Something happens in the second half of Scenes from a Marriage — either version, Bergman's original 1973 television miniseries or the recent American remake — that is almost impossible to retain with fidelity. The emotional logic is layered across hours; the resonances between scenes depend on remembering a gesture or a line of dialogue from earlier. The first time through, you feel the weight of it. The second time, if you kept notes, you understand why you felt what you felt.
This is a minor but real example of what a film diary does. It converts an experience into a record, and that conversion changes the experience itself — during the film, in the hours after, and across years of watching.
What Writing Does to Attention
The act of anticipating a note — even a brief one — reconfigures how you pay attention. You begin to hold the film slightly differently: not at arm's length, exactly, but with a corner of your mind on what you'll say about this. This sounds like it would distance you from the work, and it can, if you're not careful. Done right, it focuses you. It makes you ask: what is actually happening here?
This is the difference between watching a film and reading it. A close reader doesn't just feel what's in front of them — they ask why the scene is constructed this way rather than another, what the director is doing, what they've noticed that the film itself seems to notice. The diary is the place where that kind of reading happens.
Critics develop this capacity professionally and sometimes lose the original pleasure of unconscious absorption. But there's a middle register — attentive without being analytical, open without being passive — that the journaling habit tends to cultivate. You watch better. More precisely: you watch in a way that will let you remember better.
The Problem of Cinematic Memory
Human memory for film is notoriously unreliable. We remember emotional climaxes — the image that burned itself in — but we compress, confabulate, and elide everything around them. Ask yourself what actually happens in the middle third of a film you love. The answer is often more vague than you'd like.
This is partly a feature of how narrative memory works: we encode meaning and emotional arc more reliably than sequence or detail. But it means that without a record, our taste evolves on a thin evidential base. We claim to love a director and find, on rewatching, that the films in our head are chimeras — better or worse, differently shaped, than the originals.
A film diary doesn't solve this entirely. But it creates a second record — alongside memory's impressionistic sketch — that is more factual, timestamped, anchored to what you actually thought the night you watched, before the film got revised by everything that came after.
The Lineage: From Private to Social
People have kept film diaries for as long as there were films to see. The cinephile tradition includes a long history of private notebooks — lists of showings attended, half-finished notes on what was seen at which repertory cinema on which afternoon. Susan Sontag wrote Against Interpretation partly from the position of someone who had been keeping score on her own viewing for years. François Truffaut's cahiers were famously obsessive before he directed a single frame.
The practice went social with Letterboxd, which arrived in 2011 and did something genuinely interesting: it made the private record visible, social, comparable. You could see not just what you'd watched but how your taste looked from the outside — which films you'd given five stars, which directors you'd explored in depth, the shape of your watching across a year. The "Year in Film" recap was the first popular visualization of film taste as data.
Newer still is the idea that the diary could respond — that a record of your watching, annotated with what you said about each film, could be the basis for genuine conversation with something that knows cinema well and remembers what you've told it.
The AI Companion in the Journal
Reely, Intertitle's film journal feature, is built on this premise. It's spoiler-aware — it draws from your Letterboxd history, so it doesn't have to treat you as a first-time viewer of films you logged years ago. If you want to talk about the ending of Caché (Haneke, 2005), it won't hedge around the ambiguity it knows you've already sat with. If you mention you found The Favourite (Lanthimos, 2018) colder than you expected, it can take that as a data point about your relationship with distancing irony, not just a note about one film.
This is different from a search engine or a recommendation widget. It's closer to what a good film-conversation partner does: it builds a model of your taste and engages with what you say in light of that model. The difference between talking to someone who doesn't know your viewing history and someone who does is the difference between starting from zero and continuing a conversation already in progress.
The Habit Is the Point
The value of a film diary is cumulative. A single entry is a snapshot; two years of entries are a portrait. You see patterns you couldn't have seen in advance: that you tend to rate films from a particular decade higher on first watch and revise them down later; that you keep returning to films about families despite thinking of yourself as primarily drawn to political cinema; that your one-sentence notes from the year you were living somewhere new carry a different emotional register than the notes from the year you were settled.
The diary becomes a record of the viewer as much as the films. And the films become entry points into something larger — into what you were like when you watched them, what you were looking for, what you found.
None of this requires an app. A physical notebook works fine. But a notebook can't search itself, can't surface what you wrote three years ago about the same director, can't draw on your full watching history when you ask it what to see next. The practice is ancient; the tools are new.
Intertitle started with the idea that the watching record you've already built — the ratings, the logs, the occasional journal entries — contains more information about your taste than almost any quiz or questionnaire could. Reely is built to work with that record, to be a companion in the continuing project of figuring out what cinema means to you.