What happens to your days if we stop

This is a promise, not a policy. Everything in it has a number attached, because a promise without one is just a nice sentence.

The short version

Your days are on your phone. Not on our servers with a copy on your phone — on your phone, with a copy on our servers.

If this app disappeared tomorrow, without warning, you would keep every day you have ever recorded.

Four things we promise

1. Export is free, and always will be. Not behind a subscription. Not behind an account. It works with the network switched off and aeroplane mode on, because it is built from files already on your phone. You can check this today: turn off your wifi and export.

2. The format is public. Every archive you export contains a complete specification of the file format and a short program that reads it. The same specification is published on our website, and will stay published. You do not need our software — or our permission, or our existence — to read your own data.

3. Your phone archives itself. The app writes a complete copy of everything to a folder you choose, in your own iCloud Drive or Files, on a schedule. You do not have to remember to do it, and it does not ask us for permission. If you stop opening the app entirely, the copy is still there.

4. If we wind down, you get a whole season of warning. Not less than 180 days, and always including at least one complete ski season during which everything still works. Announced in the app and by email.

Ninety days sounds reasonable and is not, for this app specifically. Somebody who rides one week in February may not open Powdium again until December. A notice posted in May and executed in August would be seen by nobody until it was already over — they would open the app for their next trip and find it gone, which is precisely what happened to snoww's users and precisely what this document exists to prevent. So the commitment is expressed in seasons, because that is the unit this app is actually used in.

Through the whole notice period, everything keeps working: recording, uploading, downloading, and export.

What we cannot promise

We are not going to pretend otherwise.

• That this app will exist in ten years. Most do not.
• That the servers stay up indefinitely. They cost money, and money runs out.
• That the social side survives us. Who you follow, your leaderboard positions, other people's activities — that lives on our servers, and if they go, it goes. We can hand you your own data. We cannot hand you a community.

Why you do not have to take our word for it

Promises are worth what the person making them is worth, and you have no way to assess that from an app listing. So the important part of this is not a promise at all — it is how the thing is built.

Your raw recordings live on your device permanently. The server holds a copy for the social features and for getting your history onto a new phone. That is the opposite way round from most apps, and it is deliberate, because it means the worst case is survivable:

If we vanished tomorrow with no notice, no email and no explanation — the account gone, the website down, nobody answering — you would lose the leaderboards and the people you follow, and nothing else. Every run, every day, every metre of vertical would still be on your phone, still exportable, in a format we have already published.

That is not a commitment we could break by behaving badly. It is a property of the design, and you can verify it yourself in about a minute.

What a wind-down would actually look like

If it comes to it, this is the sequence:

1. Notice, at least 180 days and spanning a full ski season, in the app and by email.
2. Everything keeps working through the notice period. No feature is switched off early, and nothing is put behind a paywall on the way out.
3. A reminder to export, more than once, with the button in front of you rather than buried in settings.

4. The format specification and the reader stay published afterwards, at three addresses, because one is not permanence:

   • the public repository, for working use;
   • Software Heritage, which archives public source code permanently and is backed by INRIA and UNESCO — it harvests the repository automatically, so this needs nothing from us at the time it matters most;
   • the format specification as a versioned PDF on Zenodo, with a DOI, run by CERN.

   None of these depends on us paying a bill, which is the entire point. A promise to keep a website up is a promise to keep paying for a website.

5. The reader is open source under the MIT licence — already, today. Not "on wind-down". A promise to open source something later is worth whatever the company is worth at the moment it stops existing, which is nothing. It is published now, so it is a fact rather than an undertaking, and reading your own data has never depended on us at any point.

Why this document exists

An app called snoww tracked skiing for tens of thousands of people between 2016 and 2024. In April 2024 its servers stopped answering. There was no notice, no final email, and no way to get anything out. People with seven years of ski days lost all of them, and are still annoyed about it, which is fair.

Nobody set out to do that. It is what happens by default when the data lives on a server and the money runs out — and the money running out is ordinary. The only unusual thing would have been planning for it in advance.

So we did. That is all this is.

This commitment takes effect the day Powdium first accepts money, and applies to everybody from that day — free accounts included. It is linked from the purchase screen for the Founder's Pass, because a lifetime promise is only worth what the wind-down plan behind it is worth.

If this page ever contradicts what the app actually does, the app is wrong and we want to know.