Hi HN.

Today I'm showing off Chow.fit, a plain-text FOSS food diary written in Gleam. You can self-host it with the Dockerfile or just use the one I host on chow.fit.

Why does the world need another food diary app? Well, I can't speak for the whole world, but as for me:

I used MyFitnessPal back in 2017-2019. It was mostly tolerable with an ad blocker. A month ago due to numbers I didn't like, I wanted to start tracking calories again. I quickly noticed that in the MFP android app, the barcode scanner is now a paid feature. It felt like paying to use my own phone camera. Their web app is slow and stuck in 2013, so I made my own web app which better aligns with my values and goals. I've lost 5% of my body weight!

I like to call an app like this "un-SaaS" because it subverts a lot of the usual SaaS playbook.

- Distraction-free. No AI, no ads, no tracking, no engagement-hancing e.g. no activity/news feed. It logs food and gets out of the way.

- All features except sync work without a subscription.

- An optional subscription enables sync. A year covers my hosting costs and buys me a T-shirt.

- Sync is local-first. You can still log food e.g. in an airplane. The app syncs up when you go back online.

- If you stop the subscription, the app keeps working without sync.

- Your data is plain text files on your device. When SparkPeople shut down in 2021, its users could only download their daily totals. They lost everything else. That can't happen with Chow.

- It uses public food data (FDC and OFF), not proprietary

- It can import from MyFitnessPal using a bookmarklet to get your auth token which stays local.

- It has a desktop app, Android/iOS app, and a CLI. None currently shipped, but you can build from source.

Folks who track what they eat: Let me know what else you'd like to see!