Why Latr has no server
Every reminder app wants an account. Shipping without a backend turned out to be a product decision, not a shortcut. What it costs, and what it buys.
Almost every reminder tool starts the same way: sign up, verify your email, and now your data is on someone else’s computer. Latr does not have that step, and it is not because we ran out of time.
What “no server” actually means here
There is no backend. There is no database. There is no API. Your reminders are stored by your own browser, in the storage area the extension owns, and the free version of Latr makes no network requests at all — not for updates, not for telemetry, not for anything.
Even the share links work this way. When you send someone a reminder, the reminder is
encoded into the part of the URL after the #. Browsers never transmit that fragment to
a server, so a shared reminder is never seen by any machine of ours, and could not appear
in a log even if we kept one.
What it costs
It would be dishonest to present this as free of trade-offs.
Reminders only fire while your browser is running. This is the big one. chrome.alarms
does not run when Chrome is closed, and there are no extensions on mobile browsers. If
your laptop is shut at 10am, the reminder arrives when you next open the browser — and
Latr tells you it is late rather than pretending otherwise. A server with push
notifications would not have this problem.
Our answer is the calendar handoff: one click puts the reminder into Google Calendar or Outlook, which does reach your phone. It is the second channel, and it exists precisely because the first one has a known edge.
Sync is limited. Reminders can travel between your own signed-in browsers through your browser vendor’s sync storage, but that is a much smaller mechanism than a real backend, and it is opt-in.
There is no recovery if you uninstall without exporting. Nothing is held anywhere else. That is the flip side of the same coin.
What it buys
It cannot leak what it never collects. This is not a policy promise that depends on us continuing to be trustworthy — it is a property of the architecture. There is no database to breach because there is no database.
It costs nothing to run, so it can stay free. Most free tools are free because you are the product, or because someone is burning investment. Latr is free because serving zero requests costs zero dollars.
It cannot be shut down by us going away. If this project stops being maintained tomorrow, your reminders keep firing, because nothing about them depends on a service staying up.
Would we do it again?
Yes, though with clear eyes about the trade. For a tool whose entire job is remembering things about pages you visit, the browser is already where the relevant context lives. Shipping that data to a server would have meant asking people to trust us with their browsing in exchange for a benefit that mostly accrues to us.
The honest limitation gets stated on the landing page, in the store listing, and in the notification itself when a reminder arrives late. That is the deal, and it seems a better one than the alternative.