomes obsolete’. It incorporates a very wide range of activities that are extensively described in the reference model. The matter of implementation is however, open to interpretation, and whilst the relationships between sub-functions in the OAIS model are well defined, the process and in particular the format of a preservation plan is not. So what does that mean for what we understand as a ‘preservation plan’?

The most widely known preservation plan format within the community is probably that associated with the preservation planning tool PLATO, developed by the PLANETS project. The PLANETS project defined a preservation plan as ‘a series of actions to be taken by a responsible institution due to an identified risk for a given set of digital objects or records’. PLATO, broadly speaking, is a ‘decision support tool’ that guides the user through this planning process, whereby requirements are identified, preservation options identified (for example source and target formats, and conversion tools), and outputs of each option assessed against a set of success criteria, leading to a decision on which specific option should be implemented more widely. PLATO is complicated though, and it isn’t widely used in the community. The online PLATO tool, judging by the failure of the ‘enter Plato’ link at the page above, also appears to be no longer available. 

A small number of simpler, but similar preservation plan formats have been developed at individual institutions, including BnF, NLNZ, and Beeld en Geluid. All of these propose or utilise perfectly valid approaches to preservation planning. But, preservation planning is still not a widely implemented concept. As Yvonne Tunnat observed in 2017, ‘we still don’t do preservation planning in our productive system. More interestingly, nobody seems to be surprised or shocked about it. People I talk to usually do not even consider doing preservation planning in the near future.’ This has changed a little recently, not only through Yvonne’s update to this blog a few months back but also with the release of NARA’s ‘preservation action’ plans (though the distinction between a preservation plan and a preservation action plan warrants more discussion, at some point). The NARA plans take a different form from the PLATO-type approach described above and are based on the outcome of a format risk exercise and subsequent ranking, rather than practical testing with specific tools. Nonetheless, in this way NARA therefore does have preservation plans for all of the formats in its collection. But many other institutions don’t. 

So where does that leave us, and in particular, IPS? Our approach is that we only need to develop preservation plans in response to specific triggers. This is broadly in line with the monitoring element of the preservation planning function within OAIS (I hesitate to use the word ‘compliant’ because, as a friend said, anything can be OAIS compliant, even a fridge, but you get the general idea).

Our trigger function is what leads to the production of a preservation plan. What might those triggers be? Well, a good example would be the release of a new version of a format with features that our current rendering software can’t support. How might we pick up on that? There’s a few options, and monitoring is the key to all of them. The first level of monitoring is probably our format sustainability assessments – these would alert us in the first instance to format changes that might need investigation. The second level is monitoring the formats – and ideally versions of formats – that we ingest into our repository, which would notify us of the introduction of this issue to our collection. The tertiary level – though ideally we would catch it before this point – would be monitoring of problems reported by users or reading rooms staff. All of these can be automated to a degree, and that’s what we will seek to do either within IPS or related work. There are other triggers too that we’ll be working on over the course of the next year. Our preservation plan format will be simpler than that in PLATO, but still capture key information that we need to know to evaluate different preservation options and make a decision, based on the information in the IPS knowledge base. 

In the meantime, we continue with a whole range of other preservation planning activities of the sort we’ve mostly described elsewhere, everything from risk assessments to digital collection profiling, policy and strategy development, and investigation into new types of content like mobile apps that have yet to be formally acquired. Preservation plans and preservation planning really are important elements of the preservation toolbox. It’s just that sometimes, we use the same terms to mean slightly different things.