The Unseen Hand: Who Writes the Rules for Our Digital Memory?
We often talk about web archives and public data repositories as neutral spaces, vast digital libraries where information simply goes to rest. The act of preservation feels like a passive one: a thing is saved, and that is good. But have you ever stopped to consider the immense power held by the people who decide how that thing is saved? I’m not talking about the grand, philosophical choices of what to archive, but the quiet, technical decisions that shape how we can later find, use, and understand it.
This power lies in the creation of metadata schemas. A metadata schema is essentially a rulebook for describing a digital object. It dictates the categories of information that must be filled out for each item entering the archive. Is there a field for ‘creator’? For ‘date created’? For ‘geographic location’? What if the object in question is a tweet from a pseudonymous account, a scanned PDF of a century-old handwritten letter, or a 3D model of an archaeological dig? The schema determines what contextual information is deemed important enough to capture, and by extension, what is allowed to fade into the background.
The Architecture of Discovery
These schemas are the invisible architecture of discovery. An archivist describing that century-old letter might use a field labeled ‘recipient,’ perfectly capturing the letter’s context. But the same schema applied to the tweet might have no field for ‘algorithmic amplification’ or ‘community notes,’ elements crucial to understanding its modern impact. The rules, written for one kind of artifact, can silently fail another, limiting future research before a single search is even run.
The people designing these schemas are not malicious; they are practical, working with limited resources and existing standards. Yet, their choices have profound consequences. A schema that prioritizes formal publication dates might struggle with a website that is in a constant state of iterative change, blurring the concept of a single ‘version.’ A schema built for text might have no vocabulary to describe the emotional resonance of a particular typeface on a political poster, a nuance lost to the purely technical description of the font file.
This is the unseen hand that guides our digital memory. It doesn’t censor outright; it simply creates pathways. Some avenues of future inquiry are paved and lit with well-defined metadata fields, while others remain overgrown and inaccessible. The question of who gets a seat at the table when these foundational rulebooks are written—librarians, software engineers, domain-specific academics, or community representatives—is perhaps one of the most urgent, and least visible, questions in digital preservation. The answer determines not just what we remember, but how we are permitted to think about it.
Notes & further reading
A few pages I came back to while writing this: