The Peril of the Perfect Copy: How Digital Preservation Can Erode Authenticity
We speak of digital preservation in the language of salvation. Our mission, we say, is to save, to rescue, to salvage data from the brink of oblivion. The enemy is bit rot, format obsolescence, the silent decay of magnetic charge. Our victory condition is the perfect, unaltered copy, a file checksummed and replicated across data centers, safe for eternity. But what if, in this single-minded pursuit of perfect fidelity, we are inadvertently robbing preserved objects of something just as vital as their binary integrity? What if we are preserving the body but losing the soul?
The counterintuitive argument is this: our technical obsession with creating flawless, timeless copies is creating a sterile, ahistorical record. We scrub away the very marks of time and use that give an object its authentic character. We are like art restorers so focused on removing every fleck of age from a painting that they erase the patina, the subtle cracks, the evidence of the painting’s own journey through the world. The result is technically perfect but historically hollow.
Consider a tangible example: a physical book. Its value isn't only in the words printed on its pages. The cracked spine tells you which passages were most loved. The marginalia in faded pencil is a conversation across decades. The faint scent of a previous owner's bookshelf, the dog-eared corner, the coffee stain—these are not flaws to be corrected. They are data. They are the evidence of lived experience, a provenance that is irreproducible. In the digital realm, our preservation efforts are the equivalent of taking that book, meticulously re-typesetting it into a pristine new edition, and pulping the original. We've saved the text, but we’ve discarded its history.
The Ghosts We Systematically Evict
This problem is amplified in web archiving. When we 'fix' broken links in a saved webpage or update embedded media to a modern, playable format, we are sanitizing the past. The infamous 'Error 404' is not just a technical failure; it is a historical artifact. It marks a rupture, a loss, a dead end that was very much a part of the internet experience at a specific moment. By auto-correcting these errors in our archives, we create a frictionless, anachronistic past that never truly existed. We evict the ghosts that haunt the machine, leaving behind a simulation that is suspiciously clean.
The same goes for file formats. Emulating a 1990s video game on a modern system to ensure perfect playability is a marvel of engineering. But it often smooths over the pixelated roughness, the slow loading times, the specific sound of a period-appropriate processor struggling to render a scene. That struggle was part of the artifact. The experience of waiting for a tape to load on a Commodore 64, listening to the chaotic chirps and buzzes, was an integral part of the software's identity. Our perfect digital copy loses the texture of the original medium.
This is not a call for negligence. We cannot allow data to crumble. But it is a plea for a more nuanced philosophy of preservation. Perhaps, alongside our perfect master copies, we need to preserve 'witness copies'—snapshots that include the errors, the corruptions, the platform-specific quirks. We need to document not just the file, but the environment in which it lived and aged. Authenticity in preservation isn't just about the bits being correct; it's about preserving the truth of the object's existence, complete with its scars and its history. The most faithful archive might just be the one that is brave enough to show its age.
Notes & further reading
A few pages I came back to while writing this: