You cannot build an automatic grader for aesthetic taste, creativity, novelty, etc. because there is no correct answer to grade against.

Speeding up human processes is the least interesting things we can do with the power of AI.

The thing that stays durable and ungradeable is not getting better at solving problems that already have a known answer; it's choosing what to build and judging whether it's any good.

OKHST keeps OKHSL's hue and saturation and replaces lightness with that tone axis.

When you optimize a creative process for raw speed, you don't get better products. You get the same products, just sloppier, sooner and shortsighted.

Federating demands more overhead and operational diligence than any centralised team would, leaving chaos in their wake when federated eventually comes back around to centralisation.

While making becomes cheaper and perhaps deciding does too, accountability does not.

I don't think that there is a system without guidelines. Guidelines force us to take a concept and systematize it and figure out what the tokens are and what the naming structure is and how we actually think about it and talk about it.

A free, open-source color-identification system. Simple, memorable symbols map to base and combined colors so anyone can tell them apart.

No design system team can solve every problem at once, and any effort to do so will only dilute its impact.

Creative provenance is emerging in response to the growing presence of AI in how things are created.

Compound icons are effective at getting more specific. What are we adding? A phone, a folder, a user, a location.

See the element, learn its real name, and prompt your coding agent with precision.

Experienced people don't merely get better final output. They get more useful work from each prompt.

The real work of learning happens when you get into the messy part of the thing you're trying to learn.

Use this maturity model to assess your design system’s health and identify where to focus next.

Optimization improves what already exists. Judgment decides what should exist next.

When a tool can generate a thousand plausible screens before lunch, the scarce skill is no longer making one; it is knowing which one deserves to exist.

Over the last three years the browser quietly grew most of the toolkit I had in Quark, and almost all of it is plain CSS with no JavaScript and no layout library.

The more you move away from a singular, visual representation, the more room there is for confusion and greater cognitive effort for discernment.

A weekly live show from Dive Club about design, tools, and the people building software.

When you’re designing motion, you’re designing with time. Understanding the mechanics turns movement into meaning.

The steady state of a good designer is permanent dissatisfaction with how things are, which is an uncomfortable but essential fuel.

A product can be technically flawless and still fail. If anything in the experience creates a flicker of doubt, a confusing screen, copy that hedges too much, a moment where the user has to stop and wonder what's happening, the trust collapses.

The setup that worked for one team, one product, one platform starts to buckle under the weight of multiple verticals, competing release cycles, and a product org that wants to experiment with everything.

The best version of this component is not the one with the most clever positioning code. It is the one where the panel feels physically attached, behaves predictably, and leaves you with less code to distrust.