The Myth of the Immortal Bit: Why Data Rot is a Feature, Not a Bug
We in the digital preservation community often speak in hushed, reverent tones about the specter of "bit rot." It’s our great nemesis, the silent entropy that threatens to erase our collective memory. Our mission, as it’s so often framed, is to fight this decay at all costs. We build redundant systems, migrate file formats, and emulate obsolete hardware in a grand, noble war against oblivion. But what if we’ve fundamentally misunderstood our foe? What if data decay isn’t a bug in the system, but one of its most crucial features?
The popular wisdom is that a perfect archive is a static one, a frozen moment preserved in digital amber. This ideal assumes that a bit, once written, should remain immutable and perfectly legible forever. This is the myth of the immortal bit. It’s a seductive idea, born from a computational logic that prizes perfect, un-changing replication. Yet, it ignores the messy, beautiful, and essential nature of how human memory actually works. Our own biological memories aren't pristine recordings; they degrade, shift, and are rewritten with every recollection. This isn't a failure of our brains; it’s a mechanism for relevance. It allows us to prioritize the now, to let go of the trivial, and to reinterpret the past through the lens of the present.
Our frantic scramble to save everything forever creates a different, perhaps more insidious, problem: the burden of the undifferentiated archive. By attempting to preserve every tweet, every blog comment, every minor revision of a public document with equal fidelity, we risk creating a vast, undigested corpus of information. We preserve the signal, but we also preserve an overwhelming amount of noise, stripping context and making it impossible for future historians—or even ourselves—to discern what truly mattered. In trying to remember everything, we make it harder to remember anything of value.
Perhaps, then, we should stop viewing data decay as a purely technical problem to be solved and start seeing it as a curatorial force to be understood. The natural degradation of file formats, the breaking of links, the silent corruption of files on aging hard drives—these aren't just failures. They are a form of natural selection for information. They ask a vital question of every piece of data: is this worth the effort to save? Is its meaning robust enough to justify the resources required to migrate it, emulate it, and explain its context to a future audience?
This isn’t an argument for negligence. It is, instead, a call for a more thoughtful and intentional preservation strategy. Rather than trying to build a perfect, un-changing vault, perhaps our goal should be to build a dynamic garden. We tend to what is vital, we let what is no longer relevant return to the digital soil, and we accept that some loss is not just inevitable, but necessary. It is the process of decay that allows for new growth. By embracing this cycle, we might just create a digital memory that is more human, more manageable, and ultimately, more meaningful.
Notes & further reading
A few pages I came back to while writing this: