okay, to start, an analogy: long ago, a sing le person could build the best video game that existed. Adventure didn't require 200 developers and 18 months to get out the door. as technology improved, so did expectations. Adventure was my goal. Today's kids' goal is fortnite. —@steveklabnik

One of the things I keep coming back to is the value of visible seams. The idea that you can look at something and develop a mental model of how it works or how it is made.

The mmm.page editing controls (both mouse and touch) are nicely done. They pack a good set of options in while still feeling lightweight.

From: https://build.mmm.page/

Experimenting with gif frames in space.

Exploring some of the inception possibilities of three.js.

Texture warp with movable control panes.

Modify triangle drawn shape or texture source shape.

Warp network.

3D failure.

Getting closer to a workable distortion system, I think.

More triangulation and barycentric coordinate experiments.

Got a point to maintain it's position in a changing triangle using barycentric coordinates and d3-ternary.

Delaunay triangulation and texture warping debug. Still getting my head round this.

Triangulation progress. Some bugs still in there.

Getting closer to... something.

Not really what it is supposed to do.

The new project starts with BufferGeometry.

Flow - an experimental image editor that lets you set and direct pixel-flows.

From: https://flow.constraint.systems

I built a new desktop, complete with see-through keyboard shelf.

Interface for a digital library collection. Part of Ireland's exhibit at the London Design Biennale.

From: https://collection.dropeverything.net/

Flow prototype work.