Why I Gave People Four Ways to Destroy Things

The Easy Answer Was One Method

When I was designing Unheavy, the core loop was clear early: write what you're carrying, then destroy it, then drift. The writing and the drifting both felt settled. The destruction part kept pulling at me.

My first instinct was one method. Simple, clean, consistent. Everyone gets the same ending. I built a shredder prototype, it felt right, and I almost shipped that.

Then I noticed something. When I thought about my own use cases — the things I'd actually want to write and destroy — I wanted different things for different content. Not because I was attached to variety for its own sake, but because grief and rage are not the same experience. Something you've been carrying for three years is not the same as something that happened this morning.

Different emotional states seem to call for different kinds of endings. I decided to bet on that.

What a Destruction Method Actually Signals

Each of the four methods in Unheavy is designed to feel genuinely distinct — not just visually, but in what it asks of you and what it communicates back to you.

The shredder is methodical. Steady, systematic, complete. It cuts the thing into pieces and accounts for all of them. Good for anxiety — the kind of thought that accumulates in layers and needs to be gone thoroughly.

The 50,000 PSI press is immediate and total. It doesn't fragment — it obliterates. One motion, no remnant. For things you just want gone without ceremony.

The incinerator is transformation. The thing burns and becomes something else — smoke, ash, heat. For things that feel too significant to cut up. It changes form rather than deleting.

The tesla coil is chaos that resolves into silence. The electricity is wild, then it's over. For the things that feel electric and hard to contain — the kind that crackle when you try to hold them still.

The method of destruction isn't just mechanics. It's a cue to your nervous system about the nature of the ending.

The Research on Perceived Control

There's a consistent finding in the psychology of stress and coping: perceived control matters, sometimes independently of whether that control is doing anything mechanically useful. When people believe they have agency over how an outcome unfolds — even a bad one — they process it differently.

Ellen Langer's 1975 work on the illusion of control (Journal of Personality and Social Psychology) showed that even perceived agency over random outcomes changed behavior and experience. People bet higher on dice they threw themselves than dice thrown for them, even though the odds were identical. The sense of choosing matters to the brain.

I'm not claiming there's research specifically on choosing destruction methods. But the broader principle — that 'how' you end something affects what the ending means — felt reliable enough to build on. The choice of method is a small act of authorship over the ending. 'This is what this deserves.' That tiny judgment seemed worth preserving.

The Feature I Almost Built (And Killed Immediately)

At one point I sketched out a version where the app would analyze what you wrote and recommend a destruction method. NLP sentiment analysis — detect anger, suggest the press; detect grief, suggest the incinerator. Adaptive, personalized, technically interesting.

I killed it in about ten minutes.

The first reason: privacy isn't just about where data goes. It's about what you allow yourself to write. If any part of you suspects your text is being read — even locally, even by a machine with no memory — you write differently. You soften the edges. You perform instead of confessing. The whole point is to write the actual thing, not a curated version of it.

The second reason: a ritual in which a system decides for you is not a ritual you performed. It's an animation you watched. The choice is part of what makes the destruction yours. Handing that off to an algorithm removes the one moment of agency the whole experience is built around.

Why Four, Not Three or Twenty

You could imagine more methods. Freezing, dissolving, imploding. Why stop at four?

Partly because of the well-documented relationship between option count and decision quality. Iyengar and Lepper's 2000 study (Journal of Personality and Social Psychology) — the one with jam samples — found that presenting too many options reduces both engagement and satisfaction. More choices doesn't mean better choice. At some point the menu becomes the obstacle.

But more fundamentally: four forced me to be intentional. Each method has to justify its own existence as a genuinely distinct emotional register. If I couldn't articulate what a new method was for — what feeling it matched, what kind of ending it created — it didn't get built.

Four covers the main territory without fragmenting the experience. Methodical completion. Immediate erasure. Transformation. Chaotic release. If what you're carrying doesn't fit one of those, I want to know about it.

What I Actually Learned

I don't have usage data, so I can't tell you which method gets picked most. That's by design — nothing is stored or tracked. What I hear anecdotally is that people use different methods for different things, and that the choice feels meaningful rather than arbitrary.

The decision I'd make again: give people the choice and trust them to know what their content deserves. Don't make it for them.

The decision I'd maybe revisit: whether to give any guidance on what each method is for, or whether the current approach — let people discover it — is actually the right call. There's an argument that too much framing constrains the interpretation. There's an argument that context helps. I still don't know which one wins.

Building Unheavy taught me that the small decisions about how an experience ends carry more psychological weight than you'd expect. The writing gets the attention. The ending is what the brain actually remembers.