design frontiers Ebook free download pdf pdf


1 year ago
Full text

  Design Frontiers

How Voice, Sound, Wearables, Sustainability,

and Other Factors Will Shape Experiences



  Design Frontiers

How voice, sound, wearables, sustainability, and other factors will

shape experiences

What’s the one major hurdle in the design of wearable technologies? How

do we create technologies to maximize happiness and well-being? With

this free collection of chapters from O’Reilly Design Library, you’ll take a

trip to today’s design frontier, where innovators are creating voice user interfaces, and building sustainable websites and apps that consume far less energy.


Whether you want to know how storytelling plays a role in multi-device

design, or guide people through your digital product by using patterns from urban architecture, this Design sampler has the answers. You’ll discover that the O’Reilly Design Library delivers the knowledge and advice you need to expand your skillset and remain up-to-date with today’s best design practices.

  This collection includes:

  Designing for Sustainability


Chapter 1. Sustainability and the Internet

Designing for Wearables Chapter 6. An Architectural Lens

  [ 1 ] Sustainability and the Internet What You Will Learn in This Chapter In this chapter, we will cover: • A definition for sustainability and the role it plays in business.

  How organizations use sustainability principles to innovate, dif- • ferentiate themselves, reduce waste, and function more efficiently. How sustainability applies to the Internet. • How virtual life cycle assessments might help web teams devise • more sustainable solutions.

  A Greener Internet

“The Internet is the single biggest thing we’re going to build as a spe-


cies,” says Greenpeace’s Gary Cook in an article from The Atlantic.


“This is something that if we build it the right way, with the right

sources of energy, could really help power our transition to renewables.

If we build it the wrong way, it could actually exacerbate the problem.”

This is a book about designing the Internet the right way: efficient,

accessible, future-friendly, and powered by renewables.



Throughout this book, we’re going to dive into great detail about all the

checkpoints we can put into place to help us make better digital prod-

ucts and services that not only keep users happy and engaged but that

  1 Ingrid Burrington, “The Environmental Toll of a Netflix Binge”, The Atlantic, December 16, 2015. (

  ) clouds/420744



are also more efficient and save energy, as well. We will discuss how to

translate these checkpoints into an easy-to-understand framework that

will help you and your clients make more sustainable design and devel-

opment decisions.


Rarely are the things that we set out to build the things that actually get

built. In the maelstrom of changing opinions, validated learning, con-

tractual obligations, and shifting stakeholder requests, we are expected

to produce magic. To stay on time, on budget, or on good terms with

the person signing our checks, we cut corners. We give in to impracti-

cal requests. We let the client autoplay their 30 MB video on the home

page. We add an image carousel that over time becomes filled with a

dozen photos of generic business people shaking hands. And in turn

the average web page size—according to the HTTP Archive, which

2 tracks how the Web is built—rises to more than 2.3 MB.

  FIGURE 1-1.

  The average web page size hit 2.3 MB in early 2016

2 HTTP Archive, “Interesting Stats”. (

  ) php?a=All&l=Apr%201%202016




All that data bouncing back and forth through our networks requires

electricity. Energy is used to host our content, serve our content, and

interact with our content. Unfortunately, very little of that energy

comes from clean or renewable resources. Hence, the Internet, which

many people might think is a “green” medium due to the simple fact

that it often replaces paper, isn’t as clean as we might think it is. In fact


it produces more greenhouse gases (GHGs) than the airline industry,


which produced 770 million tons of emissions in 2015. With more

than three billion active Internet users in early 2015 and more than half


the world’s population predicted to be online by the end of 2016, it is

well on track to produce more than a billion tons of GHGs any day now,

if it isn’t doing so already by the time you read this.


We don’t mean to build bloated websites that clog networks, waste elec-

tricity, and frustrate users. Somehow—be it a desire to try out the latest

design bell or programming whistle or the need to keep demanding

stakeholders happy—it just turns out that way. If design and develop-

ment teams could check themselves every step of the way by simply

asking “Is this the most sustainable, efficient solution?”, many of these

bad decisions—which have detrimental consequences like lost revenue

and frustrated customers—could be headed off at the pass. And the

Internet would be a cleaner, greener, happier place for all. But first, let’s offer a bit of context.

  3 American Chemical Society, “Toward Reducing the Greenhouse Gas Emissions of the Internet and Telecommunications”, January 23, 2013. (


  ) gas-emissions-of-the-internet-and-telecommunications.html

  4 Air Transport Action Group (ATAG), “Facts and Figures”. (

  ) figures.html

5 Internet Live Stats, “Internet Users”. ( )



  | Sustainability Defined

The word “sustainability” and its variations are thrown around often,

maybe not as much as the term “green,” but still quite a lot. Marketers

love using it to describe their products’ virtues. Environmentalists

espouse a more hopeful future in its promise. In certain circles, its

overuse has perhaps watered down the word’s meaning and trivialized

the fact that few things are truly sustainable.


The most commonly quoted definition of sustainability comes from a

paper called Our Common Future, released in 1987 by an organization


known as the Brundtland Commission, named by the United Nations

after the former prime minister of Norway and tasked with the mission

of rallying countries to slow deterioration of our environment and nat-

ural resources:

  Sustainable development…meets the needs of the present without compromising the ability of future generations to meet their own needs.


At the time, the hope of this organization was to reconcile economic

development with environmental damage. Nearly 30 years later, this

dream of reconciliation between people and planet remains elusive. As

Naomi Klein points out in her book This Changes Everything: Capitalism

vs. The Climate (Simon & Schuster, 2014):

  Our economic system and our planetary system are now at war. Or, more accurately, our economy is at war with many forms of life on earth, including human life. What the climate needs to avoid collapse is a contraction in humanity’s use of resources; what our economic model demands to avoid collapse is unfettered expansion. Only one of these sets of rules can be changed, and it’s not the laws of nature.


Those of us who create the digital future are smack-dab in the cross-

hairs of this fight between economic expansion and environmental col-

lapse. The rise of the Internet has fueled our increasing demand for

products of all shapes and sizes, each of which requires substantial

6 Wikipedia, “Brundtland Commission”. ( )





resources to create, transport, use, and dispose of. As we shall see in

upcoming chapters, although physical products might require the

lion’s share of these resources, digital products and services have their

own role to play, as well.

  FIGURE 1-2.

  With so many doom-n-gloom environmental messages out there (some of questionable motivation) it is easy to see why people feel burnt out or overwhelmed and don’t think they can make a difference



Progress has been made since the definition of sustainability was

coined. Sustainability is now its own industry, driven by data and chock

full of consultants, analysts, corporate social responsibility (CSR) man-

agers, and scientists. Many sustainability professionals have adopted a

systems-thinking approach to sustainability rather than view it simply

as a compliance or regulation issue, which is common. These people

focus on how a company’s systems and its constituent parts interrelate

within the larger system of the environment over time.





In describing the definition of sustainability put forth by the Brundtland

Commission, the International Institute for Sustainable Development

7 notes that:

  All definitions of sustainable development require that we see the world as a system—a system that connects space; and a system that connects time. When you think of the world as a system over space, you grow to understand that air pollution from North America affects air quality in Asia, and that pesticides sprayed in Argentina could harm fish stocks off the coast of Australia. And when you think of the world as a system over time, you start to realize that the decisions our grand- parents made about how to farm the land continue to affect agricul- tural practice today; and the economic policies we endorse today will have an impact on urban poverty when our children are adults.


Using this systems-thinking approach (and others), the sustainability

industry has helped many businesses adopt more sustainable practices,

resulting in reduced GHG emissions, less waste, more efficiency, and

8 often higher profitability.


Author Nathan Shedroff also covers this extensively in his book Design

9 Is the Problem (Rosenfeld, 2009):

  The essence of this definition [of sustainability], which may not be obvious immediately, is that needs aren’t just human, they’re systemic. Even if you only care about humans, in order to care for humans, you need to take care of the system—(the environment) that you live in. And this environment doesn’t include just the closed system we call the planet Earth. It also includes the human systems we live in—our societies—and the forming, changing, and constantly evolving val- ues, ethics, religion, and culture that encompass these societies. We aren’t separable from each other, and we can’t ignore the effects of the whole—nor should we.

  7 International Institute for Sustainable Development (IISD), “Sustainable Development”.

  ( )

  8 Sunmin Kim, “Can Systems Thinking Actually Solve Sustainability Challenges?

  Part 1, The Diagnosis”, Erb Perspective Blog, Jun 4, 2012. (

  ) erbperspective/2012/06/04/systems-thinking-part-1 9 Nathan Shedroff, Design Is the Problem (Brooklyn, NY: Rosenfeld Media, 2011).





  7 To date, few people have applied this same type of thinking to dig-

ital products and services. With all of its figures, metrics, and sys-

tems-based thinking, the field of sustainability should be a natural

fit for those of us who create the nervous system of the 21st century

known as the Internet. But what does it really mean to be sustainable?

The reality is it’s next to impossible. Few things we humans create are

truly sustainable. In a 2004 Grist interview, Patagonia founder Yvon

Chouinard noted: 10 There’s no such thing as sustainability. There are just levels of it. It’s a

  process, not a real goal. All you can do is work toward it. There’s no such thing as any sustainable economy. The only thing I know that’s even close to sustainable economic activity would be organic farming on a very small scale or hunting and gathering on a very small scale. And manufacturing, you end up with way more waste than you end up with finished product. It’s totally unsustainable. It’s just the way it is.


In the green building movement, some have suggested that we move

beyond thinking about sustaining an already degraded planet and more

toward regenerative design rather than simply sustainable design. In

other words, defining success simply by being impact-neutral is not

enough. As a species, we need to renew and regenerate, to change

things for the better. According to the folks at Thrive Design: 11 This entails going even deeper than the leading edge of systems think-

  ing (the uncovering of the interconnected and complex nature of the world) and into the realm of systems being. It recognizes that we are the system, that there is and can never be any separation of humans from the web of life that we wholly depend upon. When we release toxins into the environment, we quickly discover them infiltrating our own bodies through the food we eat, the water we drink, and the air we breathe. In essence, what we do to nature, we do to ourselves. This perspective can help us come to see the planet and its life sup- porting systems as part of our ‘extended body’. A natural response to


10 Amanda Little, “An Interview with Patagonia Founder Yvon Chouinard”, Grist, October 23,

2004. (



11 Joshua Foss, “What is Regenerative Development?”, Thrive Design Studio. ( http://www. ) this understanding is of care and compassion as it becomes our own self-interest to actively manage the health and integrity of the living systems that we rely upon.


This switch from focusing on outcomes to focusing on process is one

that fits well within the context of this book. Startups, digital agen-

cies, and software companies are consistently moving toward mod-

els of continuous deployment, where the features and functions of

product releases are ongoing and the systems built are actively man-

aged much like those mentioned before. When we talk about design-

ing greener digital products and services, we do so with the idea that

the Internet will never be a truly sustainable place. It will always use

resources. There will always be work to do. If you have ever designed a

website or mobile app, you know that meeting the needs of the present

without compromising those of the future is a monumental challenge

indeed. After all, when was the last time you played a Flash game on

your phone? FIGURE 1-3.

  Systems thinking requires that we see the world as a system that connects space and time


  | Sustainability in Business

Designing greener digital products and services requires us to better

understand the role of sustainability in business as a whole, for it is

into these systems and processes where we will plug our own work.

Although it might not be the norm today, assessing the impact of a

company’s digital products and services will hopefully be just one com-

mon future component of greater sustainability initiatives in business.

The ways in which sustainability concepts are used in business are so

broad they can be hard to encapsulate in a single chapter overview.


Different companies take different approaches, use different tools, and

focus on different things based on the resources available to them and

what they want to achieve from their efforts. Each organization tends

to chart its own course using its resources, tools, and goals as drivers.

Some focus on energy efficiency, others on waste reduction. Some com-

panies are driven by marketing goals, whereas others use principles of

sustainability to drive innovation and disrupt industries. Some focus

on a triple bottom line that pays equal attention to people, planet, and

profit. The best-in-class do all of the above.

Cradle to cradle (C2C), for instance, is a biometric approach to the

aspiration of waste-free products and systems design. It models busi-

ness and manufacturing practices on natural processes and suggests

that industry must protect and enrich ecosystems and nature’s biolog-

ical metabolism while also maintaining a safe, productive, technical

metabolism for the high-quality use and circulation of organic and


technical nutrients. In other words, rather than producing waste, the

traditional waste products at the end of a product’s life cycle are reused

to give life to something new or are reintroduced to the life cycle of the

original product, creating what is commonly known as a “closed loop”

system, where outputs are reintroduced as inputs. Fairphone, a Dutch

company that sources e-waste from Ghana for use in its smartphone

devices, is a great example of C2C in practice. More on Fairphone in

later chapters.

12 Michael Braungart and William McDonough, Cradle to Cradle: Remaking the Way We Make Things (New York: North Point Press, 2002).




  FIGURE 1-4.

  Data drives sustainability principles at many businesses

Cradle to grave, on the other hand, refers to a company taking respon-

sibility for the disposal of goods it has produced and is most commonly

associated with life cycle assessments, which we cover in the following


section. C2C approaches are generally more favorable than cradle to

grave due to the closed loop system concept mentioned above.

Some companies use a helix of sustainability, another systems-based

approach which also maps models of raw material use onto those

of nature. The helix guides six categories of key business functions

(governance, operations, corporate culture, process, marketing, and

13 The Dictionary of Sustainable Management, “Cradle-to-Cradle”. ( http://www.






stakeholders) through five levels of commitment to sustainable prin-

ciples (no commitment, exploration, experimentation, leadership, and

14 restoration).

  FIGURE 1-5.

  The helix of sustainability

Factors such as number of employees, size of supply chain, financial

resources, company mission, and product and service types also play

important roles. Some have internal departments to manage these

affairs while others hire consultants or firms to assess impact and

make recommendations for them.

Because reducing an organization’s existing environmental impact and

devising processes for minimizing future environmental impact can

take so many forms, let’s explore some of the ways in which this can

play out:

  • Identifying (and, later, implementing) efficiencies

  Life cycle assessments (LCAs) •

  • Benchmarking

  Marketing, branding, and greenwashing •

  • Innovation and disruption


Although some of these processes do include assessments of electricity

used by internal data centers and employee workstations, they rarely

take into account energy used by websites, mobile apps, cloud-based

services, or other digital products and services. This offers unique chal-

lenges and opportunities.

14 The Dictionary of Sustainable Management, “Sustainability Helix”. ( http://www.







Many companies start down the road to greater sustainability by iden-

tifying efficiencies that let the organization reduce waste of any form

while also saving money. This is the low-hanging fruit to sustainabil-

ity professionals. These could be efficiencies in electricity use, produc-

tivity, process improvements, and so on. This is a practice particularly

common in larger organizations, as saving money is an easy sell to

shareholders and decision makers. Unfortunately, when organizations

just care about cost, they might only identify the lowest-hanging fruit

and not do more because the payback is longer for higher-hanging fruit

(instead of bundling a mixture of both).


Identifying efficiencies is often part of the larger goal and scope pro-

cess of a life cycle assessment, discussed in more detail a bit later. To

accurately identify efficiencies, companies must first accurately iden-

tify sources of waste. Companies need good (and ongoing) measure-

ment for this, which might be easier for energy but more expensive

and/or challenging for physical waste.


By assessing the entire life cycle of their products and services, from

cradle to cradle, as noted earlier, companies can identify sources of

waste, and in turn, identify corresponding efficiencies to improve

performance. This is a common workflow in sustainability practices,

yet to date very few companies that offer sustainability services or life

cycle assessments apply this process to digital products and services. A

framework for doing so will be presented later in this chapter and dis-

cussed throughout the book.



Life cycle assessments (LCAs) are commonly used to calculate the

environmental impact of a product or service during its entire lifetime.

Also sometimes referred to as cradle-to-grave, or more recently, cradle-

to- cradle assessments, as noted earlier, they entail multiple steps: 1.

  Goal and scope definition (What are we trying to accomplish?) 2. Inventory analysis (What are we trying to assess?) 3. Impact assessment (What is the impact of our inventory?) 4. Interpretation (What does the data tell us?)




After these steps have been defined, a rigorous process begins to help

the company or organization better understand its environmental

impact and then devise a plan to do something about it.

To apply these concepts to digital products and services you might

answer the preceding questions as follows: Goal and scope definition

  Define the environmental impact of our online properties and implement a plan for mitigating that impact.

  Inventory analysis Web, apps, cloud services, social media, and so on—what is included and what isn’t?

  Impact assessment How much CO e is generated by these properties? Is there other 2 waste, as well?

  Interpretation Create a plan to reduce impact—renewable energy, carbon offsets,

increases in product/service efficiency, e-recycling, and so on.

  FIGURE 1-6.

  Life cycle assessments take a product or service’s impacts into account from cradle to grave, or better yet, from cradle to cradle




  Setting goals and scope

It is important with any complex project to set specific goals, timeline,

budget, scope of work, and so on. However, enough flexibility should

be built into a project’s scope to allow for iteration, experimentation,

and the ability to pivot from a specific process should it not yield fruit-

ful results.


We will talk about Agile workflows in later chapters, because they

relate to creating more sustainable digital products and services, but it

is important to note that building iteration and collaboration into the

process of any project is often a more sustainable approach that bears

better results than those that require extensive scoping up front with

little room for flexibility after a project is in process.


If, for example, the focus of your efforts is solely on greenhouse gases,

15 your scope might look like this: Set boundaries

  Define an inventory’s physical, virtual, organizational, and opera- tional boundaries. Traditional LCAs are rarely full scope (cradle to cradle) and the assessment can only begin when materials arrive at a company’s gate. Power use for designer workstations, for exam- ple, would fall under the scope of a traditional LCA, but if apply- ing this to digital, are you including website or social media, too? If it is in the cloud, is it your problem? How much do you care about frontend, user-driven electricity during product use? For additional considerations, see the sideba.

  Define scope Decide which emissions source and/or activity categories should be included in the inventory. Does the scope include both physical and virtual properties? The details you define here will drive the level of effort required.

  15 US Environmental Protection Agency, “Climate and Energy Resources for State, Local, and Tribal Governments”. (

  ) inventory.html




  15 Choose quantification approach Depending on the data available and the purposes of the inven- tory, define how you will approach collecting data and where it will come from. In a traditional LCA, this might be identifying a source

for emissions factors (EFs) and a formula for measuring them.

  Set a baseline When choosing a baseline—the point that marks the beginning of your measurement efforts—to benchmark progress going for- ward, consider the following: {

Whether the data is available for the time you’ve chosen

{ Whether the chosen timeframe is representative { Whether the baseline is coordinated to the extent possible with

baseline years used in other inventories (if they exist)

  Engage stakeholders Bring stakeholders into the inventory development process early on to provide valuable input on establishing a baseline. If a specific department or position will be affected by the data derived from this process, it will help to have them on board from the beginning.

  Consider certification Consider a third-party certification such as ISO 17024, EPt(GHG), or those offered by CSA Group. This will ensure quality and that your inventory is complete, consistent, and transparent.

  The Power of Users

  Although you can’t control what end users put on their phone, how they treat their battery, or whether they keep their laptops running at peak efficiency, you can make sure you’re not part of the problem. As we’ll see throughout this book, slow-loading digital products and services that are unreliable have a real impact on the bottom line as well as on your reputa- tion with customers. But is their power use something you should measure? Isn’t that their problem? If you can measure how much use your application gets across devices and platforms, you can estimate how much electricity it needs. This is a stan- dard LCA concept that can be applied to digital. If you can measure it, you can reduce it by using many of the optimization techniques we’ll discuss in the chapters ahead. You can also offset it, which we’ll discuss momentarily.


  Inventory Analysis

This is where data collection happens. Life cycle inventories (LCIs) can

be complex affairs. Analysts track all of the inputs and outputs to a

business system, including (but not limited to) raw materials; energy

use; emissions to air, water, and soil (tracked by substance); and so on.


The complexity of a company’s supply chain is one of many factors that

will dictate the complexity of the inventory.

In Internet parlance, this would be what is commonly referred to as

analytics. A popular tool like Google Analytics, for example, offers you

all the things you can measure—with the notable exception of electric-

ity use. In the name of efficiency, it is up to you to take an inventory of

the metrics you should measure.

  FIGURE 1-7.

  Life cycle assessments can help you measure what really matters in your organization

For reference, common components of standard (i.e., physical prod-

ucts) LCIs include materials, manufacturing, packaging, distribution,

usage, and disposal. We’ll discuss how to apply these components to

digital products and services later in this chapter. For each of these

components, consider the following questions:



  Materials What are the materials used to make the product? Where are they sourced from? How much energy is used in that process? How much emissions are produced in both sourcing and creating the materials? How can they be made or sourced in a more sustainable manner?

  Manufacturing How much energy does the manufacturing process use? How much waste does it produce? Where can you reduce emissions in the process?

  Packaging What materials are used in the product’s packaging? Does the product use soy-based inks or compostable materials, for example, versus a less sustainable equivalent? How much energy is used in the packaging process? What about waste? Can that be reduced at all?

  Distribution How much energy is used in the process of distributing the prod- uct? Does it need to be shipped in a cargo container from China or can it be walked down the street to the store where it is sold? How much waste or emissions are produced when distributing the product?

  Usage Does the product require energy when being used? Is waste pro- duced? How can product usage be more efficient in energy con- sumption? How can it produce less waste?

  Disposal When the product is disposed, can its materials be reused? Are there ways in which it can produce no waste at the end of its life cycle?




  Impact assessment

This process, also part of a full LCA, is where the data collected is ana-

lyzed and the impact of a company’s product or service assessed. In

other words, a certain manufacturing process might require a specific

amount of oil or natural gas, something that is typically included as

part of the inventory that we just discussed. The impact assessment

will decipher the environmental effect of that process. For digital prod-

ucts and services, an assessment will use data pulled from analytics

tools to gauge a product or service’s environmental impact.


Finally, all that data needs to be interpreted and a plan of action put into

place to decrease environmental impact. This includes sourcing renew-

able power, increasing energy efficiency, reducing waste, and in gen-

eral making the company’s supply chain—including its digital prod-

ucts and services—more sustainable.

  The Greenhouse Gas Protocol

An important part of an LCA is calculating GHG emissions. As emis-

sions from electricity use can comprise the majority of a digital product

or service’s waste, this is critical to successfully analyzing their impact.

The Greenhouse Gas Protocol from the World Resources Institute

serves as the foundation for nearly every GHG standard and program

in the world. It is the most widely used international accounting tool for

government and business leaders to understand, quantify, and manage

greenhouse gas emissions across a product’s life cycle.


Product life cycle emissions are all the emissions associated with the

production and use of a specific product, from cradle to grave, includ-

ing emissions from raw materials, manufacture, transport, storage,

sale, use, and disposal. 16 The Greenhouse Gas Protocol can help organizations do the following:

  • • Determine and understand risks and opportunities associated

    with value chain emissions.
  • • Identify GHG reduction opportunities, set reduction targets, and

    track performance.

16 Greenhouse Gas Protocol, “FAQ”. ( )


  | Engage suppliers and other value chain partners in GHG manage- • ment and sustainability. Enhance stakeholder information and corporate reputation • through public reporting.

It is also important to note that the GHG Protocol only focuses on

greenhouse gases. It is not meant to show an indication of a prod-

uct’s entire environmental impact, which could include other kinds of

waste. Because electricity powers everything on the Internet, its big-

gest source of waste is GHG emissions, but as we will cover in later

chapters, the hardware used to create, serve, and interact with web con-

tent produces other forms of waste, too.

