Against the Grain: Why We Should Stop Archiving by 'Significance'

The most common mantra in digital preservation is to be selective. We’re told, with great authority, to focus our limited resources on what is "important," "historically significant," or "culturally valuable." This sounds like wise stewardship. It is, in fact, a deeply flawed and potentially dangerous filter, one that guarantees the erasure of the very context we claim to preserve.

The problem is that our judgment of significance is a product of the present, a lens ground by contemporary biases, power structures, and blind spots. We archive the speeches of politicians, the declarations of institutions, the works of celebrated artists. These are the "oaks" we deem worthy of the archive. But what of the "weeds"? The sprawling, chaotic, and seemingly trivial undergrowth of daily digital life—the forgotten group chats, the derailed forum threads, the ephemeral dashboards of defunct local businesses, the raw sensor data from a weather station that no one thinks to question? These are the materials that future historians will need to understand not just what we said was important, but how the world actually functioned and felt.

The Tyranny of the Curator

When we pre-judge, we enact a silent tyranny. We decide, today, what future generations will be allowed to know about us. We are building a canon of the present, a curated highlight reel that will be mistaken for the full picture. This is not preservation; it's the creation of a sanctioned narrative. The truly radical, counterintuitive practice is not smarter selection, but strategic non-selection at the point of capture. To save everything within a defined, technical scope—every byte from a municipal server, every crawl of a neighborhood association’s site—is a more honest form of humility. It says: "We do not know what will matter, so we will defer that judgment."

This isn't an argument for hoarding without purpose. It is an argument for shifting the intellectual labor from upfront curation to backend description and access. The work becomes less about "is this significant?" and more about "how can we make this pile of stuff navigable?" Let the researchers, the artists, the community members of 2124 sift through the digital sediment with their own questions. Their definition of significance will be utterly different from ours. A seemingly mundane log of pothole repair requests from 2025 might, to them, be the richest dataset on pre-autonomous vehicle urban infrastructure and civic engagement.

Yes, storage has a cost. But the cost of lost context is incalculable. By archiving only what we anoint as important, we are building a future of disconnected monuments, standing in a field where the city once was. We must have the courage to preserve the messy, the boring, and the ostensibly insignificant. For in that sprawling, undifferentiated mass lies the true texture of our time—a truth we are not wise enough to pre-edit. The most ethical archive may be the one that consciously refuses to play the role of arbiter, and instead commits to being a keeper of the whole, bewildering field.

Notes & further reading

A few pages I came back to while writing this: