Another CSCW paper for this week’s reading group; this one on digital archives. What does data preservation mean in a digital world? How do adults see their legacy? This paper draws on archival theory to study these questions.

Authors: Sarita Y Schoenebeck, Paul Conway

Links: ACM DL, free access

Background and motivation

The paper opens:

If you could preserve just some of your digital data 15, 30, or 100 years into the future, what would you choose to save?

This choice underlies “archival appraisal theory,” which is the foundation of the paper. The field of archival science, meanwhile, “studies the theory and practice of preserving, organizing, and providing access to records of human activity.”

I was unfamiliar with archival theory before reading this paper, but the authors helpfully include an overview. While archival was once considered an act of passive curation, the field now recognizes that curation is inherently subjective and political. “Professional practices for selecting and preserving records … are about choosing which narratives will continue to exist through time while discading others.”

On archival appraisal: appraisal is how we assign value to items that might be archived. This, too, is naturally subjective. And perhaps most importantly, it’s an act of power.

Archivists appraise, collect, and preserve the props with which our notions of identity are built. In turn, notions of identity are confirmed and justified as historical documents validated with all their authority as ‘evidence’ the identity stories so built.

This paper draws upon archival theory in a computing context. The authors interview 17 adults ages 51 – 72, who “are—or soon will be—experiencing decisions about what to keep, what to get rid of, how, and when.”

In this work, we build on archival appraisal theory as a foundation to explore personal data stewardship principles that apprize data and its management as instruments of power, control, subjectivity, and emotion.

Interviews with older adults

The authors’ farming of the interviews was interesting (I added line breaks for clarity):

The interview began with warm-up questions about family, friendships, and daily life.

The first set of questions was about remembering: what were memorable moments, people who were important, and things of value. We also asked participants what they thought having a legacy meant and what they thought would represent their legacy.

The next set of questions was about forgetting, such as: what things have you gotten rid of or actively destroyed? Are there moments of your life you wish you could forget? Are there people you wish you could forget?

Knowing that most studies in this field are conducted on undergrads or early career professionals, these questions already have my attention.

Data as legacy: some participants hoped to document their impact on the world—that they helped people, that they were kind, that they worked with youth. Others expressed the importance of family values or creative work (music! art!).

The participants in the study were widely invested in the sharing and communication capabilities of Facebook without a particular consideration of the long-term availability of Facebook content. … The urge to share data that could contribute to the lodging of a personal legacy is more ubiquitous than the urge to save such data in a personal archive.

Precarity and forgetting: most participants described negative experiences and relationships as things they’d like to forget, but often said they’d not want to forget people themselves. Others experienced detachment; the following quote (from a participant) is incredibly powerful to me—

What legacy? Poor people don’t have legacies, do they? My name is [spells last name]. It’s pronounced [pronounces]. My legacy is that my youngest son has my last name. I grew up where family names are… your heritage is important. I hope he has some boys and regardless if he marries the woman or not, that they can come to an agreement that their last name’s going to be [last name]. That’d be my legacy. I have nothing to give my sons when I die. I have nothing to give my grandson when I die. I have nothing, but a name, memories, so that’s going to be my legacy. -P9

Institutional and personal archival: participants were aware of privacy risks with companies holding personal data, but some mentioned that they stuck around on e.g., Facebook because their family was there.

People are readily embracing the power of digital data to document their lives and to share with those around them the evidence of their personal and creative impact. Our research is a small indicator of a growing disconnect between what people think a personal archive is, on the one hand, and what happens to the pieces of that digital archive when agency and control are dispersed and distributed across technology platforms.

My thoughts

I’m reminded of Hamilton, here: “you have no control / who lives, who dies, who tells your story.” The idea of a legacy is explored throughout the musical. And to me, in a way, the creation of Hamilton is itself an archival process, about reimagining the founding of America by telling different stories and centering different people.

This paper was interesting. I learned a lot from reading it, and I’m glad I read it, but it wasn’t quite what I was expecting. I came in looking for a discussion of personal data and archival practices, of people’s relationships with digital platforms, and of how these ideas might inform technology design.

Instead, I found interviews on what it means to have a legacy, what’s worth remembering and forgetting, and how they think about creating their own archives. While I liked reading them, I think I need to reflect more on what this means for computing.

Archival theories propelling an understanding of the archive in the 21st century are a natural fit with computing scholarship. CSCW researchers generally subscribe to the belief, articulated by Pinch and Bijker [63], that technological systems are socially constructed. Archivists share this belief, arguing that archives, and the ideas, feelings, actions, and transactions embedded in them, are social constructs.

I don’t disagree with this, but I had hoped that the authors would expand more on what this means. How can platforms use the learnings from this paper? How can they place the power inherent in archival appraisal in the hands of the user? What responsibilities do technology companies have with respect to archival?

I hope that these questions are discussed in future work. As I wrote above, this was interesting, but for me it raised more questions than it answered!