1. The condition

Yesterday produced more passively-recorded information than your grandmother’s entire week. Photographs taken without thinking about them. Routes logged in the background. Songs registered, heart rates sampled, messages timestamped, weather noted, words spoken into devices that quietly listened. Most of it was made without you trying.

Almost none of it sits in the place where the day actually happened. The records are scattered — across applications, services, accounts, devices — each written for a different purpose, none of them organized around the day they came from.

We are drowning in records and starving for memory.

Six months from now, you will mostly forget yesterday. The texture will be gone. You will remember it as a feeling, with three or four anchors — a meeting, a meal, a phone call — and the rest will be silt. The records will still exist. You just won’t have a way back to them.

2. The day as unit

Not the note. Not the photograph. Not the workout, the meeting, the meal. The day.

A day is the smallest unit of personal history that still carries texture. A note is a fragment. A photograph is a frame. A workout is a number. A calendar event is a label. But a day is the rhythm of a life — morning into evening, weather into mood, the people you saw and the songs you played and the route you walked and the small thoughts you had between things.

Most modern instruments record fragments. None of them remember days.

To practice retrography is to start with the day, then assemble the fragments around it.

3. Memory as reconstruction

Memory is not a recording. It is a reconstruction — built each time, from cues, in the present moment.

What you remember about last June is not stored as last June. It is rebuilt from what survived: a photograph in a familiar place, a song the algorithm played you again, a building you walked past last week. The reconstruction is the memory. The cues are the medicine.

A practice of memory, then, is a practice of better cues. Of gathering what life already records — the photographs, the steps, the places, the songs, the words, the weather — and arranging them, by the day, so that returning is possible.

The point is not to recreate the day. The point is to be able to walk back into it.

4. Against autobiography by machine

It is now possible, technically, to ask a machine to write your life for you. To summarize a year in three paragraphs. To distill a character from a decade of messages. To produce, in seconds, the kind of autobiography that once took a person ten years to compose.

The novelty is real. The temptation is real. The substitution it offers is also real.

We are skeptical for a reason that ought to be obvious: a summary is not a memory. A summary is a substitute for one. To outsource memory to a machine is to remove the work that makes a life feel lived.

The work is part of the point. The slow return to a Tuesday in March — picking it up, turning it over, finding the song you were listening to, recovering the weather — is the act in which memory happens. A machine can deliver the materials; it cannot do the remembering on your behalf, because the remembering is not happening in the machine. It is happening in you.

There is a useful role for assistance, of course. When the stretch is long and the eye gets tired, a careful summary can help you step back. So can a librarian. So can a friend who knew you that year. These are tools. They are not the same as the work.

The archive remembers the facts. You remember the life.

5. The practice

Anyone can begin. The traces are already there — in the photographs you have not deleted, the routes your devices record, the songs that come back to you on shuffle, the words you have written in messages and journals and the margins of books.

The work is to arrange them by the day they came from, and to return.

Some of this can be done by hand. Some of it is helped by instruments. The instruments matter less than the practice. The practice is the slow assembly of evidence; the patient return to ordinary days; the refusal to let years pass without ever looking back at how they actually felt to live through.

A life is a great quantity of days. Most of them are forgotten.

Retrography is the practice of forgetting fewer of them.

6. Principles

The full set lives at /principles. In short:

• The day is the basic unit of personal history.
• Capture is not memory.
• Traces become meaningful when returned to context.
• Memory is reconstructed in the person, not the device.
• The archive should belong to the person it describes.
• Assistance is a lens, not an author.
• Quantification should support memory, not optimization.
• Ordinary days are worth preserving.
• A private archive should create return, not addiction.

7. Studies

Three reconstructed days. Imagined, but ordinary. They show what the practice looks like when applied to a single, particular life.

→ A Tuesday in Linz (coming) → A Sunday in Lisbon (coming) → A Wednesday at home (coming)