Although the Greenhouse Gas Protocol is currently the gold standard,

it is also worth noting that the Sustainability Accounting Standards

Board (SASB) has a mission to devise sustainability standards for

different industries based on accounting standards but for environ-

mental, social, and governance (ESG) issues as opposed to financial



It should be noted, too, that an important component of sustainability

in business is benchmarking improvement. How do you know if you’re

improving if you aren’t continuously measuring results and comparing

them against the last time measurements were taken as well as against

competitors? It is important that companies dedicate resources to these

endeavors in amounts appropriate for the organization. Otherwise, ini-

tial efforts are wasted and improvement is minimal.


Given the similarity in process between this and many digital endeav-

ors—such as Agile and iterative design strategies, website performance

optimization, or digital marketing campaign measurement—one

might think there would be more synergy between technology, design,

marketing and sustainability departments, but with few exceptions

this isn’t often the case.

  Which brings us to our next section.




  FIGURE 1-8.

  Benchmark it: an important concept in this book.



Some companies pursue sustainability initiatives to meet marketing

objectives or to appease shareholders, customers, suppliers, and so on.

If the company doesn’t take its environmental initiatives seriously, this

can offer opportunities for what is commonly referred to as greenwash-

ing (discussed i). Fossil fuel

companies might, for example, spend large sums of money on cam-

paigns about how they’re “going green” even though the product they

sell is the biggest contributor of GHG emissions. Examples include

“clean coal,” “clean natural gas,” and Volkswagen’s “Dieselgate” emis-

sions scandal (VW engineers intentionally programmed diesel engines

to activate certain emissions controls only when the vehicle was under-

going emissions testing, while in real-world driving, the vehicles’

nitrogen oxide output was up to 40 times higher than during tests and

were less efficient on performance and fuel efficiency than what the




company claimed). Coca-Cola, Air France, BP, and many others have

come under fire from environmentalists for touting their commitment

to the planet while also polluting or putting people and communities

18 at risk.


“In reality, every company with a green claim is greenwashing,” says

sustainability consultant JD Capuano. “Our current economic system

is based on unlimited growth, but our resources are limited. Even a

company that makes the most responsible products can still have sig-

nificant environmental impact.”

This is often a design problem as well. Because marketing campaigns

are so heavily reliant on good design to communicate their message,

designers are often hired to greenwash, and in some cases they might

even consider themselves “green” because of their work when in reality

that is not the case.


It is up to the individual designer to define where he draws the line in

the sand, and for many—especially those who freelance for a living,

where a single project could make or break your monthly finances—

that can be a moving target. In the early days of my career, I was hired

for a motion graphics project that marketed a certain type of cigarette

as an active lifestyle brand. Yup, you read that right: snowmobiles, ski-

ing, extreme winter sports, and cigarettes apparently all go hand-in-

hand. At the time I really needed the work so I took the gig, but in ret-

rospect it’s one of several freelance projects that, given the chance, I

would reconsider. Projects like these inspired me to focus more heavily

on doing mission-driven work that mattered as Mightybytes grew and



Given the lack of consistent, common standards for what CSR or sus-

tainability initiatives should entail and how truly good companies

should perform, it can be difficult to discern between a good company

and just good marketing. Some companies—sustainable brands, B

Corps, and those that subscribe to the philosophy of conscious capi-

talism, for instance—are able to align profit with purpose in ways that

  17 Robert Duffer and Tribune staff, “ Volkswagen Diesel Scandal: What You Need to Know” Chicago Tribune, September 22, 2015. (

  ) automotive/ct-volkswagen-diesel-scandal-faq-20150921-story.html

  18 Breena Kerr, “The Culprit Companies Greenwashing Climate Change”, The Hustle, December 6, 2015. ( )





help the company thrive financially while also solving social or environ-

mental problems. By undergoing the rigorous B Impact Assessment,

however, only certified B Corps prove that they adhere to higher stan-

dards of accountability and transparency, giving consumers confidence

that their products and services haven’t been greenwashed. We will dis-

cuss this in more detail in Chapter 3 when we cover triple-bottom-line

business models.

  FIGURE 1-9.

  Greenwashing: it might sound nice, but don’t look under the hood



Some leaders in this movement are disrupting the status quo with

approaches like biomimicry, a process that seeks sustainable solutions

to human challenges by emulating the patterns and strategies found


in nature. New York–based Ecovative Design, for instance, grows

packing and building elements from biodegradable mushroom-based

materials (see Figure 1-10) rather than manufacturing them from more

environmentally toxic sources. These products can replace polystyrene,

a more hazardous material.

19 Ecovative Design, “How It Works”. ( )



  FIGURE 1-10.

  Ecovative Design creates packaging and building materials from mushrooms. 20 Or take Nascent Objects, a more sustainable electronics platform that

could disrupt how consumer electronics are designed, manufactured,

and sold (see Figure 1-11). Their system of modular electronics compo-

nents operates on the premise that filling landfills with outdated gad-

gets or purchasing the latest edition of a device just for a single feature

is wasteful. The company’s business model is shaking up the way con-

sumer electronics have been brought to market for the last 60 to 80

years. Rather than releasing single-use products that lock the electron-

ics inside, they have created a system of 15 or so common electronics

modules—speaker, camera, microphone, and so on—and made them

interchangeable. The result is a system with untold potential and a com-

pany whose mission is to create a sustainable electronics revolution.


Choosing from various modules, customers can use the company’s

software to create electronics devices specific to their needs, such as

baby monitors, WiFi speakers, water meters, security cameras, and

more. The company uses 3D printing to create the device’s housing

and ships the product to the customer. If a product is outgrown—such

  20 Patrick Sisson, “Nascent Objects, a Sustainable Electronics Platform, Wants to Make Gadgets More Green”, Curbed, March 11, 2016. ( http://www.curbed.

  ) com/2016/3/11/11201000/nascent-objects-modular-electronics-sustainable





as a baby monitor—customers can reuse the internal modules to cre-

ate something new. So far, the company has designed a water meter, a

wireless speaker, and a security camera with many more products in

the pipeline FIGURE 1-11.

  Nascent Objects is disrupting the way consumer electronics are designed, manufactured, and sold

Company founder Baback Elmieh believes that an increase in the num-

ber of distributed supply chains coupled with better technology lowers

the barrier to entry for designers. Using a system like the one offered

by Nascent Objects means that startups can now release products that

previously required the work of a much larger company.

It is unclear what happens to device housings once the product is no

longer of use—are they sent back to Nascent Objects to be reused? Plus,

  3 D printing has its own environmental implications: 3D printers can

waste about 40% of nonrecyclable materials and because they require

keeping plastic materials melted in order to work, they use a lot of elec-


tricity. A design shop that keeps them running all day could ostensi-

bly run up a large carbon footprint for each printed piece. Plus, a study

from early 2016 showed that some 3D printers emit styrene particles, a

22 possible carcinogen.

21 Adele Peters, “Is 3-D Printing Better for the Environment?”, Co.Exist, January 29, 2014.


  ) environment


22 Steve Dent, “Study Shows Some 3D Printing Fumes Can Be Harmful”, Engadget, February

1, 2016. (

  ) be-harmful




Still, given the huge environmental impact of large-scale electronics

manufacturing, Nascent Objects’ model offers enormous potential to

minimize impact by rethinking the role of devices in our lives, offering

solutions with reuse built right into the business model.


Similarly, sensors associated with Internet of Things (IoT) applications

offer vast opportunities to monitor energy use and control energy-pro-

ducing hardware like solar panels. A Nest thermostat, for example, pro-

vides consumers with accurate real-time data on energy use, helpful

monthly reports, and the ability to turn down or shut off your HVAC

system when you’re not home, resulting in significant savings in both

energy and money. There are many business applications for IoT tech,

as well. We have only just scratched the surface here.



In 2014, then New York–based sustainability consultant JD Capuano

worked with a well-known online community platform to map out the

environmental impact of their business. The goal was to define and

quantify the impact, which included estimating the carbon footprint

related to running its website—how much carbon was being released

into the atmosphere by its servers’ energy use—and devise a strategy

for reducing that impact.

Because this company’s primary source of income came from its

online platform, it was important for it to include data centers and web-

site performance analysis as part of the process. “Our first step was to

interview the head of their data center to understand everything they

were doing,” JD said. The company colocated its servers in a few data

centers, meaning it shared server rack space with other companies. It

also had a number of geographically distributed servers on multiple

content delivery networks (CDNs). A CDN caches redundant copies of

content on servers in different geographical areas. When users request

that content, it is served from the location closest to them, resulting in

speedier delivery and less data transmission.

This was helpful in terms of efficiency, but meant that JD had to spend

a lot of time collecting data from disparate sources. “The hardest part

was getting some form of usable data from the colocated data center

host. It required some negotiation. JD then established a regular feed





of this information as it came in to his client’s IT team. “The CDNs, on

the other hand, were transparent,” he said. “One of them even calcu-

lated our client’s carbon footprint and emailed it over.”

JD combined CDN emissions and data center information that was

adjusted with partial feeds from power strips that recorded energy the


IT Team installed on some racks in the collocated data centers. The

power strip feeds let his team quality control the extrapolation method

on the provider’s data to estimate a unit emission figure that made

sense for the organization.


It is important to note that website and data center analysis were part

of a larger sustainability audit that JD did for the entire organization.

This included measuring and diagnosing its energy, emissions, water,

waste, and purchasing impacts. The emissions included its offices, data

centers/CDNs, business travel, employee commuting, and shipping.

Combining the overall data culled from the organization’s efficiency

efforts with that of the online components—website and data cen-

ters—enabled JD to make overall recommendations for emissions



Considering this was the first iteration of its platform’s carbon footprint,

JD decided to exclude embodied energy—the sum of all energy required

to produce goods or services—of the equipment, and the energy

required at end-of-life for disposal. Separate from GHG emissions, JD

devised a process for measuring waste that included e-waste—elec-

tronics hardware and other devices—collected in its offices and colo-

cation spaces. He also ensured that they were using a recycler certified

by e-Stewards, an organization that provides globally responsible elec-

tronics recycling. Chapter 3 includes more information about e-waste.


  | Sustainable in Digital?

  To give you an example of how otherwise innovative and sustainable com- panies can easily overlook the impact of their digital properties, neither Nascent Objects nor Ecovative Design appear to power their websites with renewable energy. Both sites have room for improvement in performance optimization, as well.

  FIGURE 1-12.


Both Nascent Objects and Ecovative Designs have made public commitments

to sustainability (hopefully, their websites will soon follow)

  That said, these startups are just getting off the ground, so the small amount of emissions their websites produce is not likely a high priority item for either company. They have many other fish to fry in this early stage of their businesses. However, one hopes that as these companies grow and scale, they will look at the impact of their digital products and services alongside the physical ones. The most sustainable startups will be those that bake sustainable thinking into their business model from the beginning—and that includes considering digital products and services.



  | Sustainability and the Internet

Now that we have covered a brief overview of how sustainability prin-

ciples are used in business, how can we apply these practices to the

Internet? As mentioned previously in this chapter, when it comes to dig-

ital considerations, many life cycle assessments stop at employee work-

stations and data centers. If we look at where energy is used throughout

a web application’s life cycle, however, we can see that energy is needed

in three key areas: The creation, testing, launch, and maintenance of these applications •

  The hosting and serving of these applications • The download and interaction of users with these applications •

Clearly, making data centers and workstations more energy efficient

plays a huge role in the aforementioned areas, but it’s not the complete

picture. If we want to calculate the environmental impact of a digital

product or service during its entire lifetime, we must apply the same

LCA process to its entire life cycle, as well.

Let’s look at Internet supply and demand from a very high level. Then,

we will dive deeper into how you can create a virtual life cycle assess-

ment for your own digital product or service.



