Interviewer
If you were given unlimited funding to design a system for storing and preserving digital information for at least a century, what would you do?
Rebecca Frank
That’s a great question for a couple of reasons. When I look at risk assessment and risk mitigation, a lack of stable long-term funding almost always surfaces as the most significant threat to digital information.
The glib answer is the money itself would be the solution. It’s not that there’s some technical barrier that we haven’t overcome or there’s not a willingness on the part of digital preservation professionals to do the work, it’s that they can’t get anybody to care enough to create a stable funding stream, right? There’s not something like a Carnegie Library system for digital preservation.
I think if you just give the money to the people who are doing the work and that probably gets you 90% of the way to a solution. It’s not super exciting technically or in any other way, but right now the biggest problem is nobody knows if they’ll be able to do what they’re doing three years from now because that’s when their grant runs out. They can’t maintain staffing with the technical expertise they want because people with technical expertise are desirable and can have stable jobs outside of cultural heritage that allow them to have a life that is planned more than a year in advance. Just giving people money is a very good solution, but if we want to think about what we would do with money in other interesting ways, I do think some amount of advocacy and education also gets you quite a long way. The people who have the ability to put money towards the preservation of things for the public good are not choosing to spend it on something that requires steady, continuous work. Because of the continual replacement of technology, the people who have the power to put money towards it don’t feel like they can plan ahead that far ahead.
I also think maybe the answer is to take that money and move it away from the institutions, away from government, away from organizations and send it to communities that have a very real stake in preserving their own types of information.In contrast, Amelia Acker made a very strong case that it would be worthwhile to restore our faith in institutions. Communities that have proven extremely resilient over very long periods of time at preserving their culture, their heritage, their data, not just in the face of obsolescence, but in the face of active destruction.
Interviewer
Your work has a deep and multifaceted focus on risk. I’m curious if there are trade-offs present in thinking about the risk to a preservation effort from something like a disaster, versus something like everyday neglect. Can those choices negatively affect each other? Are they mutually exclusive?
Rebecca Frank
I began by looking at disaster planning in my earliest work and I did the thing that everybody does, which is look at the big flashy risks. I did a project at the University of Michigan libraries where I worked with our preservation department over the course of a year to customize their disaster plan template for each of our libraries on campus. One of my jobs was to think of scenarios, almost like a tabletop exercise for people to walk through each scenario, like what do you do when this specific event happens?
If data is lost, it doesn’t matter how it came to that point.
It’s really enticing to lean into that, these choose-your-own-adventure disaster scenarios, but I actually don’t think they are particularly useful. Yes, have a disaster plan, know the number for your insurance company, have the plastic sheeting for your books, for digital preservation, have offsite backups that do what they’re supposed to do. But I’ve come to believe that the actions that help mitigate and minimize the smaller everyday risks also work on disasters. Having multiple backups does both, having staff who know and understand their jobs addresses both, solid documentation of how to replace equipment, how to restore whatever is lost, how to reconstruct something, all do both right? The question isn’t was it flood or snow or fire? The question is, did you lose power? Did you lose connection? Do you need to replace equipment? Do you need repairs at your facility?
BitRot is a disaster because it requires restoration of some kind. That’s an answer nobody likes to hear, but I think it’s very practical. What happens is we get caught up in the idea of big disasters. This is something that like risk research has sort of borne out over the years, we have this bias towards exciting, flashy events. People tend to view big disasters as more likely to happen. They tend to view big disasters as more likely and more impactful, and then you overlook the small everyday things that are much more likely to happen and that you actually need to be prepared for on a more regular basis. Like not having enough money. Losing your funding is also a disaster.
If data is lost, it doesn’t matter how it came to that point.
Interviewer
It’s a little unfair to ask you for the answer, but what do you think are the viable long-term financial models for digital preservations efforts?
Rebecca Frank
Digital libraries do not have endowments for the most part. That’s a very unexciting, real answer. Somebody needs to care enough to create a huge endowment that will maintain itself for digital preservation. The very uncool answer is that somebody needs to create endowments that are specifically for digital preservation, the way that we have for extremely large cultural heritage institutions, like the Met, MoMA, or the Getty. We need a Getty for digital preservation specifically. Because the cycle of grants and funding worked for a while, but it’s gone now. We are about to enter an era where we are going to find out what happens when institutions can no longer tell their preservation folks to apply for grants. If the library cares about maintaining their repository, they’re going to have to do it on their own. Or somebody out in the world is going to have to step up.
Interviewer
When it comes to long term preservation, what else do you wonder about?
Rebecca Frank
There are two things you didn’t ask about that I think are worth mentioning. One, I think we’re trying to save too much. And two, I think I’m triangulating from something David Wallace, Paul Conway and Finn Brunton all said, but when archeologists are finding things, most of what they find are in trash heaps, right?
So maybe what we should be doing is putting a bunch of hard drives in a bag and throwing them away. The longest long-term preservation, maybe it’s more likely to happen with the things we throw away. I think it’s such an interesting idea. I mean, I’m not seriously telling people to throw stuff away, but I do think if we’re not talking about short-term access, we’re only talking about long-term future access, maybe we need to trust the future to be able to figure out how to access what we have and just put it somewhere that they’re likely to find it. Nobody in preservation likes to hear that, just like they don’t like to hear that we’re saving too much.
I think the part of it that I find most interesting is it requires a lot more trust in the abilities and motivation of people down the road. We spend a lot of time assuming that people are not going to be able to find or understand or access things. In fact, when people are motivated, they are really good at figuring stuff out. So maybe the problem isn’t software migrations, maybe the problem is just getting data somewhere where people will find it.