Golden Krishna
Listen to podcast episode

“There’s an app for that” used to be a blessing—now it’s a curse. The proliferation of touchscreen interfaces for every purpose are the problem, not the solution. Golden Krishna explores a world where designers think outside of the smartphone to build practical, delightful products.


Interfaces used to be the solution. “There’s an app for that” was music to our ears.

But now interfaces are overused. They’re getting in the way. Instead of pragmatic solutions, we’re building clunky apps.

The solution is not a better app. The solution is no app – no interface.

Example: an app to open your car. Krishna explains how such an app can take up to 13 steps to open your car: take out your phone; unlock it; close last app; find car app; open app; wait for it to load; so on…

Instead, all you want to do is open the car door. So all you should have to do is open the door. Reach out, pull handle, and done.

Maybe the car senses its you because of a key fob in your pocket. Or maybe there’s some biometrics involved. But the point is it reduces friction.

An interface is merely one of many possible tools to accomplish a task. Consider all ways to solve a design problem. Only use an interface if it is genuinely the best way for a user to accomplish a task.

How did we get here?

We’re interface addicts. Krishna gives a long list of products that have been “re-invented” with interfaces where there was no need for one. Cars. Fridges. Trash cans. Restaurants. You don’t need screens in these places. And yet…

User experience is NOT UI. User experience is about creating a smooth, intuitive, enjoyable experience for the user. Sometimes it requires a UI. But sometimes it shouldn’t – but has one anyway!

Calling the role “UI/UX” is damaging. It blurs the lines between what should be two distinct disciplines.

“Blurring the two disciplines upon hiring designers has played an important role in our approach to creating technological experiences… The result is that much of our lives today is dominated by digital interfaces asking us to tap, touch, swipe, click, and hover because that’s what people were hired to create… When you hire someone to generate UI, you won’t get new, innovative solutions. You’ll get more UI, not better UX.”

Krishna has two excellent lists to drive the point home. One is properties of UI, which include navigation, menus drop-downs, buttons, links, windows, rounded corners, updates, search fields, hover states, iconography, colors…. so on.

The other is UX, and includes people, happiness, understanding needs, solving problems, pleasure, personality, satisfaction, and effectiveness.

Confusing the two is the root of our problem. Moral of the introduction: UI is not UX.

Ads have ruined company priorities

It used to be that we design for the best possible experience. But then “free” came around. Nothing is really free, however – you may not pay for it but you do with your data to improve ad targeting.

As a result, the incentives for the creators of these products is not to create a better experience for you. It’s to keep you glued so they can increase ad revenue.

Naysayers will say that keeping you glued is not a bad thing. You wouldn’t be glued to something you don’t like. That’s true, though it’s true the way folks prefer sugar to vegetables.

Ad-based businesses drive designers down one path and one path only: keep people on the platform and looking at ads. It doesn’t matter if there’s a better experience that requires less (or no) time using your interface.

“When Twitter’s powerful tools allowed an increasing number of third-party services to elegantly extend Twitter’s platform, for example, analysts reacted with concern. Investment arms expressed caution because a quicker route to getting things done allowed 14 percent of Twitter’s user base to avoid strategy ad placement”

Krishna shares a powerful anecdote of a Facebook News Feed redesign which resulted in users being too efficient. The experience was too good! Users were able to get in and out of Facebook quickly. They reverted the design in order to keep folks on Facebook for longer.

“As a Citigroup analyst boasted about Facebook before it went public, “It has a greater potential value than Googl because it was beating the other giant in ‘time spent.’” For many of these companies, it’s not necessarily about fewer or more ads, it’s about interface addiction.”

Krishna is at his best here, laying out his manifesto.

“I believe our job as designers is to give you what you need as quickly and as elegantly as we can. Our job as designers is to take you away from technology. Our job as designers is to make you smile. To make a profit by providing you something that enhances your life in the most seamless and wonderful way possible.

The path of addiction for profit pursued by some of our largest technology companies and encouraged by their investors is not good for us, or the next generation. Some of these services have become more addictive than alcohol or cigarettes, and can make us feel worse about ourselves even when we use them.

It’s time for us to pursue a new course.”


Studies show we’re unable to focus on more than one thing at a time. Multitasking is a myth. All you’re doing is rapidly task-switching—focusing on one thing at a time, then the other, then the first, then the other… which diminishes your ability to do either.

Our devices are incredibly distracting. They’re designed to be so. But that results in worse cognition, worse productivity, and missing out on our everyday life.

Another big distraction? Light. Screens.

Everyone has a circadian rhythm that aligns us to the sun. We get tired at sundown. Wake up when there’s light.

