A terabyte of photos is not a memory

The cost of capturing anything fell to nothing and the cost of describing it did not move. That gap is why a person can hold forty thousand photographs and still be unable to find the one of their mother in the kitchen, and why the person who inherits them will not even try.

What archivists know that consumer software does not

An archive is not a pile plus search. It is material plus description, and the description is most of the work.

Provenance. Where this came from, who made it, when, and what it was part of. Without it, an item cannot be trusted or placed.

Context. What was happening around it. A photograph of a kitchen is nothing; a photograph of the kitchen in the house the family lived in for twenty years, taken the week before they left, is a record.

Relationships. How items connect to each other and to people. This is what makes an archive navigable rather than merely searchable.

Appraisal. Deciding what to keep. Professionals treat this as a core skill, not a failure to store enough, and the discipline that guides it is a century old.

Consumer software does the first partially, through automatic timestamps and locations, and essentially none of the rest. Face recognition and semantic search help find things somebody already knows to look for. They are close to useless for somebody who does not know what is in the collection at all.

Why volume actively hurts

Every unlabelled item raises the cost of every search. At some size the collection crosses a threshold where nobody attempts it — the effort to find anything exceeds the expected value of finding it — and the material becomes functionally lost while remaining perfectly intact.

That threshold arrives much sooner for an inheritor than for the person who made it. The maker navigates by memory: they know roughly when, roughly where, who was there. Remove that memory and the same collection is a haystack with no needle established.

What a record looks like instead

Smaller, and described. A few hundred items with real description are worth more than a hundred thousand without, and this is the finding that surprises people most.

Selection is the work. Choosing what represents a life is a task the person should do themselves, while they can, because nobody else can do it for them accurately.

Description at the point of selection. Who, when, where, why this. Captured in conversation rather than by filling in fields, because nobody fills in fields.

Narrative to hold it together. The connective account that makes items meaningful in relation to each other. Without it, an archive is a set of artefacts with no story, which is what a box of unlabelled photographs already is.

The awkward implication

A good personal archive is mostly a deletion exercise, and that is an unpopular thing to say to someone who has spent years accumulating.

But the alternative is not preservation of everything — it is loss of everything, arriving later and in a form that looks like safety. A described, curated record that somebody will actually read is preservation. Forty thousand unlabelled files are a decision deferred onto somebody who will make it by not looking.