Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
I killed every houseplant I owned, so I built an app to stop doing that
2 points by gongshaojie12 29 days ago | hide | past | favorite | 3 comments
I've killed more houseplants than I'd like to admit. Not from neglect exactly — from the specific, stupid way I neglect them: I water everything on the same day because I can't keep track of which plant wants what, so half of them rot and the other half go thirsty.

So a few months ago I started building GrowBook. A plant care tracker. Flutter, offline-first, Android first.

I know. There are already a hundred plant apps. I want to be honest about why I built another one instead of just downloading one of those.

I tried them. The good-looking ones are subscription walls with a plant drawing on top — you can't add a second plant without paying. The free ones drown you in a feed and a shop and "plant coins." And almost all of them assume you're online, all the time, which is a weird assumption for an app whose entire job is to remind you to do a thing at home.

What I actually wanted was small and boring:

- A list of my plants with a photo of each real one, not a stock image of the species. - Per-plant watering / fertilizing / pruning schedules, because my snake plant and my fern do not live on the same clock. - A reminder that actually fires, even with no internet, even after I reboot my phone. - The whole thing usable without an account. Cloud sync should be a thing I opt into, not the price of entry.

That last point turned into the spine of the whole architecture. GrowBook is offline-first: everything lives in a local SQLite database (I'm using Drift), and the app is fully functional before you've ever seen a login screen. Sync to the cloud is a feature you turn on, not a wall you climb over.

I'm building this mostly solo, and I want to do it in the open — including the parts that go wrong, which so far is most of them. The next few posts are going to be very specific: the week I lost to Android's notification system, the light sensor Flutter refused to read, the AI plant identifier I had to rip out and rebuild. Real bugs, real dead ends.

Where I'm at right now, honestly:

- Users: 0 - Revenue: $0 - Plants kept alive by the app: a suspiciously healthy count, but I'm the developer, so I don't count.

If you keep plants: what's the one thing every plant app you've tried gets wrong for you? I'd rather build from your annoyance than my assumptions.

— building GrowBook in public, #1



I also build a plant app. Identification was never my pain point, placement was: a Tineke I had filed under "medium light" turned out to be sitting above 20,000 lux.


I found this article at indiehackers

https://www.indiehackers.com/post/growbook-i-killed-every-ho...

And your home page

https://www.indiehackers.com/gongshaojie12

Looks very interesting.

Got more?

My son-in-law has a moisture sensor (blue = good) In every plant. I'm more of a "finger test" person, but I have only one indoor plant and some asparagus in containers outside that I keep alive somehow.

Just one comment at this stage. Why complicate life with AI identification if they're your own plants? Who should know better, and KISS is a good idea.

Now, if you REALLY want to get complicated (and please, do not try this at home) have a photo decide if a plant is healthy and what might be ailing it.

Sorry.

And BTW GrowBook is out there in corporate services and for (surprise) baby growth records.

GOOD LUCK


URL?

Some photos of the corpses?

Indoor or balcony?




Consider applying for YC's Winter 2027 batch! Applications are open till November 2.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: