The Preservationist's Paradox: Why Saving Everything Might Save Nothing
There’s a powerful, seductive idea that has taken root in our digital preservation efforts: we must save everything. It’s a noble impulse, born from a century of watching priceless cultural artifacts vanish into the ether. The logic seems unassailable. Storage is cheap, data is fragile, and we cannot possibly know what future generations will deem significant. So, we point our crawlers at the entire web and hit ‘archive.’ We champion the idea of the comprehensive, unbiased record. But I want to propose a uncomfortable counterpoint: in our frantic effort to save everything, we risk creating archives so vast and undifferentiated that they become, for all practical purposes, useless.
The received wisdom is that selection is a form of bias, and bias is the enemy of a true historical record. By archiving without human intervention, we avoid the subjective flaws of a human curator. But this ignores a crucial truth: the algorithm is not neutral. It, too, has a bias—a bias towards what is easy to capture. It favors large, stable websites with clear structures over fragile, personal, or complex web spaces. It endlessly duplicates popular news articles while missing the ephemeral conversation happening in a now-defunct forum. The result isn’t an unbiased record; it’s a record biased towards the loud, the link-friendly, and the technically simple.
The Weight of the Infinite Archive
This creates what I call the Preservationist’s Paradox. We build these infinite archives in the name of access, but their sheer scale makes meaningful access nearly impossible. It’s the digital equivalent of storing every pebble on a beach to ensure we don’t lose a single grain of sand. When researchers of the future are confronted with a petabyte of undifferentiated data from our era, how will they find the signal in the noise? Without the context provided by selective curation—without someone having made a conscious choice that *this* matters—the archive becomes a desert of data where finding an oasis of meaning requires a miracle.
This isn’t an argument for abandoning bulk preservation. The work of organizations like the Internet Archive is foundational and invaluable. Rather, it’s a plea to stop seeing selective, curated archiving as a lesser or biased practice. We need both. We need the sprawling, automated ‘wilderness’ of the bulk archive, but we also desperately need the carefully tended ‘garden’ of the curated collection. The curated archive provides context, narrative, and, most importantly, a statement of value. It says, ‘We, as a society, looked at this digital moment and decided it was worth the focused effort to save and explain.’ It provides a path through the wilderness.
True preservation isn’t just about throwing bits into a digital vault. It’s about ensuring that what we save can actually be found, understood, and used. It requires not just storage, but storytelling. It demands that we complement our powerful, automated crawlers with the equally powerful, irreplaceable force of human judgment. Otherwise, we may find that in saving everything, we have successfully preserved nothing of meaning at all.
Notes & further reading
A few pages I came back to while writing this: