I built this because I needed it

My daughter was born, and within about a week I had four apps open.

One was good at feeds and knew nothing about sleep. One tracked sleep beautifully and had no idea a nappy existed. One did growth charts and wanted twelve euros a month for them. The kraamzorg left a paper booklet on the kitchen table, which was genuinely the best of the lot until it filled up and I started writing in the margins.

None of them was bad. They were just each built for a different slice of the same fortnight, and the slices did not add up to anything. When the consultatiebureau asked how she had been feeding, I had the answer in three places and the confidence to give none of them.

That is the whole origin. Not a market gap — a kitchen table with a booklet, a phone with four icons on it, and a question I could not answer about my own child.

Centralising it was the actual feature

I want to be precise about what was wrong, because “an app that does everything” is usually a bad idea and I do not think this is one.

The problem was not that I had to open four apps. It was that the connections between the things were where the information lived, and no app could see them. A short feed followed by a long sleep is a different fact from a short feed followed by a scream. A nappy at 04:10 is the reason the feed at 04:00 stopped early. A temperature means one thing on its own and another thing next to two days of poor feeding.

Split across four products, all of that is gone. Not hidden — gone, because nothing holds both halves at once. The booklet on the table could see all of it, which is why a piece of paper was outperforming several million euros of venture capital.

So the goal was never “more features”. It was one place, so the things that only mean something together are together.

I am an engineer, so I had opinions about the rest

Having decided what to build, I found I had strong feelings about how — the kind you develop after fifteen years of watching what happens to other people’s data. Three of them turned into rules I have not broken, and I want to write them down publicly so it costs me something if I do.

Nothing you write leaves your device

Hodierna uses analytics. I am not going to pretend otherwise — I need to know whether a screen is being used and whether a sync is failing, and guessing is worse for you than measuring.

But the analytics can only ever know that something happened, never what it said. If you log a feed, the event is “an entry was written” and the property is feed. Not the duration, not the side, not the note you typed at 3am, not the photograph, not her name.

The way this is enforced matters more than the promise. The properties that may be sent are an allowlist — a fixed list in one file — and anything not on it is dropped before it leaves the phone. Not a denylist, which is a promise to have thought of everything in advance. An allowlist fails closed: if I add a field and forget, the field silently does not go, and the bug is that I lose a metric rather than that you lose your privacy.

No data is sold. No data is shared with third parties for advertising. There is no advertising, which is the easiest way to be sure.

You can take everything with you

Every entry, every measurement, every photograph, every document — exportable, always, in formats that outlive me.

This one is partly ethical and partly a promise I am making about my own future behaviour. An app that holds your data hostage does not have to be good. An app you can leave in thirty seconds has to earn the next thirty seconds, every time. I would rather be in the second situation, because I think it produces a better product and because it is the arrangement I would want as a customer.

It is also a hedge you should want. I am one person. If I am hit by a bus, or if I lose interest, or if in five years I turn into exactly the company I am describing — the export button is what makes those survivable for you.

The parts you need at 3am are free, permanently

The logbook is free. All of it. Feeds, sleep, nappies, temperature, growth, sharing with your partner, the export, all three languages.

The reasoning is not generosity. It is that there is no version of this where a parent at 3am with a crying newborn hits a paywall or a network error, and once you accept that, most of the pricing questions answer themselves. Every write completes on the device without waiting for a server. The whole thing works with no signal, on a train, in a hospital basement.

What that leaves is the small number of features that genuinely cost me money per use: storing your documents, and the AI that reads them back to you. Those are paid, because a subscription is more honest than the alternatives — and the alternative to charging for expensive things is either advertising or selling data, and I have already ruled both out.

I am aware “free forever” is a thing people say before a funding round. All I can offer against that is the export button and this paragraph.

What it will never do

One more, and it is the one I feel most strongly about.

Hodierna will never tell you your baby is fine. It is not a medical device, it does not diagnose, and it will not reassure you — because it cannot know, and a piece of software that says “that looks normal” to a frightened parent at 4am is doing something genuinely dangerous.

It will tell you facts. “3.28 kg, 185 g less than last time” is a fact. It will point you at your verloskundige or your huisarts unconditionally, never on a threshold. The line between those is machine-checked in the test suite, in all three languages, because I do not trust myself to hold it by hand at midnight when a feature would be nicer with a little reassurance in it.

Where it is

Hodierna is on Google Play, with the iPhone version still with Apple. It is named after hodiernus — “of today” — because that is the only day a logbook is really about.

She is fine, by the way. I am allowed to say that. The app is not.