The Neutral Net: A Critique of the 'Archive Everything' Imperative
In the world of digital preservation, a powerful and largely unquestioned dogma has taken root: we must save everything. It’s a sentiment born from a noble fear—the terror of the digital void, the silent erosion of our collective memory. This impulse drives the heroic efforts of web archivists and open data advocates. But I want to question this foundational principle. Is the goal of a perfect, complete archive not only impossible, but perhaps also undesirable?
The argument for totality is seductive. With storage costs plummeting and crawler technology advancing, why wouldn’t we aim to capture the entire digital ecosystem? It feels like a logical extension of the open data ethos: if data is good, more data is better. This mindset, however, often overlooks the inherent editorial nature of the act itself. An archive is not a mirror; it is a portrait painted by its creators. Selection is unavoidable, from the seed URLs chosen for a crawl to the frequency of capture and the depth of a site’s hierarchy that is deemed worth preserving.
More critically, the ‘archive everything’ mantra risks creating a false, and dangerously complacent, sense of neutrality. A petabyte of data feels objective. It has weight. It implies comprehensiveness. But this vastness can mask the silent, systemic erasures that occur long before the crawler is deployed. What about the communities that primarily exist on closed platforms, their conversations happening in ephemeral, encrypted spaces? What about the vast stretches of the internet that are deliberately excluded from indexing? Our grand, total archive becomes, by default, an archive of the indexable, the public, and the technically convenient. It is a map that mistakes the coastline for the entire continent.
The Burden of the Uncontextualized Past
Furthermore, preservation without purpose is a peculiar form of hoarding. It offloads the immense ethical burden of contextualization onto an unspecified future. We are creating a digital Pompeii, burying countless artifacts under layers of data, assuming that future archaeologists will have both the tools and the wisdom to interpret them correctly. But without the surrounding social context—the jokes, the nuances, the unspoken rules of a now-defunct platform—what meaning can be derived? A saved hate forum is just data; without the stench of its original vitriol, it risks being neutered into a mere historical curiosity, or worse, misappropriated.
This is not an argument for inaction. It is a call for a more thoughtful, and frankly, more humble approach to preservation. Instead of striving for an impossible totality, our efforts would be better spent championing intentional, contextual archiving. This means being transparent about our gaps. It means prioritizing the voices and spaces most vulnerable to disappearance. It means archiving not just the data, but the story of why it was saved and how it was meant to be read.
True preservation isn’t about building a perfect memory bank. It’s about curating a meaningful record, one that acknowledges its own limitations and biases. It is an act of interpretation, not simply collection. By abandoning the fantasy of a neutral net and embracing our role as active, responsible stewards, we can build archives that are not just large, but truly wise.
Notes & further reading
A few pages I came back to while writing this: