I keep coming back to this idea of “alive software,” which I think is software that’s…well, alive, lol.
Basically that means software that’s updating itself in real time based on how it’s being used.
There’s a couple layers i see to this - one that’s simple and one that’s not.
simple aliveness
I’m building this orchestrator tool for all my claude code sessions called Boo. It’s a fork of Ghostty, hence the boo.
It’s basically a workspace and session organizer/manager:
There are workspaces on the left and each workspace can have multiple sessions, and then there’s the Boo orchestrator level, which is basically the boss / chief of staff. In that layer I plan out my day and, based on the conversation, it then creates and updates and removes workspaces and sessions.
I’ll share more on how that works in a later post or something, but for now I wanna focus on the latest little addition which is here:
That’s basically a live ‘update’ feed of what’s happening in the worksessions that’s being written and rewritten as i go through the day.
The idea and hope here is that it can sorta be a lil sidekick buddy that does things like:
- keeps me honest by telling me what I’m working on relative to what I said i was working
- notifies me when things need my attention
- remind me about things i said were important that aren’t even showing up on my workspaces
There’s infinite potential for how that could grow and evolve, but the basic idea is - its non static. it is alive.
also the structure of the software could be alive
The other way that the software could be alive is that it could literally recreate itself in terms of how it looks and feels and functions over the course of the day that I’m using it.
So if I want to go into a deep focus on one project, the current option is FOCUS MODE - every workspace dissapears but the one I’m working in.
But what if i could ask it to create an entirely new mode based on the particulars of what I’m needing to focus on. Or what if it suggested a whole new mode and asked if I wanted to do it?