The Jevons paradox is an economic term coined by 19th century econ-

omist William Stanley Jevons to explain how technological progress

increases efficiency and also increases demand, and subsequently con-

sumption due to what he called a rebound effect—when something is

cheap and convenient, more people want it. Jevons was initially talking

about coal use. He observed that when technological advancements

increased efficiency and reduced prices, it also drove demand for coal

and increased consumption across many industries.


Though Jevons was originally talking about energy, his paradox can be

applied to almost any resource and is especially relevant to Internet use.

In most examples—such as automobiles or lighting—a convenience

factor such as price, availability, location, and so on led to widespread

adoption, and in some cases innovation. In the context of this book,

when more efficiency leads to more consumption, we get 3+ billion



  23 Internet users taking advantage of cheap broadband, inexpensive net-

books, and of course the now ubiquitous smartphone. This number

is growing exponentially with some people predicting that the entire


world will be online by 2020. Widespread adoption always has impact.


The choices we make to buy a cheaper soda, drive a roomier car, or get

faster access to the Internet have consequences when they’re ampli-

fied to include even half the world’s people. When you consider the

amount of virtualization—migrating offline processes online—that

global businesses are currently undertaking, the impact is significant.

Though these migrations will themselves ostensibly reduce CO 2 e emis-

sions, they also have an effect of their own, which is often overlooked.


Similarly, the IoT will see 50 billion connected devices by 2020, or about


devices per person on the planet, according to a popular report by

25 26 DHL and Cisco. Gartner put the number closer to 20 billion devices.


Whether it’s a car self-diagnosing faulty brakes or a chip in your pet’s

collar that texts you Fluffy’s location when she has run away, anything

that can be assigned an IP address and transmit or receive data is a

“thing” in IoT parlance.


The consumer-side promise of the IoT is that its smart, self-aware

nature will help people make more responsible choices in real time.

On the business side, industrial automation—in everything from traf-

fic lights to buildings—while less visible, will save significant amounts

of electricity. Remote monitoring will also save on transportation costs

and their accompanying emissions.


“If we can put computing into goods so they can self-describe, and pro-

vide access on how to safely dispose of them—where ‘dispose of’ means

return for recycling, or remanufacture, or reuse,” says Chris Adams

  23 Internet Live Stats, “Internet Users”. ( )

  24 Chris Greenhough, “Eric Schmidt Predicts Entire World Will Be Online By 2020”, Inquisitr, April 15, 2013. (

  ) world-will-be-online-by-2020

  25 DHL and Cisco Trend Report 2015, “Internet of Things in Logistics”. (


  ) things.pdf

  26 The Register, “Gartner: 20 billion things on the Internet by 2020”. ( http://www.theregister.





  • • Manufacturing all these often disposable devices expends large

  • • It is very difficult to gauge the overall footprint of a device because

    often its components are created in many different places.
  • • Whereas some, some devices, such as the Nest thermostat (dis-

  • • IoT devices will often replace older devices, which need to be dis-

  • • Each device transmits and receives information to a server some-



from Product Science, a London-based digital agency that works pri-

marily with organizations working on social or environmental prob-

lems as part of their business, “then we have a chance to close a lot of

extremely wasteful flows of resources.”

This is all great for the environment. There are, however, some signif-

icant concerns as well:

  amounts of energy and uses potentially hazardous raw materials or conflict minerals.

  cussed in more detail in a moment), use features and UX to save energy, others, such as fitness monitors or home automation sys- tems, can have much heavier energy footprints.

  posed of. Also, the sometimes disposable/replaceable nature of these devices—their “planned obsolescence”—means that they might end up in a landfill. Some companies even build in self-de- struct functionality for security purposes so that a device can’t be used after a certain date. 27

  where in a data center, which requires power 24/7/365. Few of these data centers are powered by renewable energy.

“This ubiquity is a double-edged sword,” Chris says. “Computing is

now cheap to the point that chips are disposable, so we need to provide

other reasons to make them last beyond them just being expensive to

deploy, or possible to biodegrade in some safe fashion.”

Though all of these Things on the IoT will ostensibly give us better ana-

lytics for identifying efficiencies and making more sustainable choices,

they also require power sources to continue transmitting and receiv-

ing data in real time. As mentioned previously, these devices offer

  27 Klint Finley, “The Internet of Things Could Drown Our Environment in Gadgets”, Wired, June 5, 2014. (



unprecedented possibilities for monitoring and controlling energy

use, but current IoT standards are scattered at best and very few of the

device manufacturers are collaborating to fix the issue. So, although

it’s great that we can monitor energy use, the lack of data standards can

make for unnecessary transmission of information. This makes for a

rather significant sustainability challenge when it comes to electricity

use. Networks, too, will need to be very robust to accommodate this

rapid growth in data transmission.


If you own seven of these devices, for example, and none of them talk

to on another, there is likely great redundancy in data transmission as

each device transmits and receives in its own proprietary format. While

Apple—with its HomeKit framework—and a few others are making

in-roads to developing common hardware and software standards for

these devices, it is still very much the Wild West, with many companies

working in silos. To develop a more sustainable IoT, companies must

work together to devise common standards that will streamline data

transmission and alleviate redundancies.

There is, however, a great untapped opportunity in IoT to help consum-

ers make more sustainable choices through UX practices when creat-

ing interfaces for these devices, as in the case of the Nest thermostat

(Figure 1-13).

  FIGURE 1-13.

  The Nest Thermostat’s leaf icon helps users make more sustainable choices





Human behavior is modified more by immediate feedback rather than

delayed feedback. Take your electricity bill, for example. You are far less

likely to modify electricity use when you receive the bill only every 30

days after you’ve already used the electricity you pay for. The Nest ther-

mostat, on the other hand, not only detects your presence in a room and

makes temperature adjustments accordingly to minimize electricity

use when you’re not around, it also shows a handy leaf icon when you

adjust the thermostat to a more energy efficient setting. This instant

feedback helps customers make more informed and energy efficient



If immediate feedback isn’t enough, Nest also sends a monthly report

(Figure 1-14) via email that not only includes information about your

own usage, but also compares it to other Nest users in your area, incen-

tivizing cost and energy savings.


Although Nest’s use of iconography, real-time feedback, and incentives

offer great lessons in creating more sustainable user experiences, there

are others, as well. We will cover more on sustainable UX practices in

Chapter 5.



  FIGURE 1-14.

  Nest also sends a helpful monthly email that shows your electricity use as compared to others in your area





As global Internet usage has grown, so too have our apps and web pages.


As mentioned previously, the average web page size as of this writing


is 2.3 MB, more than 24 times the size it was in 2003. When served

up on sluggish Internet connections to mobile devices with slower pro-

cessors, these pages waste time and energy and frustrate the bejeezus

out of users. Our love affair with video backgrounds, multi-image car-

ousels, complex social-sharing features, high-res images, rotating ban-

ner ads, and other frontend bells and whistles has caused an epidemic

of slow, overloaded pages. Meanwhile, many studies have shown that

users are unlikely to wait more than a couple seconds for a page to load.

  There seems to be a bit of cognitive dissonance happening here.


Chapter 6 is devoted entirely to helping you optimize pages for better

performance and faster delivery. VIDEO STREAMING

Let’s not forget one of the most bandwidth-intensive uses of the Internet

to date: streaming media. Consider these statistics: With more than 1 billion users as of this writing, YouTube alone • streams hundreds of millions of hours of content each day. 300 hours of video are uploaded every minute. Still, according to broad- band Internet tracking firm Sandvine, YouTube made up about 29


% of overall Internet downstream traffic at the end of 2015.

Netflix use, on the other hand, made up about 37% of all Internet • 30 downstream traffic.

  Amazon video comprises slightly more than 3%. •

  28 Website Optimization, “Average Web Page Breaks 1600K”. ( http://www.


  29 Emil Protalinski, “Streaming Services Now Account for Over 70% of Peak Traffic in North America, Netflix Dominates with 37%”, VentureBeat, December 7, 2015. ( http://

  ) north-america-netflix-dominates-with-37

  30 Sandvine, “Global Internet Phenomena Report”. (

  ) global-internet-phenomena




In all, video streaming services like YouTube, Netflix, Hulu, Vudu, and


others already make up more than 70% of consumer Internet traffic.


In 2013, Cisco said streaming video was expected to grow to 76% by

32 2018 but by current numbers, that estimate might be quite low. One

can see why it is important that the electricity used to power these ser-

vices is generated by renewable sources. Currently, there are big draw-

backs to that, which we will cover in more detail in Chapter 3.


Progress is being made on this front. In 2013, according to Google

Green, Google’s website on its environmental impact, 35% of the com-

pany’s energy came directly from renewable sources, leaving 65% pro-

vided by other nonrenewable resources. Greenpeace’s 2015 Clicking

Clean report, which bills itself as a “a guide to building the green

Internet” put that number at 46%, with 15% coming from natural gas,


% from coal, and 13% from nuclear sources. Google claims to offset

its nonrenewable power sources, but as we will see in Chapter 3, renew-

able energy credits have their own set of drawbacks. Still, with more

than $2.5 billion worth of investments in renewable energy projects

and a commitment to 100% renewable power by 2025, it is one of the

33 largest corporate investors in renewable energy in the world.


Another example: Netflix and several others such as Vimeo, Vine, and

SoundCloud use Amazon Web Services (AWS) to host their content, as

do hundreds of other popular apps we all use, like Dropbox, Pinterest,

The Huffington Post, Yelp, Reddit, and so on. In fact, a 2012 study by

Deepfield noted that one-third of all Internet users will access an AWS-

34 hosted site or app on average of at least once per day.


This can be problematic for sustainability purposes because AWS has

consistently come under fire by Greenpeace and other organizations for

its lack of transparency about energy use and sources. Though it has

made a long-term public goal to be 100% renewable energy powered,

the lack of transparency from AWS makes it difficult for the public to

  31 Emil Protalinski, “Streaming Services”. (



32 Cisco Visual Networking Index: Forecast and Methodology, 2013– 2018 (2014).

  33 Google, “Renewable Energy–Google Green”. ( )

  34 Robert McMillan, “Amazon’s Secretive Cloud Carries 1 Percent of the Internet”, Wired, April 18, 2012. ( )





ascertain how it will actually accomplish this goal, leading Greenpeace

to grade the company with C, D, and F ratings across its 2015 Clicking

35 Clean Scorecard.


With more than 60% of the Fortune 100 having carbon and renewable

energy goals in place, the company’s lack of transparency will likely


become of larger concern to their customers. But they are not the only

player for which this has become an issue. As you can see from the

chart in Figure 1-15, other companies such as eBay and Digital Realty,

which hosts LinkedIn, have a ways to go, too.

  FIGURE 1-15.


This infographic from Greenpeace’s 2015 Clicking Clean report shows which of

the big Internet companies have committed to renewables and which still have

a ways to go

When you couple the meteoric growth of streaming video with the lack

of transparency and available resources for hosting powered by renew-

able energy, the problem becomes crystal clear. Again, Jevons paradox

in action.



36 Ceres, “Power Forward 2.0: How American Companies Are Setting Clean Energy Targets

and Capturing Greater Business Value”. ( forward-2.0-how-american-companies-are-setting-clean-energy-targets-and-capturing-

  ) greater-business-value/view





With the release of devices like the Oculus Rift, Sony’s PlayStation VR,

and others entering consumer homes for the first time in 2016, vir-

tual reality (VR) has seen a resurgence in popularity that Dr. Jonathan

37 Waldern, a VR pioneer, calls “as big an opportunity as the Internet.”

Projected revenue by 2020 is expected to be at $5 billion for sales of

virtual reality game content, but that’s just scratching the surface.

Projected 2020 sales for virtual and augmented reality content are


expected to be at $150 billion. Although traditional game markets

make logical sense, online providers like Netflix, Hulu, and Amazon

are also reportedly exploring ways to add VR content to their subscrip-

tion services. This could put significant pressure on already stressed


  FIGURE 1-16.

  Virtual reality: an opportunity as big as the Internet

The ultimate goal of virtual reality is to create as much detail in the

rendering of a virtual setting that it is indiscernible from the real

thing. This includes visual, aural, touch, temperature, and eventually

  37 Charles Arthur, “The Return of Virtual Reality: ‘This Is as Big an Opportunity as the Internet’”, The Guardian, May 28, 2015. (

  ) may/28/jonathan-waldern-return-virtual-reality-as-big-an-opportunity-as-internet

  38 Janko Roettgers, “Hardware Giants Bet Big on Virtual Reality and a Market That Doesn’t Yet Exist”, Variety, March 22, 2016. (

  ) oculus-rift-consumers-1201735290





olfactory details. According to an early 2016 article in Forbes, “humans

can process an equivalent of nearly 5.2 gigabits per second of sound and

light—200 times what the US Federal Communications Commission

predicts to be the future requirement for broadband networks (25

39 Mbps).” Because it is estimated that the human eye can perceive up to 150 frames per second, “assuming no head or body rotation, the eye can

receive 720 million pixels for each of 2 eyes, at 36 bits per pixel for full

color and at 60 frames per second: that’s 3.1 trillion (tera) bits!”

We’re not there yet, of course. And although compression algorithms

can reduce some of this, for VR to realize its true potential, the fact

remains that exorbitant amounts of bandwidth will be required to

transfer huge data payloads to end users. According to Dr. Markiewicz:

  The game industry, which will have a major role in virtual reality, is not known for efficiency or environmental awareness. The cultural zeit- geist glorifies Hummer computers over Prius rigs, and game design has emphasized ever more elaborate 3D and animation without regard for power consumption. If graphic design has ignored the energy con- sequences of bad web design, the game industry celebrates its defi- ance of sustainable norms.


With all this in mind, virtual reality content creators will need to

embrace new breakthroughs in performance optimization to maintain

a good balance between limited bandwidth, great UX, and optimized

performance without latency.



Finally, no discussion of Internet sustainability would be complete

without mentioning the power used by data centers. From those owned

by large companies like Google and Facebook to the colocated servers

used by web design firms and small startups across the globe, every

server on the Internet pulls energy from our aging electricity grid


/7/365. Nearly all of them have redundant methods in place to serve

data from alternative sources if a server goes down and to guard against

  39 Bo Begole, “Why The Internet Pipes Will Burst When Virtual Reality Takes Off”, Forbes, February 9, 2016. (

  ) internet-pipes-will-burst-if-virtual-reality-takes-off/#774f287064e8




power failures, so these redundancies also require power in addition to

that used by their live servers. They also typically have banks of genera-

tors—which run on diesel fuel—as backup power sources.

This is important because according to a 2013 post in the New York

Times, digital warehouses use about 30 billion watts of electricity, about


the same output as 30 nuclear power plants. A single data center can

use as much electricity as a small town (Figure 1-17). According to that

same article, on average, data centers only use 6 to 12% of the electric-

ity powering their servers for computations. The rest is used to cool

servers down, keep them idling, and to guard against surges that could

crash their systems. In other words, it is possible that data centers can

waste around 90% of the electricity they pull off the grid.

  FIGURE 1-17.


A single data center can use as much electricity as a small town and many also

use diesel-fueled backup generators to maintain uptime during outages

40 James Glanz, “Power, Pollution and the Internet”, New York Times, September 22, 2012.


  ) energy-belying-industry-image.html





In countries like the United States, only about 13% of the nation’s

energy comes from renewable sources, according to the US Energy

41 Information Administration. This includes hydroelectric power,

which some environmentalists don’t consider a true renewable source

because of the potential damage it can have on natural habitats, but

doesn’t include nuclear power (another 20%), which has environmen-

tal issues of its own.


When one considers this alongside the 30 billion watts of energy used

by data centers, 90% of which is wasted, one can quickly see a pic-

ture emerging of an industry poised to be one of the planet’s biggest


  Moving to renewables

Although some of the larger Internet companies like Apple, Google,

and Facebook have made huge strides in powering their centers with

renewable energy, as of this writing, others, like Microsoft and AWS,

still have a ways to go yet. Many cloud providers, like Heroku, as an

example, build on top of AWS’s existing infrastructure. In late 2015,

Rackspace, which struggled as a competitor to AWS in cloud-based ser-

vices, followed Heroku and others’ path by offering managed cloud ser-

42 vices on AWS.


Similarly, in the world of colocation and web hosting providers, some

companies have made strides toward powering their centers with

renewables, but many are completely opaque about where their energy

comes from, including providers that tout their “green” credentials.

This is tremendously important to conscientious designers and devel-

opers who want to build environmentally friendly digital products and

services. Where we host our apps matters.

We will cover data centers, hosting providers, and renewable energy in

more detail in Chapter 3.

  41 US Energy Information Administration, “How Much US Electricity Is Generated from Renewable Energy?”. (

  ) cfm

  42 Yevgeniy Sverdlik, “Rackspace to Provide Managed AWS Services Before Year’s End”, Data Center Knowledge, August 12, 2015. (

  ) archives/2015/08/12/rackspace-provide-managed-aws-services-years-end


  | Virtual LCAs

Now that we have a better understanding how and where energy is

used to power the Internet and a working knowledge of sustainabil-

ity principles in business, let’s apply the LCA to digital products and

services. Remember that list of processes we discussed earlier in the

section on inventory analysis? Here is where that specifically applies to

digital products and services.


In an article published on Creative Bloq (formerly .Net magazine),

Dr. Pete Markiewicz proposed a framework for equating virtual LCAs


to their physical counterparts. Components include those listed in

Table 1-1.

  TABLE 1-1.

  This chart from Pete Markiewicz compares standard life cycle assessment components to their digital counterparts LCA


  Materials Software and visual assets Manufacturing Design and development Packaging Uploaded to the Internet Distribution Downloaded through the network Usage Interaction, user experience, completing tasks Disposal Data erased from client

  According to Dr. Markiewicz:

  Compared to the physical world, sustainable web design will de-em- phasise the manufacturing (read production) stage, making it qualita- tively different from print design. Why? Unlike physical products, web pages don’t produce waste paper or ink after the page disappears, only heat from the electronics. for this reason, the cost of web page manufacture will be small compared to the cost of ongoing use. On the other hand, the longer a web page is viewed, the more bits it burns, so efficient use based on good user experience will be very important.

  Let’s take a look at each of these virtual components individually.

  43 Pete Markiewicz, “Save the Planet Through Sustainable Web Design”, Creative Bloq, August 17, 2012. (

  ) sustainable-web-design-8126147






Rather than asking how much energy a manufacturing process uses,

let’s look at the software and visual assets needed to build our digital

products and services. If you use cloud-based services, are they hosted

on servers powered by renewable energy? If not, do you tend to leave

applications open and running in the background on your computer?

Many of these applications regularly use network connections to check

for updates or provide usage information to the company that created

them. When left open, they also use RAM and processing power, which

wastes energy.

It can be next to impossible to decipher how energy efficient the cre-

ation process of a piece of software is. Did they use Lean Startup princi-

ples, creating only what was necessary to validate their product? Or did

they load the software package up with features that are rarely used?

(I’m looking at you Adobe Photoshop and Microsoft Word.) Are the

software company’s offices powered by renewables? Is the equipment

energy efficient? The same questions apply to visual assets used during

the creation process. It can be challenging to answer these questions,

but if we are equating the manufacturing process of a physical prod-

uct to that of a virtual one, they are worth considering in the name of

reducing wasted and emissions.



It is in the design and development of a product where you can make

many efficiency gains and energy savings. Strategies like progressive

enhancement—an approach that uses feature layering to allow anyone

access to basic content and functionality while also providing enhanced

features to those with more advanced browsers and/or greater band-

width—and mobile-first makes digital products and services accessi-

ble to more people. Web standards like HTML5 and CSS 3 as well as

emerging open source frameworks for virtual reality like WebVR and

OSVR will allow for more efficient delivery of content to users across

devices. Similarly, accessibility standards for users with disabilities

who might need to access your content with screen readers or other

assistive technologies will mean that your content can be viewed across

a wider array of platforms and devices.




When running an inventory analysis on a digital product’s design and

development, some questions to ask include the following: Does the product use outdated or nonstandard technology that • make hardware work harder or requires plug-ins or other addi- tional resources to run? Is it optimized for mobile devices? • Are images, scripts, and other assets compressed or otherwise • “minified” for fastest possible delivery?



Our digital products and services might not require packaging or cargo

ships to transmit them from place to place, but there is still energy

needed to get these applications to their virtual destinies. The US

Energy Information Administration estimates that about 6% of elec-

tricity used in the United States is lost in transmission and distribu-


tion. Typical modernized countries have similar rates between 6 to



%. In developing countries, it’s another story. In 2000, India lost

around 30% of its electricity in transmission, but has made improve-

ments and is currently at about 18%. This is due to the simple fact

that by running electrical energy through wires some of it eventually

becomes heat.


Designers and developers can create applications that are more effi-

cient and served from computers powered by renewable energy to make

this less of a problem. Use of CDNs, by which pages are served from

the server closest to the person requesting them can help minimize

this, as well.

Thus, when running an inventory analysis on a digital product or ser-

vice, it is important to ask questions about how much energy is lost in

transmission as well as that which is used by infrastructure, bandwidth,

and frontend use. Are there ways to reduce those amounts?

44 US Energy Information Administration, “Frequently Asked Questions”. ( http://www.eia.

  ) gov/tools/faqs/faq.cfm?id=105&t=3

  45 Robert Wilson, “How Much Electricity Is Lost in Transmission?”, Carbon Counter, February 15, 2015. (

  ) electricity-is-lost-in-transmission





All digital products and services require energy to run. Interaction rep-

resents the most profound way they differ from their physical coun-

terparts, as noted by Dr. Markiewicz’s quote earlier. Question is, how

much energy do they use? How can one answer this for a virtual LCA

inventory? Here are some questions to ask: How much traffic does my site or app get? •

  How much of that traffic comprises computers versus mobile • devices? How much energy do each of those devices use per second? Per • minute? How much data is uploaded/downloaded per device per minute? •

  • • What is the average amount of time spent (per device) using my

  digital product or service?

You can find answers to many of these questions via Google Analytics

or other metrics measurement applications.



In the inventory analysis for a physical product, one considers how

much waste is created and energy used when a product is discarded.

Often the focus is on recycling disposed materials. With a virtual prod-

uct, recycling isn’t an issue, though churn can leave residual data for

closed accounts in databases, increasing an application’s size and hence

footprint. On the frontend, power is used to delete applications or doc-

uments, though that is minimal in comparison to usage and interac-

tion. Also, many online systems, like Git, for example, back up data

rather than deleting it. So, if you store thousands of backups, even if

only changes are stored, does that increase electricity use and hence

environmental impact while also decreasing efficiency? Although disk

storage might be energy efficient, backup management might be inef-

ficient. Also, as our applications grow in size and complexity, this can

become more of an issue.

  Some questions to ask include the following:

  • • What processor resources are required to delete my application?

  Does it leave residual information on the user’s device? • How many unused accounts are in the database? •



  Code obsolescence

Many digital products and services we rely on every day are built on a

foundation of obsolete code and outdated proprietary systems, hacked

together with the digital equivalent of dental floss and chewing gum.

Airline reservations and stock trading are just two examples of large

systems our society depends upon to function but which sit upon

unstable and outdated code. In July 2015, both the NYSE and United

Airlines’ systems went down on the same day (as did the Wall Street

Journal, for that matter). Outdated code not only leads to potential secu-

rity and performance issues, but it also wastes energy and puts your

users’ data at risk.

