When your smartphone tries to be too smart

6th June, 2024

Back in the 1980s, the design expert Donald Norman was chatting to a colleague when his office phone rang. He finished his sentence before reaching for the phone, but that delay was a mistake. The phone stopped ringing and, instead, his secretary’s phone started ringing on a desk nearby. The call had been automatically re-routed. Alas, it was 6pm, and the secretary had gone home. Norman hurried over to pick up the second phone, only to find it stopped too.

“Ah, it’s being transferred to another phone,” he thought. Indeed, a third phone in the office across the hall started to sound. As he stepped over, the phone went silent. A fourth phone down the hall started ringing. Was the call doomed to stagger between phones like a drunkard between lampposts? Or had a completely different call coincidentally come in?

Norman tells the story in The Design of Everyday Things, the opening chapter of which is a collection of psychopathic objects from bewildering telephone systems to rows of glass doors in building lobbies that simply offer no clue whether to push or pull or even where the hinges are.

“Pretty doors,” jokes Norman. “Elegant. Probably won a design award.”

Reading Norman’s book more than three decades after its publication in 1988, it is striking how much the surface of things has changed. We no longer have to deal with incomprehensible telephone systems or VHS recorders. Good design is not a niche luxury now, but viewed as an essential part of business. The world has scrambled to imitate the success of Apple, one of the world’s most valuable and admired companies, which is built on good design: beautiful, easy-to-use products.

And yet I wonder. The aviation safety expert Earl Wiener is famous for “Wiener’s Laws”, which include “whenever you solve a problem you usually create one”. The truth is that modern devices may seem simple and easy to use, whereas they are in fact fantastically complicated. Those complications are elegantly obscured until something goes wrong.

I thought of Wiener and Norman recently as I arrived in Amsterdam, equipped with a Eurostar ticket barcode on my phone. Problem: the Eurostar exit barrier in Amsterdam is also the ticket gate for a variety of metropolitan rail services. As I tried to scan the barcode, the ticket barrier perceived my phone as a wannabe contactless credit card, and charged me for a local rail journey instead.

This is the logical result when two paths of technological improvement collide. Path one: replace a fussy magnetic strip on a paper train ticket with a more flexible barcode. Then display the barcode on a phone. Path two: replace a paper travelcard with a contactless travelcard. Then replace the travelcard with a more flexible contactless credit card. Then add the contactless credit card function to a phone. Problem solved, and, as Wiener declared, when you solve a problem you create another one.

Not that I’m complaining. Paying for stuff with a phone is convenient and, I am told, very secure. Travel is vastly easier, waving my phone to buy anything from tram journeys in Amsterdam to smørrebrød in Copenhagen, all at fair exchange rates.

And yet the point remains: technological improvements can have unanticipated consequences. One example, from Guru Madhavan’s new book Wicked Problems, is the theft-prevention system installed in Seattle rental cars by a car-sharing company. The system was designed to prevent cars being towed away by thieves. It disabled the cars remotely if they were detected to be moving with the engine off.

But beautiful Seattle is served by numerous ferry services and, in 2017, renters taking the boat found themselves unable to restart their cars when the ferry docked. An anti-theft system in a car caused major delays to a regional ferry system in a way that was obvious in hindsight but hard to foresee.

As the systems engineer Nancy Leveson argues, safety is a property that emerges from how an entire system fits together. The same is true for everyday usability and reliability. Both the ferry fiasco and my problems at the Amsterdam ticket barrier resulted from an unexpected interaction of two systems.

There is a way to switch off the contactless card function, but it is buried deep in the settings menu lest the phone seem too complicated.

Donald Norman argues that a well-designed product should make functions visible and intuitive: users should be able to grasp how it works, what their options are and get feedback about the results of their actions. That is all very wise, but our modern devices have managed to become so intuitive and versatile by concealing from us how they really operate. Laying bare the true complexity of the supercomputers in our pockets would boggle the mind. We cannot be exposed to how these things really work, lest we lose our grasp on reality. (See also: ChatGPT.)

And so we carry around these pocket miracles, and very useful they are too — until something unfortunate happens. It’s brilliant to have your tickets, keys, phone, address book and cash all in the same little box of delights, as long as you don’t drop the box of delights down the toilet. As Earl Wiener put it, “Digital devices tune out small errors while creating opportunities for large errors.”

Next time, I’ll print the ticket.

Written for and first published in the Financial Times on 10 May 2024.

Loyal readers might enjoy the book that started it all, The Undercover Economist.

I’ve set up a storefront on Bookshop in the United States and the United Kingdom. Links to Bookshop and Amazon may generate referral fees.

Pin It on Pinterest

Share This