But with artificial light, we become alert. (Blue and white light suppress melatonin production).

“Computers have D65 or higher screens; the high color temperature seems to suppress melatonin when used at night; and that suppressed melatonin might be throwing off your internal sleep cycle and possibly even preventing your body from fighting off some types of cancer.”

Krishna’s point here is that NoUI products don’t distract – they just get the job done. And they don’t ruin your sleep and health.

Principle 1: Embrace Typical Processes Instead of Screens

The fragility of phones is a sign of bad design. We know people drop their phones. We know they get wet. We know we want longer battery life. And yet, phones are thinner with smaller batteries, not waterproof, and shatter-prone. Why aren’t phones designed to handle that? This is bad design, a failure to prepare for real behaviors.

“Camp Grounded, run by Digital Detox, provided Craig with a getaway from tech to remember what it was like to live without hashtags and with human contact—bond, share, and have fun with roughly two hundred adults in the wilderness… Craig was glowing from having paid money for someone to take away his technology. Take away his notifications. Take away his ringtone. Take away his phantom vibrations.”

The bad solution to this problem: smartwatches. It just pushes the notifications and interface onto our wrist.

Principle 2: Don’t serve the computer

Tech should figure out as much as it can about you and use that information to require less input and work from the user. It can do this using sensors – whether they are the ones built into your phone or something additional (and small, unobtrusive) that you wear or carry.

Example: headlamps. You want it to be bright when you stare into the abyss. But when you’re looking at a glossy map or your friend, you want it to dim. The old solution is to click a button and tell the machine what you want. Not a digital interface, but still an interface. A better solution? Put a light sensor and proximity sensor on the headlamp. When you’re staring at something far away and dark, it gets bright. Something close and less dark? Then the light dims slightly.

“Sensors are one way to provide richer information for machines. They can seamlessly enable a machine to read the needs of the outside world without a submit button. And there are many ways to gather the signals that lead to machine input.”

Principle 3: Adapt to individuals

Data science and algorithms allow computers to adapt to individuals more than ever: to predict and serve what you want when you want it.

This philosophy is much more Google Now or Nest-like than Apple or Siri-like. Google Now and Nest predict what you need and serve it up. Siri just responds to requests. Whether or not Siri responds well is irrelevant. The point is that Siri must be prompted.

Case Studies

“Great companies offer their customers the best possible solutions, whether they have a graphical interface or not.”

Krishna has a few great case studies of companies that solve problems without interfaces. Here are three great ones.

Lockitron. The first ten product was an app you launched to unlock your door. Yuck. Second get figured it out. It used Bluetooth to recognize when your phone was by the door and unlocked it automatically. No need to take your phone out of your pocket. Just open the door. Boom!

Mazda 929. The problem: your car gets agonizingly hot on a sunny day. The solution? Also the sun! Mazda put solar panels on the roof and those panels would power a fan inside the car after the car interior passed a certain threshold. Simply yet incredible.

Goodyear self-inflating tires. Instead of an app or a gauge, the tires sense when air pressure is low, and when the car is moving they open a small tube that allow them to reinflate themselves.

There are more amazing case studies in the book.


Challenge 1: privacy

For a machine to serve up whatever you want as you want it, it needs to know a lot about you. This book is many years older than the Facebook Cambridge Analytica scandal but still sensitive to the fact that folks don’t just want to hand over their data.

Possible solutions:

  1. Transparent, readable (ie: short), easy-to-understand T&S which make it easy for people to understand what data is being handed over, how it is being used, and how it can be controlled.
  2. Only hold data for as long as its necessary, and then to permanently delete it (and to communicate that fact).
  3. Show all data stored and option to delete all data anytime

Challenge 2: Automatic solutions are terrible.

Some are terrible, yes. Clippy for instance.

But the good ones are so good we hardly notice them. Like automatic sliding doors or airbags in cars.

So yes – automatic solutions are hard. But when they work, they work, and the market loves them.

Challenge 3: Dependence New Tech = Fragility

The more tech you depend on, the more likely it is something will go wrong and leave you stranded.

Sensors are a possible solution here. Detect something is broken and begin taking steps to solve it. Or an interface – only as a backup, Krishna is quick to remind – but an interface to troubleshoot and solve.


As he concludes, Krishna gives a fairly obvious caveat: his book’s point is not that NoUI is ALWAYS the ONLY solution – but it is USUALLY the BEST solution, and should always be considered first.

An interface is merely one of many possible solutions. If it is the best one, then use it. But if it’s not, then don’t rely on it as a crutch.

Buy The Best Interface is No Interface on Amazon