Design's future is web-native

Ask a painter what their medium is and they’ll tell you oil, or watercolor, or acrylic. Beyond just what they paint on, the medium is the material they think in, which they’ve developed fluency with over years and whose affordances shape their craft.

Ong argued that communication technologies are more than mere vehicles for transporting thought; once matured, they actually end up structuring it. Writing, for example, was intended to preserve speech, but it ended up creating distinct possibilities for distance, sequence, abstraction, and analysis. Fully-realized media become part of the human mind’s working equipment, which we familiarize ourselves with to the extent that their influence can become difficult to see.

Designers have always had forms of media to specialize in. Print designers think in paper and ink. Architects and industrial designers consider physical form and material, and the constraints that accompany them. UI designers have spent the past few decades mastering proprietary design tools that imitate digital interfaces.

McLuhan conceived of forms of media as evolutionary, rather than as isolated entities. Most new media do not exist as wholly independent mediums; instead, they incorporate, reorganize, and expand the media before them. Writing puts the form of speaking into a graphic form. Photography renders images mechanically and can contain writing. Printing takes writing and makes it multiple and standardized, and can also include images.

All of these are examples of a history where media have increasingly incorporated one another into their make-up. Writing can describe an image, an image can contain writing, and film can incorporate both along with sound and theater. Each medium that evolves incorporates all these previous media, and becomes an environment for their existence.

In the past few decades, we’ve also witnessed the meteoric rise of a groundbreaking new medium, one that’s capable of absorbing all others, while also extending them with novel possibilities for interaction. This is, of course, the World Wide Web.

Something is shifting: the web is becoming a bona fide creative medium in its own right, and has matured into a continually evolving material, not just a destination or delivery mechanism. I would go further and argue that the web is the medium toward which all other media and design practices will converge.

The substrate is moving

This absorptive capacity is now beginning to reshape the tools with which designers work. For instance, when an interface is designed with an agentic coding tool like Codex or Cursor, prototyping in Figma can be avoided entirely, as usable code is already produced. The distinction between designing an interface and constructing it begins to dissolve: the representation and the functioning artifact increasingly unify.

When thinking in terms of a medium begins to calcify its practitioner’s process, the distinction between simply representing an interface and actually working within its runtime is not only technical. A static canvas that imitates a web interface encourages the composition of states, while designing within the very architecture of the web encourages the composition of rules, relationships, and behavior.

The change we’re seeing necessitates a broader scope of materiality. In Digital Anthropology, Daniel Miller and Heather Horst argue that digital worlds are neither less material nor less real than those that preceded them. Their materials are just different, and so much so that they seem immaterial and unreal. The Internet, as seamless as it seems, is a tangled web of infrastructures, standards, code, content, and devices, and the contexts in which people encounter them. Rather than freeing design from mediation, browsers give designers more direct access to the material system through which interfaces take form.

Whereas the material properties of most traditional media are fairly well-defined, those of the web experience continual innovation and evolution. Improvements in the quality, efficiency, or accessibility of paper, ink, printing presses, or photography, for example, can be expected, but changes to their fundamental set of affordances are less likely. This isn’t so for the web, which is still actively being constructed. Novel possibilities are unlocked through constant advancements in the languages used to write the web, the frameworks and libraries used to construct applications within it, and the browsers through which its pages are interpreted.

Pretext, for example, was a recent attempt at building a tool around perhaps the most humble material on the internet. Designers usually work with text by placing it in a box and adjusting it after seeing where the lines break. Pretext enables a web page to predict where and how the text will flow, what kind of space it will take, and how it can be adjusted before it reaches the browser. This makes it easier to build layouts that respond intelligently to different words, window sizes, and amounts of content, instead of treating text as something poured into a fixed frame. Its significance is easy to overlook precisely because it deals with such basic concerns as line length, spacing, and wrapping. It demonstrates how something as old and well-established as typography may receive completely new life through computation.

From emulation to embodiment

An image captures only one instance or one possible state - a still captured from an array of possible moments. Conversely, HTML, CSS, and JavaScript establish relationships that allow for multiple states. Code allows for layout scalability, interactivity, and animation, and thus, it makes content responsive to context. Browsers, hardware, networks, and users all provide and shape messages on the web, but the code provides the material out of which their message is woven, not a representation of something to reconstruct.

Emerging generative design software is already beginning to acknowledge this. Paper’s entire design environment is built on HTML and CSS as the bedrock, which closes the gap between the design surface and the runtime environment that it generates. Rather than designing within software that emulates browser environments, designers can design within the browser environments themselves. Handoff frustrations of old, including developers having to rebuild designs from scratch, are becoming obsolete.

Quiver extends this into graphic design by harnessing AI for SVG generation. Choosing to format generated logos as SVGs rather than flattened images (which image generation models would produce) isn’t just a formatting choice. SVGs are structurally different in that they’re scalable, editable, and programmable, and enabling AI to construct them allows for systems to be more easily extended and generated rather than frozen.

A similar shift is underway in motion design. After Effects established the dominant paradigm of arranging layers and keyframes on a timeline, while Flash extended that model with vectors, scripting, and interaction for the web. Later tools such as Jitter and Cavalry treated motion more procedurally, as relationships, behaviors, and live data flows rather than a sequence of manually composed frames. Remotion and HyperFrames push this further by defining animation through code and browser technologies and integrating AI natively, making motion parameterized, reusable, and generative rather than fixed to a single rendered sequence.

This is where design’s center of gravity is moving. In the years to come, we’ll see more web-native design tools developed to encompass print design, industrial design, and other design disciplines. Different design disciplines are developing a more symbiotic relationship with web primitives. The browser is shifting from a destination for their outputs to where the work actually happens. The implications haven’t fully landed yet in how most designers work, but the tools are catching up.

A medium that absorbs (and extends) others

What’s striking about the web as a medium is its hunger. It absorbs the media that need their own formats and means of distribution, but it doesn’t leave them unchanged.

Posters can now interact with their audience. Illustration can be navigable. Typography can reconfigure itself based on language, screen dimensions, and movements. A film can branch depending on the user’s input. A photograph of a product can serve as a configurator. Beyond transferring existing media onto a screen, the web provides them with state, behavior, and the ability to react.

Even spatial experiences, which were first simulated by specialized proprietary software, are now available through browsers because of WebGL and WebGPU technologies. Using well-documented technologies like Three.js and shaders, one can create cinema-grade, interactive experiences that can be delivered with just a link.

As Miller and Horst observe, there is constructive tension inherent to the digital because the reduction of data into binary is not a driver of homogeneity, but rather of great differentiation. It is this tension that informs how the web functions – where basic things such as URLs, markup languages, style sheets, scripting languages, images, and protocols allow radically different forms to inhabit the same distributable environment. Different artifacts like shaders, essays, stores, and games can all open in the same browser without flattening into the same kind of object, yet they are all flattened, in a sense, into links.

Interactivity makes the web more than just a platform where other media converge. While other media formats like prints, films, and graphics exist as predetermined compositions, in a web environment, the artifact itself may feature several states. These can respond to users in a number of ways, including scrolling behavior, mouse movement, user input, or geolocation. The user becomes both the message’s recipient and its shaper.

This is becoming increasingly true of physical design as well. Sneakers have launch experiences. Architecture projects have immersive webpages. Restaurants have online brand worlds. All of these things aren’t the physical object, but they’re part of how it gets presence and context in the world. Technological advancements such as VR, ray tracing, and gaussian splatting are even allowing us to experience physical design on the web in higher fidelity. We can still have print and physical media, but they’ll be the outputs of online models and instructions rather than static artifacts.

With these boundaries eroding, there’s no reason for individual design disciplines to remain compartmentalized. An identity can easily become an interface, a motion system, a data set, and a system for creating future instances. The web thereby proximates various media but also makes them interoperable. To design for the web means to define the ways in which the concept can exist in different forms without making any one of them absolute.

Web apps and interfaces will be the future home for design and designers because they’re practically frictionless. Even apps have significant friction. They require installation, storage, platform approvals, and user commitment before users can access the experience they offer. A website only requires a link. It can be shared in a Slack message, indexed by a search engine, embedded in a newsletter, and opened on any device with a browser. It’s crawlable, bookmarkable, archivable, and remixable.

This is where the meaning of designing for distribution is changed. The web should no longer be thought of as a medium where one publishes after designing content, but as one that transforms content into something accessible and alive. Apps may be more appropriate for certain use cases, but websites remain the front doors and showrooms, and sometimes even the products themselves.

What fluency means now

I’m not suggesting that other media or design practices will cease to exist or lose importance. Print, motion, graphic, and industrial design will remain relevant and distinct practices.

The change is more subtle: the web will become an underlying layer that connects all of them. Print designers will engage with web technologies for tooling and distribution purposes. Industrial designers will model browser-rendered digital prototypes. Motion designers will experiment with shaders and 3D libraries. Architects will exhibit their work in immersive web environments.

Ong calls the process by which technologies are absorbed by the way we think “interiorization”. This is the level of fluency required for web-native design. Web designers who have internalized HTML, CSS, SVG, Canvas, WebGL, components, and tokens empower themselves to think of responsive systems, states, behaviors, and relationships because those become possible in their mind.

Just as painters don’t need to produce their own pigments, this doesn’t mean that designers will need to become programmers. I am in fact encouraging designers to develop an understanding of materials that constitute the web, which will yield dividends in sharpening their craft.

Designers can no longer afford to treat the web as just a medium on which their work is exhibited; with every increase of responsiveness, programmability, and the connectivity of artifacts, the web increasingly shapes both how design reaches its audience and what kind of designs are possible in the first place. Its primitives will serve as a common language for different interfaces, identities, motion, space, and physical products.

The designers best prepared for this shift will be those who learn to think through that language, not simply translating static ideas into webpages, but conceiving work from the outset as a system of structures, behaviors, and possible states. As McLuhan famously remarked, “the medium is the message”.