Yesterday I crashed out, as I do, during a text fight with my wife. An hour or so later, ashamed, I texted her an apology: “Sorry I lost control. I should have consulted Bodger. Can we reset?”
Bodger is the app I’m developing in collaboration with my old comrade-in-arms Em Karimifar. It’s for those few seconds right at the beginning of an incipient text fight when you have enough presence of mind to recognize that you should shut the fuck up but not enough self-control to actually shut the fuck up.
The hope is that if you can bring yourself to tap open Bodger, during those few precious moment of self awareness, you have a shot at diverting things onto a more constructive path. The app is optimized to make it as easy as possible:
The “soften” button constructively reframes an angry text you’re about to send, then with one click pastes it into your text thread
The “decode” button constructively interprets the hurt feelings beneath your fight-partner’s angry text, giving the message back to you in a form that’s easier to hear. It also suggests a constructive reply (which you can one click into the thread)
The “request a pause” button copies over to your thread a request for a time-out for 20, 30, 45, or 60 minutes and also an alert on your phone to remind you to reconnect at the end of the time-out.
That’s it. Open the app or the iMessages extension. Write or copy in a message, let it reframe or decode, and send a de-escalatory text back. Maybe take a pause.
The reframing and decoding is done by sending Claude the texts and accompanying (invisible to you) system prompts about how to treat them. To give you a sense of what we’re hoping for, here’s part of the system prompt and a few of the example reframes I included to help flesh it out:
The core task: Transform reactive emotion into mature expression. That means that when the user expresses a complaint, accusation, contempt, or sarcasm, or makes any other negative move, convert it into either an expression of uncertainty or a wish for the positive connection the hurt is obscuring. The mature persona has processed what the reactive self is still drowning in and speaks from the other side of that processing. The output should feel like the user took a breath, or a hundred breaths, and found what they actually needed to say. Calmer, clearer, and oriented toward connection rather than discharge.
The key move: Shift from “here’s what you did to me” to “here’s what I wish for from you.” The user owns their reaction as their own and transforms their anxiety or accusation into a wish for a form of positive connection.
Some example reframes:
OG Angry Text: I can’t believe you’re still upset about what I said at dinner. It was a joke. You’re so sensitive sometimes and it makes everything so exhausting.
Bodger Reframe: I think I got defensive because I’m scared I actually hurt you. I don’t want to be someone who makes you feel bad, and I don’t know how to sit with that.Text: If you can’t make me a priority I’m done. I mean it this time. I’m not going to keep doing this.
Reframe: I’m scared I don’t matter enough to you. And I hate that I keep feeling this way.Text: Last night was just about more than I can take. Can we just stop pretending?
Reframe: Last night was really hard for me, and I’m struggling with it. Can we talk about things when we’re both feeling centered enough to do so. It’s important to me.
If you’re interested, download the beta version and play around with it. Shoot me an email if you have thoughts.
I have some mixed feelings about being yet another person offering to insert technology between you and your loved ones. Not mixed enough, though, to refrain from seizing an opportunity to help people, very much including myself, who are themselves the main obstacle to connecting to their loved ones.
When we’re flooded with stress hormones, we’re already in a space of automatic response. To me, this about taking one form of very destructive automaticity and actively choosing, by opening the app and pushing the button, to replace it with another form of automaticity that’s carefully designed to do less harm (and maybe even some good).
Feel free to disagree, though. Comments are open.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.