Humane functionalism

I’ve released humane-functionalism, a skill for AI coding agents that applies ideas from Danish functionalism to UI and UX.

I wanted it to start with the person using the interface and what they’re trying to do. That means looking at the hierarchy, how comfortable the controls are to use, whether the feedback makes sense and how somebody gets back when something goes wrong. It works with the product’s existing character, then checks the changes against that task.

The inspiration comes from Danish architecture and product design, with related lessons from Danish Modern furniture. Arne Jacobsen’s attention to the complete experience, from a building to its smallest interior details, translates into navigation, typography, and controls that belong together. Poul Henningsen’s care for how light is experienced informs the skill’s attention to visual comfort. Kaare Klint’s study of human proportions brings it back to the person using the thing.

These are my interpretations for software, and I’ve put the connections in the research notes. They include Børge Mogensen, Kay Fisker and Vilhelm Lauritzen, with lessons in construction from Hans J. Wegner and Poul Kjærholm too. There’s plenty of room for colour and expressive form in it; I’m interested in how those choices help the experience.

It’s available in my skills repo. Install it with:

npx skills add kemiljk/skills -g --skill humane-functionalism

Then give your agent a screen or flow and ask: “Use humane-functionalism to improve this flow.”