I was at happy hour with my wife one Saturday. I’d recently launched Bassists.com, and I was riding high and already thinking about what to build next.
That’s when I thought about my old notebooks. The ones filled with ideas I’d shelved because they felt too big at the time.
One of the most exciting ones came back to me: an interactive music connection map.
The music connection map I couldn’t build for years
Years earlier, I’d drawn a version of this on a four-by-eight foot whiteboard. I started with Miles Davis’s “Bitches Brew” in 1969 and traced the musicians outward: Joe Zawinul to Weather Report, Weather Report to Jaco Pastorius, Jaco Pastorius to Joni Mitchell, and on and on. One connection led to another, which led to another, until I’d connected “Bitches Brew” all the way to Madonna.
It was a mess of lines and arrows, and it was beautiful. You could see how much of modern music was connected through the musicians who played in each other’s bands.
When I originally conceived it as a digital tool, I talked to developers. We couldn’t figure out how to build it with the technologies of the day. The idea stayed in the notebook.
Sketching an interactive music discovery tool at a bar
Sitting at the bar, it hit me: that discovery map could work inside Bassists.com. The site already had the data. Bassists, bands, albums, all connected through real relationships. The map was already there in the database. It just needed a visual layer.
I asked my wife for her notebook and sketched it out right there. Jaco Pastorius in the center, with lines out to Weather Report, Joni Mitchell, and the other bassists and bands connected to him. She looked at the sketch and understood it immediately.
I could see it would work. I couldn’t wait.
From bar sketch to working D3 prototype in 40 minutes
After happy hour, I sat down with Claude Code. I described what I wanted in detail: an interactive map using the D3 JavaScript library and the existing site data. You click a node, and it expands. Animated. That feel-good exploratory interface where you keep clicking because you want to see what’s next. I shared the sketch.
Forty minutes later, I had a working prototype. I spent a little time refining the UX, then worked with Claude Code to integrate it directly into the site. We set it up so the map updates automatically whenever a new bassist, band, or album is added.
From a paper sketch at a bar to a working tool on a live site. In a few hours.
Why removing friction changes how you think about building
That’s the part people fixate on: the speed. And the speed is real. But that’s not the point of this story.
The point is what changes when nothing feels daunting anymore.
That whiteboard drawing sat in my head for years. I’d talked to talented developers about it and we couldn’t crack it. Not because they weren’t good, but because the tools weren’t there yet. So the idea sat, along with a bunch of others.
Now the friction is gone. The gap between a big idea and a working prototype has basically vanished. You stop filtering your own ideas. You stop thinking “that would be cool, but I could never build it.” You go back to the notebooks.
I’m driven to push AI with these old ideas and just see what happens. The explore tool was the first one. It won’t be the last.
AI wrote the code. But experience drew the map.
