[edit] Calm Technology Presentation

Given at Bell Labs in acceptance of the Claude Shannon Innovation Award on August 28th, 2016.

52 slides. 45 minutes.


Hi, my name is Amber Case and I’d like to talk to you about a concept called Calm Technology. I first learned about it 10 years ago, and the time for it is now.


How many of you have heard this quote from Cisco that 50 billion devices will be online by 2020?

Whenever I hear quotes like these I like to consider how they usually come across. The future will be so exciting! How many of these devices can our company own? There still seems to be this idea that at some point in the future we’ll have this perfect society complemented by automation, robotics and the “Internet of Things”.


I don’t think the same way. I like to ask the question of whether this kind of future actually sounds good, and what things might happen if we were to live in a world of 50 billion devices.


Let’s consider some products that already exist…


The smart watch. It gives you a very similar experience to your mobile phone right out of the box. Most smart watches ship with all notifications by default on.

This is a big issue, because you can’t just make the same experience for one platform (a phone) work for another platform (a watch!). Each device is different and has a different context! With a smart watch you get interruptions from another source!


Let’s consider what I keep encountering from big tech companies: the concept of a “smart fridge”.

The “smart fridge” might have a payment plan that monitors your diet and prevents you from opening the fridge if you ate too much food. But what if your friend comes over and has a diabetic attack? Maybe you or they can’t get into the fridge. Or maybe you get frustrated at the fridge for limiting your dining options and you just eat all of the junk food you left out on the counter. Or maybe the fridge beeps you every time one of your items goes bad. Like “Hello Dave, your bananas have gone bad!” But we don’t need a fridge to tell us our bananas have gone bad! The banana already has a skin that tells us that it has gone bad! The peel! It turns a different color! We don’t need a machine for that. And we don’t need a fridge that talks to us an interrupts our day.


Put these devices together and you get what I call “the dystopian kitchen of the future”. It’s a bunch of different technology with a ton of little beeps and alerts, all vying for your attention, and all vying for different software updates. Plus, it’s all written in different programming languages, many of which aren’t really supported anymore. Some of the companies go out of business and your devices don’t even work anymore!

But it gets worse. Heaven forbid having to “inherit” all of this smart home technology when you move into a new house. Instead of simple light switches you have all of these smart light switches! You have to be a system administrator to live in your own home! No one should have to be a system administrator to live in their own him! Or if you break up with someone while living in a “smart home”. Who has the passwords? Who gets the tech? Where did the original thermostat go? Did it get thrown away? How do we get everything back to defaults?


We live in an era of interruptive technology. The world beeps at us incessantly. We are asked to update the software on our Apple TVs before we watch a show, not right before we turn the device off. There are ways to deal with this technology overload. We don’t need an era of interruptive technology.


We need a Calm Technology.

A Calm Technology describes products that are there when we need it, not when we don’t, where our devices recede into the background and allows us to be human.


Where did this concept come from? Let’s go back in time to the 80s and 90s at XEROX PARC, a highly innovative and quirky research center based in Palo Alto, California.

You might know XEROX for their impact on the printing and copying world, but their main contributions to society came from their research center. The place was filled with interesting thinkers. Scientists, anthropologists, artists and engineers filled the halls. Conference rooms were furnished with beanbag chairs instead of traditional tables. Why beanbags? To solve a serious human problem. Engineers were constantly jumping up from their chairs and interrupting their work on the chalk boards. Technologist Alan Kay put bean bag chairs in because they’re difficult to get out of! This caused the momentum of the engineers to slow down, giving each of them time to get their points across. This unconventional space paid off. PARC gave us a number of crucial innovations, including what we now know as the Graphic User Interface (GUI) and modern desktop computing.

What caught my attention most were two research scientists, Mark Weiser and John Seely Brown. In the 80s and 90s they uses PARC resources to come up with a working model of the future. It was a world in which devices were cheap and numerous, and you could connect almost anything together. Weiser imagined that the future would be a world filled with “pads, tabs, and boards”, and this is what we have today. We have large displays, or “boards”, small tablets to carry around and do work on, and small pads to consume and distribute information with.

After Weiser, Brown and others set up all of these devices and the PARC research team used them, they found that the biggest issue was attention. “We don’t need smarter devices!”, Weiser said, “We need smarter people!” They began to write papers together.


