The Digital Cemetery: Why Preserving Everything Might Cost Us Our History

There’s a guiding principle in our field, a pious mantra repeated at conferences and in funding proposals: save everything. The sheer, staggering scale of the digital realm demands a strategy of capture-at-all-costs. We build bigger server farms, develop more complex crawling algorithms, and celebrate each new petabyte of archived data as a victory against the void. But what if our greatest triumph—our boundless capacity to hoard—is also our greatest failure? What if, in our rush to preserve everything, we are merely constructing a digital cemetery where information goes to be forgotten?

The counterintuitive argument is this: uncontrolled preservation doesn’t safeguard history; it buries it. A graveyard is not a library. A library has a catalog, a system of organization, a selection process that gives its contents meaning through context and accessibility. A cemetery, by contrast, is a vast collection of inert objects, arranged by convenience, where the primary interaction is one of mourning, not discovery. Our terabytes of “preserved” data, lacking the curatorial care and intellectual scaffolding of a true archive, increasingly resemble the latter.

The Tyranny of the Terabyte

Unlimited preservation creates a tyranny of volume. The human mind, and indeed our current tools for discovery, are not built to sift through an undifferentiated mountain of digital detritus. A researcher looking for a specific local government report from 2004 might be presented with a billion equally “preserved” web pages, blog comments, and cookie consent banners from that year. The signal is lost in the noise. By attempting to give all data an equal chance at immortality, we have inadvertently devalued all of it. The truly significant is submerged under a relentless tide of the insignificant.

This approach is a radical departure from the traditions of physical archiving, where scarcity of space forced a conscious act of selection. An archivist had to ask, "Why should this record be saved? What story does it tell? Who might need it?" This process wasn't about exclusion for its own sake; it was an act of narrative creation. It shaped a coherent, usable past from the chaotic raw material of the present. Our digital “save-all” strategy abandons that responsibility, outsourcing the burden of meaning-making to some future, hypothetical AI that might one day be smart enough to sort through our digital refuse.

Worse, this obsession with volume drains resources from the more delicate, human-centric work that gives data its soul: description, context, and connection. A lone spreadsheet is a ghost. Only when its metadata explains its origin, its creator, and its relationship to other records does it become a historical source. We are spending our fortune on building ever-larger coffins for data, while defunding the librarians who could give that data a voice.

This is not a call for destruction, but for a paradigm shift from preservation to stewardship. We need less emphasis on the bit-bucket and more on the curatorial act. It requires the difficult, unpopular work of asking not “what can we save?” but “what should we save, and for whom?” It means building rich, well-tended gardens of information, not leaving behind overgrown digital forests. The goal isn’t to prevent the death of data, but to ensure that the data which survives is given a meaningful life, one where it can be found, understood, and used to illuminate the past, rather than simply occupying space in a silent, digital tomb.

Notes & further reading

A few pages I came back to while writing this: