A MANIFESTO ON HUMAN-MACHINE CONNECTION

THE ROAD
IS NOT
A SCREEN.

When does digitization improve the car, and when does it weaken the relationship between human and machine?

Porsche 718 analog cockpit — fixed physical controls and manual shifter
Porsche 718 — analog cockpitPhysical controls spatially fixed, haptic, and executable without visual attention — the ergonomic standard racing derived.
Tesla Model S 17-inch touchscreen interior
Tesla Model S — 17" touchscreen, 2012Climate, audio, navigation, and drive settings collapsed into a single glass panel. Fast to update. Slower to operate at speed.
Scroll to begin
§01

NOT SCREENS
VS. BUTTONS.

That framing is the mistake everyone keeps making.

Task completion time
Touchscreen2.3s
Physical control0.4s

Screens belong in cars. Navigation, rearview camera feeds, battery state-of-charge — these are inherently visual and dynamic information tasks. They are appropriate use cases for a display. That is not the argument.

The argument is about what got moved into the screen: climate control, audio volume, wiper frequency, seat heating. Functions that previously required under one second of muscle-memory execution now demand a visual search, a mode selection, a tap, and a confirmation. At 100 km/h, a 2-second glance covers 55 metres of road without driver attention. NHTSA research identifies 2 seconds as the threshold at which crash risk begins to compound.

The decision was not ergonomic. A single integrated display eliminates 30–40 discrete mechanical components per vehicle. Software-defined controls can be updated post-sale, generating subscription revenue. Expanses of glass photograph as "modern" on a showroom floor. These are viable business-case justifications. None of them reduce driver cognitive load.

"Consolidating tactile controls into a touch display does not reduce interface complexity for the driver. It reduces component count for the manufacturer."
— the distinction that matters
§02 — evidence

THE DATA

The AAA Foundation for Traffic Safety and NHTSA's Visual-Manual Interface Guidelines provide the empirical basis for this argument. The findings are not contested. They are largely ignored by production design teams.

0SEC
0SEC
0SEC
40SEC

Mean eyes-off-road time for common infotainment tasks — navigation entry, audio selection — across tested touchscreen systems

AAA Foundation, 2017 — Measuring Cognitive Distraction
2.5×

Increased crash risk when a driver's eyes leave the road for more than 2 seconds, relative to an attentive baseline

NHTSA SHRP2 Naturalistic Driving Study, 2016
12SEC

Maximum cumulative glance time per task before crash risk becomes unacceptable — the threshold most touchscreen systems exceed

NHTSA Visual-Manual Interface Guidelines, 2013
The automation paradox

Level 2 automation introduces a documented second-order problem. MIT AgeLab research (2019) found that drivers using hands-free highway systems took their eyes off the road more frequently — not less — than drivers in manual mode. Mean secondary task engagement increased substantially when the vehicle was handling lateral control.

The mechanism is well-understood: reduced perceived risk lowers attentional vigilance. The infotainment screen functions as the nearest available secondary task. Partial automation and large touchscreens, combined, are not a neutral pairing.

§03

The iPhone
model doesn't
apply here.

Tesla Model S 17-inch touchscreen dominating the cabin dashboard
The road demands kinetic, spatial, safety-critical attention

The iPhone displaced the physical keyboard because the smartphone's primary function fundamentally changed — from single-purpose voice communication to a general-purpose information terminal. A fixed physical layout was architecturally incompatible with that shift. A reconfigurable touch surface allowed the same hardware to serve as a camera, a navigation device, a payment terminal, a document editor. Flexibility was not incidental — it was the enabling feature.

The automotive case does not share that precondition. The primary functions of a vehicle — directional control, velocity management, occupant comfort — have not changed in character. The tasks a driver performs repetitively (adjusting cabin temperature, modifying audio output, activating wipers) have a stable, non-variable structure. They do not require interface reconfigurability. They require sub-second, eyes-free execution with zero error tolerance. The TRL (2020) study found that touchscreen distraction produced longer reaction time delays than either hand-held phone use or cannabis impairment at the legal limit. Flexibility, in this context, is the wrong optimization.

SmartphoneStationary use
Vehicle70+ mph, shared road
Error costRetry
Error costCollision
2007

Apple introduces the iPhone

Three products in one — phone, iPod, internet device — all controlled through a multi-touch display. The screen became the condition that allowed the object to become flexible.

2012

Tesla Model S launches

A 17-inch touchscreen becomes the visual and functional center of a car cabin. NVIDIA described it as the largest touchscreen ever installed in a passenger car.

2012–2023

The industry copies the surface

Automakers saw that consumers associated large glass surfaces with intelligence and modernity. They copied the visible screen-centered result without copying the deeper system.

2024

VW restores physical controls

After years of criticism, Volkswagen moved toward restoring buttons for essential functions. The market corrected the assumption that fewer buttons means better design.

§04

Different bets.
Different outcomes.

Every major manufacturer made a different call on the screen-versus-control tradeoff. The market is now reporting back on who got it right. This isn't about analog nostalgia — it's about which interactions are actually faster and more reliable when digitized, and which got slower.

Precision machined aluminum control knob with knurled edge — tactile engineering
The Disruptor

TESLA

Strength

Genuinely rethought the car around electrification, software, charging, updates, and centralized digital control.

Challenge

Many people copied the visible screen-centered result without copying the deeper system behind it.

"A dashboard that photographs well but buries volume control three taps deep hasn't simplified anything. It's moved the complexity somewhere the camera can't see."
§06

What racing
figured out.

Racing driver's gloved hands gripping a carbon fiber steering wheel — human-machine feedback

Motorsport engineering has systematically resolved the same human-machine problem that road car design is currently mishandling. In a racing context, every interface decision is subject to performance validation: if a control increases task time or demands visual attention at speed, it is removed or repositioned. Brake feel conveys thermal state. Steering torque communicates grip margin at the contact patch. Throttle pedal response communicates rear axle traction. This is proprioceptive data — processed below the threshold of conscious attention, faster than any screen readout.

A current Formula 1 steering wheel integrates over 25 discrete physical controls — rotary switches, push buttons, shift paddles, a multi-position brake bias adjuster. Every one of them is operable by gloved hands without visual reference. This is not a nod to tradition. It reflects a rigorous engineering conclusion: under high cognitive load, haptic spatial memory outperforms visual search every time.

Driver Feedback Channels
Steering load
Brake pressure
Throttle response
Grip
Vibration
Weight transfer
Rotation
Sound

The road car conclusion

The F1 cockpit has stricter constraints than a road car — but the logic is identical. When a task needs to happen fast, without looking, without error, the physical control wins. Not because it's more sophisticated. Because it's in the same place every time, it responds with weight, and you don't need to find it.

Porsche kept physical HVAC controls in the Taycan. VW reversed course on the ID series. Both got rewarded for it — fewer complaints, better reviews, less distracted drivers. The ones still asking the question are doing better than the ones who stopped.

§07

Where we
actually stand.

The research literature provides a clear functional criterion: tasks that are visual, infrequent, and information-rich are appropriate for display interfaces. Tasks that are frequent, haptic in nature, and must be completed without redirecting visual attention belong on dedicated physical controls. This is not a matter of aesthetic preference — it is a question of human factors engineering.

The market correction is already underway. Volkswagen reversed its touchscreen-first strategy across the ID. lineup following documented usability failures. The UK's Transport Research Laboratory (TRL) published findings in 2020 showing that touchscreen operation impaired driver reaction times more than cannabis intoxication at the legal limit. These findings are on record. The design community is beginning to respond.

PASSIVEKINETIC
Slide right to join →

SLIDE TO ENGAGE

base44
Edit with Base44