Here’s one of my favorite papers: The Coming Age of Calm Technology. It explains the attention overload we would experience in the future and how we could create technology to deal with it.

My other favorite paper of there is called, “The World is Not a Desktop”. We can’t expect to interact with reality in the same way we interact with a desktop computer. With a desktop computer we have the complete luxury of being able to sit down and give the machine all of our attention. With mobile devices, we sit in the checkout line trying to download an app, load a coupon or send a message before it is our turn at the checkout counter. Our divided attention can also lead to dangerous situations such as trying to focus on driving directions while speeding down a congested highway in a new city.


This is an important note. We often forget the fact that we’re surrounded by invisible technology almost all of the time: Electricity! It’s there when we need it, not when we don’t. We only notice it when it breaks down, and we use it to get other tasks done, like watching a movie with a loved one or using the stove to cook a meal.


So how do we design this technology? I looked at what Weiser and Brown wrote and tried to boil it down to some key principles.


The first principle is this: Technology shouldn’t require all of our attention, just some of it, and only when necessary.

How does this work? Let’s look at an example.


Weiser used the tea kettle as an example of a technology that you set and forget. When it is ready, it calls to you. You don’t have to sit and watch it on the stove the entire time. Instead, you can go to another room and do something else. It will shout out to you when you need to give it attention!


The teapot example bings me to the next principle. Technology should be able to empower your peripheral attention.

We have a very large capacity for attention, but often times we design products that only work with our visual sense, and these require us to pay all of our attention to them. As you move away from the sense right in front of you, you can get other senses. Touch, peripheral vision, and sound! We can do so much when we make use of these additional senses. It’s a way to get the same amount of information across with the least amount of attention. Think of it as compressing information into another sense!


Here is an example of using the sense of touch to notify someone that they’re slouching. The LUMOBack Smart Posture Sensor is a device you wear around your waist. It monitors the angle of your back and buzzes you when you exhibit poor posture.

This is an example of a personal alert. I had an employee once that had an insulin pump installed. During one of our meetings he suddenly beeped. The beep caused him a bit of embarrassment, because a beep can be heard outside of the individual. It would have been much better for the device to use a more personal notification, like a buzz. Imagine beeping during a wedding or funeral and having no control over the notification type! Conversely, we can’t hear a tone in a crowded restaurant or a bar. A buzz gets the point across without disrupting others and does so in a variety of environments. This is an example of when an alert type should be changeable.


Here’s an example of a product that uses ambient awareness. There are so many futuristic videos out there where a disembodied computer voice wakes a single person out of bed in their perfect condo. Unlike real life, the computer can understand their San Franciscan accent perfectly, because that is the kind of accent the machine was trained on.

Instead, this light changes color based on the weather forecast. Will it be rainy? Easy! The light will turn light blue. Heavy rain? A deep blue. Sunny? Yellow. Grey? The light will be white.

The difference here is that you can simply feel what the weather will be like, without being told in a noisy voice. The iPad on the wall can display an additional level of information, should you want to check it, but it doesn’t blast it to you right away.


This is a light-based status system connected to a task-tracking service called Beeminder. Beeminder is a company that helps you track your goals. If you get off track, they take your money. Logarithmically.

If you’re tracking a lot of information, you don’t want to have to check it all of the time. It is better to know at a distance whether something needs your immediate attention or not. This is the idea behind this device. When everything is good, the light will glow green, making use of your ambient attention to let you know that everything is okay. If the light glows yellow, you know you have to check something, but there’s some time. If the light turns red you know instantly that something needs your attention.


You don’t need to overwhelm people with information. There are ways to present a lot of data in a less stressful way.


Compass for iPhone was an experiment in gathering all of the data already captured by your phone into an easily digestible image. With one click glace you can see a picture of your day. How much did you exercise? How much did you sit? Instead of individual numbers, you can see your day in colored bands and ‘moments’. This kind of information doesn’t try to understand how you’re feeling for you. You’re able to look at the information and make your own decisions about how your activities affect your life.


How many times have you been on hold with an automated phone tree? This kind of behavior turns humans into machines. A similar thing happens when we try to put humanness into technology. We end up with products that get us stuck in places, look and act creepy, or have other unwanted characteristics.

