Jennifer Howard

Not long ago, I had a chance to interview the cyberpunk Bruce Sterling, who’s in the process of giving his archive to the Harry Ransom Center at the University of Texas at Austin. It was a neat, all-over-the-place conversation—one that’s given me some ideas for future stories—but one thread I can’t let go of is Sterling’s doubt about the wisdom of trusting electronic archives:

He’s also fascinated by dead media—a subject he has written a lot about—and is robustly skeptical about the idea that digital media are durable. “It’s like thinking you’re going to run your Ford Edsel for the next 50 years,” he says.

Scholars won’t find any digital artifacts—no hard drives or floppy disks, for instance, in the material he’s given the Ransom Center so far. “I did not have any electronic documents to give them. Not even one,” he says. “I’ve never believed in the stability of electronic archives, so I really haven’t committed to that stuff.”

“There are forms of media that are just inherently unstable, and the attempt to stabilize them is like the attempt to go out and stabilize the corkboard at the laundromat,” he adds. “You can get into big trouble that way.”

The beginnings of a very interesting conversation on the subject sprouted up in the comments on the story. The digital humanist and textual scholar Matt Kirschenbaum worried that Sterling meant it wasn’t worth even trying to save born-digital artifacts, that “the task is doomed.” (Matt has done a lot of fascinating work on digital forensics, which you can find if you poke around his website, and last year he co-authored a report on “Digital Forensics and Born-Digital Content in Cultural Heritage Collections.”)

Brian Lennon of the University of Pennsylvania, had this to say:

Debate of the stability of electronic records aside, archivists know that not even the best and hardest work can guarantee posterity to more than a fraction of cultural life —- and that success, even with that fraction, consists in not taking one’s successes for granted. When it’s reflective (rather than merely administrative), archival theory is profoundly melancholy, for this reason. From ancient Ebla to twenty-first century Baghdad, archives, libraries, and museums (and universities!) are punctually destroyed.

I like the idea of archival theory as “profoundly melancholy.” That makes sense to me. Like Lennon, I took Sterling’s comments as a caution—a reminder that the whole business is unpredictable and that we can’t count on everything being available in perpetuity. Remember the archivist’s mantra Multiple copies, multiple formats, multiple locations—clay tablets and codex books stored alongside microfilm and hard drives deep inside mountains in Utah, for instance. If nothing else, as Lennon said in his comment, archival success “also means not taking access for granted.”

I’d hoped that the conversation in the comments thread would take off from there, but so far it hasn’t. A question I want to ask here, and it’s one that I’ve been asking myself for a long time, is how much the no-guarantees aspect of all this ought to worry us. Put it another way: How much should we try our damnedest to save, and are we deluding ourselves if we believe that we have a decent shot at saving (almost) everything? Is it crazy or quixotic or noble to imagine (quoting Brian Lennon again) that digital archiving gives us a chance of “enabling the artifact to outrun an army or a cruise missile”?

I cherish archives and what they do—archival stories tend to be among my favorite stories to write—but I also find the prospect of archiving all of humanity’s cultural output overwhelming. (“Tiring” would also work here.) Will the scholars of the future thank us for trying to save every last tweet, every born-digital document and blog post and whatnot that we produce? Is it hubris to think it’s all worth saving, assuming we even could save it all? Does anyone find the idea that not everything will persist a source of relief as well as melancholy?

Historians will argue, with some truth, that we can’t know what the future’s going to want from us—which of our vast output it’s going to have the time and inclination to sift through, assuming at least some of that output survives in consultable form. That’s not an argument I take lightly.

How do you provide for preservation of your own work? What do you hope to hang on to, and what measures do you take to preserve it, short of taking out a thousand-year lease on storage space inside a mountain someplace? If nothing else, the long-term unreliability of the technologies we work with ought to make us think very hard about what we’re trying to save and why.