  Investing in updates

The importance of keeping hardware/software up to date cannot be

stressed enough. Band-aid fixes without the equivalent accompanying

architecture planning, budgeting, and so on add more vulnerability to

your digital product or service. The duct-tape method of coding new

features is unfortunately more common than one might think and is

decidedly not a very sustainable approach. In a Medium post about this

46 topic, professor and software developer Zeynep Tufekci, states:

  A lot of new code is written very very fast, because that’s what the intersection of the current wave of software development (and the angel investor/venture capital model of funding) in Silicon Valley com- pels people to do…Software engineers do what they can, as fast as they can. Essentially, there is a lot of equivalent of “duct-tape” in the code, holding things together. If done right, that code will eventually be fixed, commented (explanations written up so the next programmer knows what the heck is up) and ported to systems built for the right scale—before there is a crisis. How often does that get done? I wager that many wait to see if the system comes crashing down, necessitat- ing the fix. By then, you are probably too big to go down for too long, so there’s the temptation for more duct tape.

  46 Zeynep Tufekci, “Why the Great Glitch of July 8th Should Scare You”, The Medium, July 8, 2015. (

  ) b791002fff03





Also, if you build an application that relies on someone else’s data, keep

in mind that every time they update their application programming

interface (API), you might likely need to update your application, as

well. Sometimes, this will result in features users love no longer being

available. A digital product or service will see its life cycle cut short if

the software isn’t fast, reliable, and up to date.


When running an inventory analysis on the end stages of your digital

product or service, ask the following questions: How will you keep your application updated, your code lean, and • your users’ data secure over time? How will you add new features that users want while still staying • efficient? When the time does come to finally put your app to rest, how will • you do so with as little waste as possible? What tools can you add to your workflow (e.g., testing modules, • real-time performance analytics, hack-ability tests) to improve code quality?

  • • What project management strategies can you implement to pre-

  vent waste of coding time, improve bug detection, and stress-test

applications during Agile sprints? (More on Agile in Chapter 3.)


  | Hardware Disposal FIGURE 1-18.

  The US produces enough annual e-waste to fill 1.15 million 18-wheel trucks.

Put end-to-end, those trucks would stretch nearly halfway around the world.

  Outdated hardware can have a significant impact on performance, stability, and security. Sometimes, it needs to be replaced and should be done so in a manner in which the hardware components can be recycled wherever pos- sible. As of May 2015, up to 90% of the world’s electronic waste is illegally 1 traded or dumped each year, according to the United Nations. This hard- ware contains toxic chemicals like mercury, lead, cadmium, and arsenic, which can cause cancer, reproductive disorders, and other health problems. The United States is the world’s largest producer of e-waste, more than one million tons ahead of China. How much waste are we talking about? 41.8 million metric tons in 2014 alone, enough to fill 1.15 million 18-wheel trucks. With the average truck trailer being about 53 feet long, lined end- to-end you could reach nearly halfway around the world with all the e-waste the United States produces in a single year.

1 America Upcycles, “About e-Waste”. ( )




  | Conclusion

The statistics shared in this chapter are meant to give context. The

Internet provides great social and environmental benefits by allowing

easy access to data and information that was once locked up in pro-

cesses with far greater environmental impact. (After all, when was the

last time you purchased a CD at a record store?) This easy access has

empowered nations, enabled communications where previously there

were none, and given people in some societies their only access to

important information like medical treatment or education.


The solution, of course, is not to use the Internet less. It is to make the

system more efficient, power it with renewable energy, and take advan-

tage of it to solve large, global problems like inequality or the climate


  Action Items

Want to explore the concepts outlined in this chapter even further? Try

these things:

  Create a list of areas where you could have made potentially more • sustainable choices on the last project you worked on. Use the framework in this chapter to run a virtual LCA of a web- • site or application you have either built yourself or use regularly. Identify areas for improvement. Where are there gaps?

  • • Ask your hosting provider what its policy is for using renewable

    energy to power its servers.


After you have identified areas for improvement, you can use the frame-

work described in the following chapters to begin making your digital

products and services more sustainable.



  1 Design Follows Technology

  With all disciplines of design, as technology progresses, the restrictions become lessened and lessened until they get to the point where that wave crashes and there's nothing holding you back from making anything you can dream of. Architecture got bolder when steel frames could support nearly any structure they could dream of without being as restricted by the need to cater to gravity. Furniture evolved with advancements in the technology of the materials available to them such as fiberglass, bending plywood, as well as technological advancements related to mass manufacturing processes. In graphic design, typefaces got increasingly more complex and detailed as printing technologies evolved to the point where those details could be reproduced, then went completely insane when computers removed nearly all graphical restrictions.

  Digital Maturity th

  Wearable technology has been around in one form or another since 17 century China when someone came up with an abacus that you could wear on your finger. So what’s so special about wearable technology at this specific point in time? There’s a reason why wearable technology is making a resurgence right now, and that’s because we’ve finally made it to the digital maturity tipping point, the point in time where overall advancements in both computing and the supporting infrastructure have lessened the restrictions of the design of technology so that they’re nearly non-existent.

  Considering that most people use their computers for simple internet-based information recording and retrieval, the gap in capabilities of a laptop and a computer the size of a watch is not very significant, at least from a technical standpoint. Though physical size of the technology is a big part of this advancement, there’s also the cost of computing which has fallen to a point where this type of technology can proliferate on a major scale, as well as the open-source hardware movement making the tools to create this technology available to a wider range of people. What caused the advancement of Wearable Technology?


Figure 1-1 Graph of Moore’s Law.

Electronics, Volume 38, Number 8, April 19, 1965

  Integrated Circuts

  On a high level, processors work in pretty much the same way they did in the 1950s when the first silicon chip was invented, a series of switches called transistors that have an on state and an off state control the flow of electrons, these are the 1s (on state) and 0s (off state) that compile into what we see on our computer screens. In 1965 Gordon Moore wrote an article for Electronics magazine where he predicted that the number of transistors we can fit on to a chip will double every two years, we now call this Moore’s Law (Figure 1-1). Right now the smallest transistor is down to 14 billionths of a meter wide. Beyond size efficiency, integrated circuits have seen increases in energy efficiency on par with Moore’s Law which is also very important for computers that can’t be plugged in to the wall all the time.

  The Internet

  The internet is obviously a major contributor to the advancement of wearable technology, the data that’s retrieved and recorded on a wearable device doesn’t have to be stored on the device if it can be stored on a server. The internet is also making Moore’s Law slightly less relevant as a computational speed limit because said computation doesn’t have to be done entirely on a local device, an example of this are image recognition programs on the Google Glass that take a photograph and immediately upload it to a server for processing instead of trying to do the computation on the device itself.

  Cellular Data Networks

  Cellular data is the primary access point for un-tethered wearable devices, whether through a phone or the device itself. 4G LTE cellular data has a maximum speed of 100 megabits per second for downloads and 50 megabits per second for uploads, which is fast enough for most wearable applications.

  Open Source

  The last major contributor to the perfect storm of wearable computing is the increased level of access to the tools used to create the technology itself. This is primarily due to the Open Source movement where the source code of software and original designs of devices are made available to the public to reproduce and modify freely. In addition to the designs and technology itself, an incredibly supportive educational community has grown around the Open Source movement.

  The Human Problem

  So we’re now free of limitations around the size of the device, the communication infrastructure, and even access to the production of the technology, but wearable devices are limited by one major thing, humans. Now that we’re free to do whatever we want with technology, we have to figure out how we actually want to move forward. The two big human-related areas are input, how we put information into the computer, and output, how the computer gives information to us.


  There are two high-level types of input for wearable devices, there’s detailed input such as text input or manipulation of objects and menu systems, basically the type of input you’d need to operate a normal desktop computer. Then there’s passive input, which refers to things like steps, heart rate, GPS location, ambient noise and other things that can be collected passively.

  Detailed Input

  Detailed input on wearable devices is difficult, primarily because of the prevalence of the keyboard and mouse. People don’t necessarily want to keep using the keyboard and mouse as primary input mechanisms, it’s just that they’re so dominant that the way we design and think about digital interactions are inevitably stuck in that paradigm. It’s similar to the invention of the television, when there just wasn’t an established content paradigm for the new medium so most of the early television shows were just people reading radio scripts, we’ll eventually get to something more appropriate, but it’ll take a while to turn the ship.


Figure 1-2 Twiddler one-handed keyboards at the Computer History Museum’s “On

You” exhibition.

  Right now there are a few techniques being tried out in terms of wearable text-input, one handed keyboards (Figure 1-2) were popular in academic circles, but are generally considered clunky and difficult to learn. Voice input is getting a lot stronger in terms of recognition, but inevitably comes with some issues such as other people being able to hear what you’re saying and background noise issues. The cleanest method of text input that I’ve seen is actually instructing the user to pull out their phone and use it as a temporary keyboard.

  Beyond text input, there has been some interesting progress in gestural interfaces, particularly with the Myo Armband (Figure 1-3) from Thalmic Labs which reads EEG signals from the movement in your arm muscles. Microsoft’s Hololens also uses gestures as a primary input mechanism by targeting an object by looking at it and gesturing towards that object. In addition to gestures, Microsoft released a “clicker” which is essentially a wearable mouse trackpad to use with the Hololens.

  Passive Input

  Passive input is a lot easier to figure out from a design perspective. It’s basically a matter of finding indicators of certain behaviors, for example, if the accelerometer on my device is moving in a specific pattern, I can safely assume that it’s an indicator that I’m walking and each cycle of that pattern is a step. The most basic example of passive input are fitness trackers, they’re worn all the time and you don’t really interact all that much with them other than swinging your arms around. A more interesting example of passive input comes from MIT Media Lab spinoff Humanyze who distributes sensor-laden lanyards to the employees of a company they’re working with in order to analyze where and how social interactions are happening within the company’s building.


  Output is a much easier problem to tackle with wearable devices, and are divided into active and passive categories. Active output being things like heads-up displays, screen-based smartwatches, and audio feedback, these are things that get your attention to tell you something. Passive output is information that’s displayed more like a dashboard that are more self-initiated or even communicate without commanding active attention.

  Active Output

  We see a lot of examples of active output in wearable devices today. These are the types of interactions that require a conscious engagement, such as when you get an email notification on your smart watch, or a calendar reminder pops up on your Google Glass. It’s pretty fair to say that most notifications are active outputs, but other non-notification things such as the display on a Fitbit are active in the sense that in order to use them, you have to actively engage with the device.

  Passive Output

  Passive output is very different from active. Passive outputs are things that might be always on, and provide information to us without commanding our full attention. A great example of this is the Lumo Lift, a “digital posture coach” that vibrates every time you slouch. After a couple hours using the Lift, you get used to it to the point where it’s almost like another sense, you’re out and about doing things and working on things and you feel that little vibration and you react to it without diverting your attention. Another example of this would be the Withings Activité watches that have a completely passive, always-on interface that you can be aware of out of the corner of your eye without actively thinking about.

  The “Us” Part of Maturity

  The other side of the digital maturity coin is that now, with little to no technological restrictions on design, we must decide how we, as a society, want to live with technology. Now that we can do whatever we want, we have to figure out what we want. What’s healthy? What actually improves our lives? This is the part where human-centered design is most important. We’ve made it through the Wild West days where we had no idea if something was going to be useful or awful, and we have to take a firm look at what exactly is going on, if we’re actually benefiting from technology, and why we’re using technology.

  Moving forward in the design of wearable technologies requires us to be fully aware of the behavior that these devices encourage, good and bad, in order to provide truly useful services that empower people with more control and awareness in their day-to-day lives. We now have enough information to step-back and take an honest look at how people actually use this technology, so we can be critical of what we’re putting into the world and be fully aware of the ramifications of introducing these products into people’s lives.

