1. The day is the basic unit of personal history.

Not the note. Not the photograph. Not the workout, the meeting, the meal. The day. A note is a fragment. A photo is a frame. A workout is a number. A calendar event is a label. A day is the smallest unit that still carries texture — morning into evening, weather into mood, the rhythm in which a life is actually lived.

A practice of memory that does not center the day will lose the day.


2. Capture is not memory.

Your phone captures more than your grandmother’s diary did. You remember less. Capture without context is just storage — a slow accretion of facts waiting to be forgotten. A photograph that lives only in a 50,000-image camera roll is a photograph the world has effectively lost.

To capture is to make a trace. To remember is to return that trace to a day.


3. Traces become meaningful when returned to context.

A single step count is a number. The same step count, on a Tuesday in February, beside a photograph of your daughter at the kitchen table, alongside the song you played four times that afternoon, is the texture of a life.

Retrography is the practice of putting traces back where they belong.


4. Memory is reconstructed in the person, not the machine.

The reconstruction is the memory. The machine only provides the cues.

This is why we are wary of any instrument that promises to “remember for you.” It cannot. It can only assemble the materials with which you do the remembering. The work — the actual mental motion of returning to a day — remains yours.


5. The archive should belong to the person it describes.

An archive is the thing itself, not a service that renders it.

This means: not rented. Not held hostage to a subscription. Not stored on a server one can lose access to. Not used to train models on a private life. An archive is yours in the way a paper journal is yours — kept by you, readable by you, exportable by you, forgettable by you.


6. Assistance is a lens, not an author.

A machine can help when you need to step back — to summarize a long stretch, to surface a pattern, to find a thread. It is useful, the way a magnifying glass is useful.

But the author of a life remains the person who lived it. The summary is not the memory. The pattern is not the meaning. Assistance is a lens you may choose to look through. It is not the eye behind the lens.


7. Quantification should support memory, not replace it.

Numbers are useful when they help you return to a day. The number of steps you walked on a Sunday is interesting because it places you somewhere, doing something, feeling some way.

Numbers are not useful when they become the day’s only record. A life measured only in metrics is a life that has lost itself to its own dashboard.

Quantification serves memory. Memory does not serve quantification.


8. Ordinary days are worth preserving.

Milestones are easy. The calendar protects them. Holidays, birthdays, weddings, the deaths in the family — these are remembered by default, by social convention, by the rituals around them.

What disappears is the ordinary. The Tuesday you walked the long way home. The Wednesday it rained and you read for an hour in the kitchen. The unremarkable Saturday that — eight years later — turns out to have been the last one with your father in good health.

The ordinary day is the form of life. To preserve only the milestones is to lose the form.


9. A private archive should create return, not addiction.

A feed is designed to never let you leave. An archive is designed to be returned to.

Retrography is closer to a bookshelf than a stream. You visit it. You sit with it. You close it. You come back next week. It does not demand your attention; it offers something to come back for.

If you find yourself opening the archive compulsively, the design has failed. If you find yourself opening it on a Sunday afternoon, six months from now, to read what you wrote on a Tuesday — the design has worked.