Technology and humanity have distinct and important roles.

The best technologies do work that humans are bad at. Going through thousands of datasets. Filtering through search results, ect.

Humans do things that technology is bad at. Customer service, warm embraces, and problem solving when something goes wrong with technology. Humans are natural curators. They understand context. No matter what, all loops should have a human at the end to check for accuracy. Otherwise we can get stuck in very unfortunate circumstances of automation. Need we bring the move WarGames back for this generation?

We need to make technology that enhances the best of humans, and the best of technology.


Take Google for example. It indexes a lot of pages made by hand. It does a great job managing search results, but it doesn’t make the final decision for us. Instead, it gives us a sheet of possible results. It is up to us as humans to make our choices. We are the ones with the context. This is why Google removed the “I’m feeling lucky” button. Humans choose from a small list. Machines cull down a large list. This is a good human-machine symbiosis.


SleepCycle is an application that uses your movement to track your sleep at night and wake you up at the right cycle of sleep in the morning. You set the alarm, plug your phone in, turn it upside down and put it under your pillow.


In the morning you get a nice sleep graph when you wake up, along with a “sleep score”. Because the alarm wakes you use at the right time, you get a fuller, fresher night of sleep. As the human, you set the alarm. The machine uses your data to wake you up.


Have you ever heard the annoying voice of a computer that is trying to sound like a human? This is an inefficient way for technology to communicate. Plus, it has to be translated into dozens of languages.

There are many more ways for products to communicate without relying on human voices.


The first example is an art installation by Natalie Jeremijenko at Xerox PARC in 1995. Called the “Live Wire” or “Dangling String”, the installation consisted of a plastic wire connected to a motor in the ceiling of one of the halls. When something interesting was happening on the internet network, the string would spin around and make a whirling noise. Researchers would hear this and go find out what was being worked on in the labs. It was a kind of digital water cooler, making use of ambient attention to attract curiosity, interest and cross-departmental human connections.


The Roomba Vacuum communicates only in melodies and chirps. The cleaner emits a happy chirp when it is done cleaning, and a sad chirp when it is stuck. Because it is small and cute and doesn’t continue to do work when it is stuck, it is a non-scary device. It waits for human help rather than trying to do everything itself. The chirp is easy to understand. It doesn’t need to be translated into many languages. It conveys the information with a light and a tone, and it is approachable enough that cats enjoy riding on it in YouTube videos!


Weiser imagined an alarm clock that would use a different tone based on what activities you had planned for the morning. For instance, if you had an early morning important meeting, the alarm clock would wake you up quickly. If you had nothing to do, the alarm would wake you up late, in a calm matter. There are ways for this to go wrong, though. A lot of people schedule lots of placeholders on their calendars, and having a computer try to differentiate between real and fake events is not ideal or possible. Only you can be the judge of that.


This principle is very important. Technology should consider social norms. It takes a while for technology to metabolize into humanity. When the first automatic elevators were installed in skyscrapers in New York, they had to be artificially slowed down because of pressure issues.

Technology has been ready for us for a long time, but it takes a while for us to get ready for it.


Let’s look at an example of different technologies. Technologies that are considered “human enhancements” are fear inducing when they are not available to everyone, are expensive and exclusive. The ill-fated Google Glass fell into this category.

Normal technologies are invisible. It is expected that most people have a smartphone. 10 years ago, an iPhone or Android would not be considered the norm, but norms change.

Restorative technologies such as crutches, eyeglasses and wheelchairs are almost universally accepted, as they get someone back to the “norm”.


Let’s look at how norms can be created and changed in just a few years.

When feature phones came out they had limited features. Text and voice calls were present, and few apps (with the exception of Centipede). When they first came out, use was limited to businesspeople, but over the next decade they were found everywhere. People got used to them, and then they relied on them.


The smartphone camera changed everything. 10 years ago the press was filled with articles about how the smartphone camera was the “end of privacy”, but because phones were now a norm, it was only a matter of years before smartphone cameras folded into normal social etiquette.


Google Glass was different. All of the features were released at the exact same time, leaving people confused as to what it really did. In the absence of information, people thought that Glass was recording them all of the time


What was different? The device was not open to developers, only to those who had the physical device. The secrecy and exclusivity launched it passed a “normal” device to an “enhancing” device in the eyes of consumers. The price point was $1600 – far too high for most people to purchase, let alone develop with. The device was also part of a closed network owned by Google, making it difficult for people to embrace application creation or use.