Chapter 1: Why storytelling matters Intro: It all starts with ‘Once upon a time…” I grew up in a family where stories were always told. My dad is a writer and reading and writing played a big

  part in our childhood. I remember being sat next to dad in the sofa with my oldest younger brother on the other side. Dad would read us the Moomin books, Narnia, the Brothers Grimm, H.C. Andersen and Astrid Lindgren and we’d sit there completely absorbed. The stories he read sent us on journeys to places we didn’t know existed. They created worlds in our minds and sparked our imagination by making us see things we’d never seen before. Just like a good dream you don’t want to wake up from, we never wanted those reading sessions to end. We wanted to know what would happen next, how it all began and how it was going to end. Evening after evening we’d get one step closer until the final page was turned and it was time to start a new chapter, sometimes in a new book. When you look back throughout history, stories have always played an important part in our lives. Long before


  the written word was invented in Mesopotamia around 3200 BC , we’d tell stories about the moon and the stars, battles that’d been lost and won and tales of the world out there that’d we’d not yet discovered. Telling stories was our way to pass along information across generations to make sure that history and facts weren’t lost. It was also a way for us to make sense of the world around us. To find a way to explain what we didn’t understand, the things we feared and that which we desired. Storytelling in various forms and shapes helped us make what we were going through more tangible. Through telling stories we were able to give others a glimpse into what had happened and share that which we believed to be the truth about the world we lived in. It was both a way to educate and entertain, but also a way to pass on and instill moral values. Something that is also evident in a number of the classic fairy tales that many of us have grown up with. Many of these classic fairy tales start with one of the most widely known sentences in the world - “Once upon a


time…”. According to the Oxford English Dictionary, the phrase has been used in one form or another since at

  least 1380 to introduce narratives of past events, typically in the form of fairy tales and folktales. During the 1600 it became commonly used as a way to begin oral narratives and often ended with “...and they all lived


happily ever after.” . Today, a quick search on Wikipedia returns translations and variations of the phrase in 83

  different languages, including my native Swedish where the phrase is “Det var en gång…”. This stems from the Danish writer Hans Christian Andersen’s “Det var engang…” - directly translated as “There was once…” - which opens many of his stories that dad used to read to us when we were kids. Whichever language “Once

  upon a time…” is in, the phrase is most commonly used in fairy tales for younger children where it refers to

  something that happened a long time ago. Amongst adults the phrase is sometimes used to refer to that there used to be a time when things worked differently. Today a number of modern variants of “Once upon a time…” have been introduced, primarily through film and TV. One of the most famous ones are “A long time ago, in a galaxy far, far away…” which opens the Star Wars films. Other well-known variations are “In a land of myth, and a time of magic…” which is the opening line in the TV series Merlin, and “In a time, before time…” from Bionicle. Whether it’s the classic “Once upon a


time…” or one of the modern variants, the way these opening phrases are structured and told, we immediately

know that it’s time to sit back because in one form or another a story will follow.

  The role of stories throughout history

  Across the 200,000 years that we, the modern form of human beings, have been on this earth, how we’ve shared and told our stories have evolved. Both in the tools we’ve used, but also in the role that storytelling has played and shaped society. eclipses. Other early forms of storytelling evolved around hunting practices and rituals and took the form of rock art. The first documented story of this sort is believed to date back sometime between 15,000 and 13,0000 BC and is a series of cave paintings that were found by a group of children in the Lascaux Caves in the Pyrenees

3 Mountains in southern France.

  Rock art has played an important part in many ancient cultures. For the Aboriginals in Australia it was part of religious rituals, but also a way to tell stories of now extinct megafauna such as Genyornis - a large flightless bird - and Thylacoleo - also called the marsupial lion - and more recently about the arrivals of European ships, all of which helped bring meaning to various aspects of human existence. Whilst the aboriginals often told their stories through a combination of oral narrative, gestures, singing, dancing and music, the pictures they painted on the stone walls were there to help the narrator remember the story. Besides rock art as a way to capture stories, carvings, sand and leaves were also used before we had developed a way to write. This allowed us to leave our stories behind and share them with others. As we developed a way to write around 3200 BC, the stories we were telling were able to be transported and shared over larger geographical areas. We used whatever we had at hand and started painting and carving out our stories in wood, bamboo, leather, clay tablets, ivory, skins, bones, textiles and more. But, despite now having a way to document our stories, oral storytelling kept having an important role and ensured that the tales we were telling could travel even further and live on. Aesop’s fables, a collection of fables credited to the slave Aesop who lived in ancient

4 Greece between 620 and 564 BCE , is a good example of this. His fables were kept alive through oral

  5 storytelling and only written down 300 years after his death in 200s B.C.

  Whilst storytelling in the early days were mainly used as a way to explain and pass on information, fables - defined as succinct fictional stories written in prose or verse that features animals, plants and other mystical creatures that are given human qualities - are one way in which stories have been used throughout history to communicate moral value. Even to this day Aesop’s fables continue to teach us lessons about life. But they have also been used for other purposes. In medieval times fables were used as a way to make fun of and satarise tribal events without the risk of retribution and in Ancient Greek and Roman times, fables formed the first of the progymnasmata - a training exercises focused on prose composition and public speaking - where students were asked to learn a variety of fables, expand upon them, make up their own and then use them in persuasive


  arguments. This has a close connection to the role that storytelling can play in design and the workplace overall today, be it for delivering presentations, giving a talk or simply presenting your work. Mastering storytelling, in whichever form and shape it takes, is a real art and something that was widely regarded during the Middle Ages when being a storyteller was seen as one of the finest professions one could have. Troubadours, or bards as they were called, travelled from town to town and village to village and were in great demand as entertainers and educators. There were even competitions where storytellers from different tribes would compete with one another around who could come up with the most compelling and captivating tales to tell. But then in the 1500s things changed and the need for oral storytellers almost disappeared.

  Around 1440 the German Johannes Gutenberg invented the printed press. He created a complete system that allowed for both rapid and precise creation of metal movable type. This allowed prints to be made in large quantities and over the next several decades the printing press spread from a single print shop in Mainz, Germany to around 270 cities across Europe. By 1500 it’s believed that more than 20 million volumes had been printed and during the 16th century the output tenfold and is estimated to have been at 150 to 200 million copies. This gave birth to a completely new industry that took its name after the enterprise of printing itself, the press. It was also the birth of the printed book, which is believed to have been in universal use from around 1480, and the rise of bestselling authors. One such example is Erasmus’ work which were sold in at least 750,000 copies during his lifetime (1469 - 1536).

  The introduction of mechanical movable type led to an era of mass communication which altered society in Renaissance Europe. Information could now circulate relatively unrestricted and threatened the power of both political and religious authorities. There was also a rapid growth in literacy which made learning and education something that no longer was just limited to the elite, but instead bolstered an emerging middle class. The printed press also gave rise to a community of scientists who circulated their discoveries through scholarly journals, and overall it made authorship both profitable and more important. Previously authorship had not mattered, as a copy of for example Aristotle made in Paris would be different from a copy made somewhere else. But now all of a sudden it was important who had said or written what and when and the rule “One

  7 Author, one work (title), one pieces of information” was established .

  In the 19th century the hand-operated Gutenberg press was replaced with steam operated rotary presses which brought the printing press to industrial scale and allowed millions of copies of a page to be printed in one day. This led to further flourishing times for mass production of printed works and dominated both how we consumed and shared information, until of course the radio was introduced to our homes in 1910, the TV in 1927 and the Internet in 1989. Each of these mediums and platforms altered the way we both told and experienced stories as well as the role that stories played in our lives. Whilst it had previously been much more about passing on information and explaining the world around us, with the introduction of the printed press, radio and later TV and internet, it increasingly became about entertainment.

  The role of personal storytelling in our day to day lives

  Though the way in which we tell stories and the role these stories have played have changed throughout the times, what hasn’t changed is the need for and the art of storytelling itself. Human beings are storytelling creatures by nature. In fact, storytelling is one of the many things that define us. Everyday we create stories of our own, from daydreaming to actual dreams at night. It’s our way to put things into context, to process what we’ve experienced and learnt and to test that which we’ve not yet come across, or which we may fear. Just like the books dad would read to us took us to new places in our imagination, the stories we create consciously and subconsciously in our minds take us on journeys too. They create emotional responses in our minds and bodies by making us see and feel things, not just then and there but for hours to come afterwards. Many of us have once had a romantic dream about someone, whether during our teenage years or slightly later, where we wake up in the morning and feel like we walk on cloud number nine. It’s like the dream has left you with a small crush and a secret - what happened in the dream - that makes you feel closer to that person you dreamt about, be it a classmate, colleague, or George Clooney as it was for me back in the good old days when ER was shown on TV. Then of course there are the dreams where we find ourselves naked in not-suitable-to-be-naked situations. Or, when we the night before a big presentation dream that we’ve forgotten everything. Those dreams can leave us with a sense of unease that we struggle to shake for hours to come.

  Whilst there has been many theories and speculations about why we dream and what the 3-5 dreams each of us on average have per night might mean, many scientists now endorse the Freudian theory of dreams which stipulates that they reveal hidden desires and emotions. Other well-regarded and prominent theories are that dreams assist in problem solving and memory formation, but also that, at times, dreams are simply random creations by our brain’s activity. Besides nightmares many of us in today’s society have positive associations with dreams. From re-telling the ‘you-wouldn’t-believe-what-I-dreamt-last-night’ dream to fascinations around what it all might mean. Back in the middle ages, however, dreams were seen as evil things with Martin Luther amongst others believing that dreams were actually the work of the devil. He wasn’t the first to believe that dreams were messages from “someone” else. In the Greek and Roman times, people believed that dreams were messages from deities and that they predicted the future. This led to some cultures practising dream incubation in order to cultivate dreams that would reveal prophecies. them so that they could be interpreted. Today there are many advocates that encourage you to write down your dreams, both as a way to help you remember your dreams better, but also as a way to understand how you were feeling and what you were going through at certain times. If neither of that is of interest, writing down those

  8 crazy dreams can at the very least provide for some entertaining reading at later times.

  As for its broad daylight sibling daydreaming, this was for many years associated with laziness and as something that could potentially be dangerous. This primarily came into play when the work we did moved into assembly lines and were dictated by the motions of the tools that we were using. These tools left little room for zoning out and the short term detachment from the immediate surrounding that daydreaming inherently involves. Today daydreaming, just like night time dreams, are widely accepted as a way for the mind to ease boredom and consolidate learning. For many daydreams involves happy associations, hopes and ambitions of things coming to pass and it’s now commonly acknowledged that rather than being a sign of laziness, daydreaming can lead to constructive results, from developing new ideas both creatively and scientifically. In this sense daydreaming is not too dissimilar from visualisations that are used to create mental images of what we want to happen or feel in real life.

  Visualisations are another form of powerful day-to-day storytelling that is used by many professional athletes as part of their training process before big competitions and events. By going through and narrating in their minds how they would like the event to unfold or feel, the athlete is doing a mental rehearsal that is scientifically proven to cultivate a competitive edge, increased mental awareness and a heightened sense of well being and confidence. This mental rehearsal is found by research to have positive effects on both mental and physical reactions and is as powerful in business as it is in sports. Through visualisations that involve the visual, kinetical


  or auditory senses we step right into the story we’re creating and this also positions us better for passing on and communicating our desired outcomes to others.

