Vibecoding
Build for the vibe, debug later
Start new thread

The 40 apps nobody opened are still in our bundle

by•

We built 111 one click apps for Arteza and 9 of them cover most of the runs. I've posted that number before. What I didn't expect is how hard the next step is.

Deleting one takes an afternoon. You check what imports it, whether a paying user has it bookmarked, whether pulling it breaks a shared component three live apps also call. Writing it took twenty minutes. Removing it takes four hours.

So they stay. Every one is something a new user scrolls past and something I have to keep working through model upgrades. Generation is cheap and deletion is full price, and nobody warns you about that half. Has anyone found a way to make cutting as fast as shipping?

33 views

Add a comment

Replies

Best

we hit this w/ content not code. 99 instagram posts since 1 aug and ive never deleted one, mostly bcoz i cant tell which ones still bring anything in.

what actually helped wasnt getting faster at cutting, it was making fewer formats in the first place. cutting stayed expensive, we just made less to cut.

numbers are here if useful

 Making less to cut is probably the real answer, I just don't like it. The code version is making every new app justify itself against the 9 that already carry the load, and I've never once run that check before building one. What I'm missing is your first bit, how you decide a post is dead. I have no signal at all for an app nobody opens, so all 111 look alive to me.

    On deciding it's dead: I look at whether anyone arrives there, not whether anyone views it. A page people only reach by clicking through from somewhere else is one that somewhere else is already carrying, and folding it in costs nothing.

Same shape on the page side, and what made cutting cheap was never deleting. A page nobody opens gets folded into the one that works and redirected at it, so the bookmark and the inbound links still land somewhere and it stops being mine to maintain. Cutting got fast once there was always somewhere to point.