Designing for Digital Archives

Elizabeth Churchill Yahoo! Research| [e-mail]

with Jeff Ubois

Fujitsu Labs of America | [e-mail]

David Gartner

Have you amassed a collection of photos and other media without quite knowing how to manage it? Have you spent hours trying to locate a precious or extremely important file? Have you ever wished you’d backed up your files after a computer crash?

More and more of our work and personal content is digital. And mobile, digital technologies like camera phones are changing the nature of capture and collection—what and how we collect. We are living in a world of continuous accumulation.

This is relatively new. Ten years ago fewer people had home computers, fewer services existed, and we weren’t surrounded by all those appealing, shiny devices that promise to record our every action in case we want to take a step down memory lane or revisit an article written a while back to snaffle some useful content. Back then terms like “ moblog-ging”, “lifelogging,” “ microblog-ging,” and “lifestreaming” were not in common parlance.

Ironically, this ease of capture and replication actually makes it more likely that we’ll lose stuff. The sheer volume of data we are able to collect makes organization daunting and specific content difficult to locate. Frankly, the logically extreme vision of life as constant accumulation

offered by Gordon Bell and his collaborator Jim Gemmell, with their MyLifeBits project, is apt to make anyone with old-time curatorial sensibilities erupt in hives.

Amplifying the challenge is the fact that content tends to accumulate in various places— on internal or external flash and other portable drives; on recording devices themselves (cameras, audio recorders, phones); and hosted at ISPs and by services like You Tube and Flickr. Few people have a centralized repository of all their stuff. We curate, consolidate, and/or back up randomly or not at all, and have muddled mental models regarding file formats, backup, and archive practices and services. Prospective retrospective—that is, imagining now what we will want to remember in the future—is hard; we have a limited ability to gauge such future value. So we have a propensity to defer decisions about whether something is worth keeping or not.

Consequently, most of us are what Microsoft’s Cathy Marshall and her collaborators have called “lazy preservationists,” who rely on “opportunism, optimism, and benign neglect.” And most of us are living in a world of digital bloat, our untamed and insecure data strewn all over the place. We skip along on

a wing and a prayer, explaining away catastrophes and rethinking data importance in the face of loss: “I guess it must not have been important if I lost it.” Sometimes this kind of loss and revision is therapeutic. Sometimes it is not. Sometimes we spend hours reconstructing content or creating passable replacements. For our own archives this is personally troubling, but as a culture it is positively terrifying that our data and our memories are at risk.

Some see this problem as a commercial opportunity. GYMA (Google, Yahoo, Microsoft, AOL) are exploring the business of archiving, backup, and storage, and services; others, like Seagate’s Mirra Personal Server, Apple’s .Mac account, EMC’s Mozy promise storage and a “data cloud” where our stuff will be safe … forever. Or until we fail to pay the subscription fee. Or until they have business or technical problems. Or, as happened to one of our own interactions columnists, some malicious miscreant masquerades as you and in a click of a button or two, deletes all your precious material. Under most terms of service agreements, users have no recourse and companies have no obligation to restore the “lost” material even if back-ups exist.

We need to develop a finer