The results? Reduced play for developers, and confusion, speculation and fear for consumers. This could have been preventable. The device could have launched with the minimum amount of features to reduce the initial price point of the product. A red record light could have helped people notice when the device was recording. The system could have been open to development for more than just those who owned the object.


Let’s compare the launch of Glass to the iPhone launch. The iPhone launched with limited features and did not allow application development until the next revision of the device. This gave people time to understand the device and get accustomed to it.


When developers finally got access to it, they were able to create lots of inventive (and fun!) applications. These apps drove consumer adoption of the iPhone, and eventually people bought the phone because of the apps. The result was excitement and widespread adoption.


The right amount of technology is the minimum amount to solve the problem.


Take the Toilet Occupied Sign for instance. It is an almost universally understood pictogram showing whether the toilet is occupied or not. The lights tell the story. The application doesn’t need to be translated into many languages. even if you’re red/green colorblind you get the message.


How many applications work on your mobile phone when airplane mode is turned on? Not many. In the future, we’ll see a return to what we saw in the desktop era. Many devices close to us, serving our information temporarily to others. Share your medical details with your healthcare provider for 24 hours in return for medical services or a diagnosis.

Computing at this point will have reached a resource limitation as well. Bandwidth will be an issue, as more and more applications rely on the far, vs. the near. Video streaming applications like Netflix will see their bandwidth costs soar and consider various solutions. One solution will be to set up their media sharing like a a file-sharing system like Napster.

Why go to a remote server to get the next installment of Game of Thrones if your neighbor just watched it? Why not get most of the pieces from them and only the missing bits from the remote server? Games like Minecraft have done this method of distributed computing well. We’ll see more and more of it in the future as bandwidth becomes expensive and people demand more information.


Here’s a graph that’s similar to what Weiser wrote about in The Coming Age of Calm Technology. In computing’s middle ages we had mainframe computers. Giant boxes that filled auditoriums with many people to one computer. As computing matured we entered the desktop era. Suddenly a home could have a computer shared by a household.

With the advent of the Internet, we started to download information from far away. In an era that I call the “remote mainframe” we started to do much more of our processing remotely. And when we got to mobile phones, there are now many devices per person, but much of our information is far away from us.


The most important part of local networks is reliability. We are carrying around supercomputers in our pockets. We shouldn't be doing all of our processing in the cloud. Touching the network will become expensive, and we need clear reliability if we’re going to have these devices in our homes.

Why do we have to go all the way to the Internet and back to control our lights? Why don’t we have a local network set up in our homes? Security vulnerabilities aside, you shouldn’t have to be a system administrator to live in your own home. And you should be able to touch the network the least amount of times possible.


Here’s a list of all of the principles of Calm Technology. There’s also on the web at calmtech.com. (I usually pause here for people who want to take a picture of this slide).


If good design allows people to accomplish their goals in the least amount of moves…


Then Calm Technology allows people to accomplish the same goals with the least amount of mental cost.

A person's primary task should not be computing, but being human.


Because as Mark Weiser said in the late 90s, “the scarce resource in the 21st Century will not be technology. It will be attention.”

Our best technology will respect and make the best use of our attention so we can go on to other things with the least amount of time.

And that’s why we need a calm technology.


A person's primary task should not be computing, but being human.


Because as Mark Weiser said in the late 90s, “the scarce resource in the 21st Century will not be technology. It will be attention.”

Our best technology will respect and make the best use of our attention so we can go on to other things with the least amount of time.

And that’s why we need a calm technology.


If you’d like to learn more, I just published a book based on this talk. It’s published by O’Reilly Books and you can also get it on Amazon.com!


And most importantly, a lot of the original papers and notes on this talk are up at Calmtech.com, the companion site for this book and talk.


Thank you so much!

[edit] Q&A

People usually ask the following questions at the end of this speech.

What do you think about AI?

What do you think of self driving cars?

What do you think about the Singularity?

Retrieved from "http://wiki.caseorganic.com/index.php?title=Calm_Technology_Slides_and_Transcript&oldid=93"

This wiki is from http://caseorganic.com.