Launched this week
FairTrack
Kalorien tracken, 100% lokal & privat
3 followers
Kalorien tracken, 100% lokal & privat
3 followers
FairTrack ist eine kostenlose App zum Kalorien- und Ernährungstracking – aber ohne Cloud, ohne Werbung und ohne dass deine Essensdaten irgendwo gespeichert werden. Alles läuft 100% lokal auf deinem Gerät. Während die meisten Tracking-Apps mit Abos, Anzeigen oder fragwürdigen Datenschutzerklärungen nerven, setzt FairTrack konsequent auf Privacy First: keine Server, kein Tracking, keine versteckten Kosten. Einfach eintragen, behalten – und die Kontrolle über die eigenen Daten nie abgeben.
Love that everything stays on device, exactly what a calorie tracker should do. One thing that would make it even better for me: a quick barcode scanner that pulls nutrition data offline, since relying on a manual food database gets tedious fast. Curious how you handle that without any cloud lookup.
@gullu97869 Hey, thanks a lot – and glad you like the on-device approach!
Quick, honest breakdown, because there's a subtle distinction here:
The barcode scanner is actually already in the app, and the scanning
part runs fully offline. Detection uses ML Kit with a model bundled
into the app (no Google Play download), camera via CameraX – not a
single byte leaves the device for that.
What honestly does NOT happen offline: the nutrition data behind the
code. For the first scan of a product we don't know yet, we pull it
from Open Food Facts and then cache it locally (Room/SQLite). So the
second time you scan the same product it's fully offline – but the very
first scan of a new product needs a brief network hop. Claiming "no
cloud lookup at all" would be a little white lie on my part.
Why not just ship the whole database in the app? Because the physics
don't cooperate: the full Open Food Facts dump is several gigabytes,
and even a heavily trimmed slice (Germany/EU only, macros only) still
lands in the hundreds-of-MB range. On top of that, such a snapshot goes
stale every day, and coverage for small/regional brands is patchy –
loose produce without a barcode is out entirely. A "forever offline"
DB sounds great but would in practice be a bloated APK carrying a
half-dead dataset.
The realistic middle ground I'd actually recommend: ship a compact
offline core of the most common products and only hit the network for
the long-tail stragglers. That gets you ~90 % of scans offline without
putting the app on a database crash diet.
In short: scanner offline – yes, already done. Nutrition data with zero
cloud – only with honest trade-offs on size and freshness. Just say the
word if you'd like me to add the offline core set.