👋 I rebuilt one of my oldest projects

Since my last update, I definitely haven’t been sitting still.
There’s been a long (and short) list of ideas I’ve been vibing with, some practical, some just fun to explore. At this point, if vibe-coding counts as a hobby… it’s probably one of mine 😅

Alongside a few new client projects and some personal experiments, I kept coming back to something older. A project that has already resurfaced twice before, and that felt ready for a proper re-imagining.

One of the clearest ways to see how fast technology is moving is to revisit your own old work.

This idea started as one of the very first things I shared on my blog (back when it was still on Medium). I tried bringing it back once more later on. And now, this week, I rebuilt it again, this time as an interactive tool that actually lets the idea breathe.

Seeing those three versions side by side is honestly wild.
Not just the technical differences, but how much more alive the concept feels each time.

Back then, I was already thrilled that the “movement of music genres” visualization ended up being used by a data-viz community in their magazine.
The funny part? Under the hood it was all stitched together with slightly painful BI tooling (hello Tableau…), plus a lot of extra design work layered on top.

The second attempt came somewhere in early 2024.
That version involved plenty of buggy Python generated by ChatGPT, a rough grid of Dutch chart data, and (again) quite a bit of Photoshop cleanup to make it presentable.

Recently, someone mentioned the design again and asked if I was planning a 2025 update. I wanted to… but it also felt like the tools we have now should let us push it much further:

• make it more visually refined, something closer to a gallery piece

• make it interactive and engaging

• give people choices in how they experience it

So I started wondering:

What if you could highlight your favorite artist?
Switch between color themes inspired by Dutch modernism or MoMA minimalism?
Toggle between a square poster and a wide landscape?

I gave it a shot… and I’m pretty happy with how it turned out.

Give it a spin and let me know what you think.

^{Quick disclaimer: even with all the different methods I’ve explored, genre labels still come down to song metadata, and genres are never fully objective anyway.}

^{For the ‘By Label’ mapping, probably the most complicated, I’ve applied quite a few rules to group releases into the major labels. Still, one might slip through now and then. That’s why I’ve added the option for you to overwrite the label directly on the canvas, if needed (use wisely & refresh :-)).}

See you next time! 👋