  Storytelling today

  Besides the daydreaming and nighttime stories that we create ourselves, we’re surrounded by stories created by others. From films and series that we watch on TV and on demand, to both shorter and longer narratives in the form of text, visuals and audio. Whilst stories back in the days traditionally was something you would observe or be a passive listener to, today’s stories invite active participants. We create campaigns where we ask users to contribute with their own stories in the form of content and we provide interactive mediums and platforms where the user or audience can influence the next turn of events in the narrative. With Virtual Reality (VR) being at the brink of becoming mainstream in 2016, we’re now at a point where we’ll soon immerse ourselves right into the world of that story. But our exposure to storytelling isn’t just happening through the traditional platforms of screens, books and magazines. We’re increasingly experiencing storytelling through politicians on TV, CEOs and leaders in their speeches and even in the restaurants, exhibitions and open space we visit. We’re living in a world that’s increasingly being curated to help cut through the noise and wealth of information out there so that we can find what we’re after easier. Or, if you look at it from the seller’s point of view, find a way to stand out from the competition. Whether it’s a restaurant owner, an exhibition organiser, a CEO or a store, we’re all looking for interesting angles and different ways to get our message and our product and service offering across. From curated dining experiences where the food we eat is matched to what we drink, to immersive experiences where the visitor becomes part of the exhibition. This shift towards curated experiences is not just something that is driven by brands. It’s also reflected in consumer behaviour and what we call the “experience economy” where we’re seeing people spending less money on clothes and food and more on holidays, cars, entertainment and


  eating out . This shift in consumer spending behaviour is partly what is driving the rise of curated experiences and content marketing and strategy overall. And it’s all with good reason. Whilst pairing menus and immersive at connecting with our audience has increased and this is where the importance of mastering the art of storytelling comes in, whether it’s for us as individuals, as leaders, startup founders, brands or as designers. In his book ‘Tell to win’, author and Hollywood producer Peter Guber mentions how we’re increasingly told that


“without a compelling story, our product, idea, or personal brand is dead on arrival.” and that “If you can’t tell


it, you can’t sell it”. In a world so full of noise and competition, whatever your view on storytelling is, there’s

  no denying that these quotes holds true. But there are other reasons why we should pay attention to the art of storytelling. For a long time we’ve only been able to speculate about the persuasive effect of stories, but over the last several decades, serious studies have been undertaken into how stories in their various forms and shapes affect the human mind and how we process information. What they’ve found is that when we read dry, factual arguments, we read with our guards up. We’re skeptical and critical. When we’re presented with a story on the other hand, something happens to us. We lower our guards and are moved emotionally and this somehow seems to leave us


  defenseless. In a Fast Company article from February 2012 titled ‘Why Storytelling is the Ultimate Weapon’ , Jonathan Gottshall writes:


“Results repeatedly shows that our attitudes, fears, hopes, and values are strongly influenced by story. In fact,

fiction seems to be more effective at changing beliefs than writing that is specifically designed to persuade through argument and evidence."

  It’s not surprising then that brands and organisations are picking up on the importance of storytelling in relation to connecting and engaging with their customers. But how much truth is there really to the persuasive effects of stories? And to what extent can we use it in design?

  Stories as persuasion tools

  Besides serving as a means to pass on information and entertain, stories and storytelling has been used throughout history from tribe wars to activists to move people into action. To get behind a cause and drive people forward, whether it’s to move past enemy lines, fight for a cause, or go on a pilgrimage. Peter Guber talks of purposeful stories and how they are essential in persuading others to support a vision or a cause. He defines a purposeful story as “A story you tell with a specific purpose in mind” and draws the parallel


  that it can be as simple as the laughter you want at the end of the joke you tell. Whether it’s a laugh we’re after or something else, in every single story we tell, there is always a desire of an act and / or a response that we’re looking for from the person or people on the other side. What many of us fail to do is be clear about what this action is, or even one step before that, to stop and think about what the purpose of our story is and how we best connect it to get the action or response we want. The reason why stories have a persuasive effect on us is because they connect with us emotionally. Whether they tap into something that’s already there, like an interest, passion or subject that matters to us, or actually convinces us through the story itself and the way that it is told, the key to the persuasive effect lies in the connection the story and storyteller make with the person on the other side. Sometimes this isn’t so much through the words that are spoken but instead through the actions that are taken. One of the most famous people who told a purposeful and persuasive story in such a way is Leymah Gbowee, an activist who back in the summer of 2002 was the spokeswoman and inspirational leader for the Women of Liberia Mass Action for Peace. Their peace movement started with local women singing and praying in a fish market and evolved into daily non-violent protests and sit-ins to protest against the then tyrannical president Charles Taylor. Gbowee’s story through actions brought women together across religious and ethnic borders and together their peaceful

  Whilst there in any good story tends to be a promise of something to come, inherent in purposeful stories is the purpose or goal that the story is leading up to and which the storyteller is trying to get people behind. Whatever purpose we might have with the story we tell, actions form a big part of how we tell that story, from the subtle actions that we ourselves take in how we tell and communicate our story, to the actions we ask the person on the other side to take. Actions are particularly important when we talk about purposeful and persuasive stories in relation to design and business. Not the least with regards to those we tell within organisations as a way to get internal team members, stakeholders and clients to buy into UX and the solutions we’re proposing. But it also forms a key part in how we design experiences and the actual call to actions (CTAs) that we include and direct users to, from the primary one to the secondary and supporting ones. These CTAs need to resonate and correspond to what matters to users, both in terms of what they need and with regards to where they are in their journey. Only then will it be something that moves the user into action. And this is where mastering the art of storytelling comes in.

  What makes for a good story

  From ancient times to the modern day, the art of storytelling has always been a well respected skill and those who were great at telling stories have often been key figures in their communities. With tribes and indigenous people, great storytellers were those who were particularly gifted at communicating a story in a memorable and effective way that not only told the sequence of events, but also evoked emotions amongst the listeners. This was key for ensuring that whatever the tale was about, be it war, deeds or events, wasn’t forgotten but could be passed on. Today technology and print has solved the challenge of how to pass on information, but the aspect of great storytellers being key figures holds true just as much now as it did back in the days.

  14 In December 2014 Raconteur published an article and infographic called ‘The World’s Greatest Storytellers. It

  was the result of a survey that included almost 500 authors, journalists, editors, students, media and marketing professionals, from bestselling authors to editors-in-chief. When asked who they considered to be the best storyteller in history, 53% of the respondents chose living storytellers and 47% chose a storyteller who had passed away. Whilst the top five included William Shakespeare, J.K. Rowling, Roald Dahl, Charles Dickens and Stephen King, there was great diversity in the responses Raconteur received. They spanned continents, genres, disciplines and mediums, as well as the reason behind the respective choices. Some picked influential people in their own fields. Others family members who’d influenced their love for writing and history. Others yet again based their choice on their own conviction of ‘great’. What unites them all is that their reason for picking their favourite storyteller is that in one way or another their stories had hit home. This points to one of the key aspects of a good story. However we tell or create a story, all good stories captures the audience's attention and resonates with them. At times it’s almost spell bound. It’s the page turner that draws us in so deep that we can’t but keep on turning the page to find out what happens next, and then after that. Other times it’s a gripping tale or faith that connects with us emotionally and moves us to excitement, or even rage.

  “This can’t be right”, “How could they let that happen?”. In all instances of good storytelling there is a bit of

  magic involved where the narrator draws their audience into the world they’re telling a tale about and to some extent holds them captive by building up the anticipation around how the story will end. Will the boy get the girl? And will they live happily ever after? The element of anticipation is a really powerful tool in both how we structure and tell our story. One of the world’s most talented storytellers, who utilised this technique, was the legendary Persian Queen Scheherazade, of Arabian Nights (or ‘One Thousand and One Nights’). She married King Shahryar who after his first wife had been unfaithful to him had resolved to marrying a virgin every day and kill his previous day’s wife, just so that she wouldn’t be able to cheat on him. By the time he was introduced to Scheherazade, he had killed 1000 wives. the time she’d finished telling her stories, King Shahryar had fallen in love with her and spared her life, for

  15 good.

  Each and everyone of us has experienced a story that completely absorbs us. From a book we simply can not put down but keep on reading late into the night, to a movie or play that leaves us reflecting and thinking about what we’ve just seen for days to come. The question of what makes for a good story is what started the journey of this book. It’s the one I called and asked my dad about when I was preparing for my storytelling talk. I wanted to find out if there was a magical recipe or formula to follow, besides a beginning, a middle and an end, and whilst I wasn’t expecting a “Well, yes there is!”, the answers he gave and the conversations and later research this sparked, turned out to be far more interesting than what I could have imagined. From the art of dramaturgy and the three ingredients my dad said should be present in any good story, to just how much that we experience on a day to day basis that’s transcended by storytelling. Everywhere you look and everywhere you go there is a story to tell and exactly what makes for a great one and how this is connected to design and multi-device projects is the story I aim to tell you through this book.

  The role of storytelling in design

  A large part of what we do day to day at work is tell stories, whether we’re aware of it or not. From using a metaphor for explaining something, to talking through the experiences we’re designing through the eyes of the persona or user journeys we’ve defined. As UX designer we’re used to using storytelling in various forms for defining and designing our work. At times it’s more explicit and other times something we’re not really aware of, but just do without giving it too much thought. As the complexities of what we design for are increasing, there are tangible and measurable benefits involved with applying storytelling principles in a more active and explicit way, from the start to the finish of your project. Storytelling is a tool that can both help us define and design our work, but it’s also a really impactful tool for ensuring we get buy-in for UX, the ideas and suggestions we have and how we recommend that we define and design those. As UX designers, strategist, marketers and product owners, we’re used to defining who our target audience is. To varying degrees we develop personas, pen portraits, user journeys, flows and other deliverables to help us, the rest of the team and key stakeholders in having the customer front of mind. We define, amongst other things, what matters to them (needs), what they want to accomplish (goals), what may worry them (concerns) and any hinders (barriers and risks) in the experience. But, we often forget to apply these same principles and tools to the people we work with and more importantly the people we report into and present our work to. Whether it’s a client or an internal stakeholder, the higher up the food chain you go, the more time poor the person in question is and the more critical it is to understand what really matters to them. Failing to do so will result in misalignments and miscommunication and in really severe cases it can damage collaboration and relationships, be they internal or external. “Death by Powerpoint” is a(n almost) real thing and more often than not we over deliver in actual work deliverables, but under deliver in the value that the work actually brings to our clients, and at times users. This is where the importance of storytelling comes in and where simple frameworks and principles can help ensure that you’re both telling a clear and well structured story, but also that it’s relevant and told in the right way for your audience in question. Whether that’s in how you narrate and present your presentation, or how you plan out the experiences of the websites, apps etc. that you work on. As for the customer on the other end, whether the user is just finding out about your product or service, or has been a loyal customer for a while, each and every user is different and though there are similarities between each user’s journey, each journey is also different, no matter how much we’d like to plan out that they go from A to B to C and so forth. When you look at these journeys and break down what a user may think, feel, need and user with buttons to click or tap on and elements that can be swiped right or left, up or down, inherent in the fact that we architect the experience and that the user controls what they do next, is that these experiences become interactive and as an extension of that, the stories we design through experiences are then also interactive. The only problem is that pretty much anything can happen and we have very little control.

Chapter 2 – Basic VUI Design Principles This chapter gets right into best design practices for designing today’s voice user interfaces

  (VUIs). It covers what “conversational design” means and the best way to achieve it. This includes knowing the best way to confirm information spoken by a user, whether your system should be “command and control” style vs. conversational, how to handle novice vs. expert users, and especially important, designing for when things go wrong.

  This book is focused on designing VUIs for mobile apps and devices. Let’s start by examining how designing for mobile is different than designing for IVRs (Interactive Voice Response).


  In the early 2000s, IVRs were becoming more common. Initially a touch-tone/voice hybrid (“Please press or say one”) and very primitive, they became an expected way to communicate with many companies. IVRs could help callers get stock quotes, book flights, transfer money, and provide traffic information. Many of them were designed poorly, and websites popped up with back doors on how to get transferred immediately to an operator (something many companies actively tried to hide). IVRs got a bad reputation, ending up the subject of satire on Saturday Night Live. [Julia / Amtrak-could put link on website to this skit] In 2004, the book Voice User Interface Design (Cohen, Giangola, Balogh) was published and became the de facto guide for making well-designed and useable IVRs. If you’re designing an

  IVR, it continues to be an excellent resource, and is a recommended text.

  IVRs were created to automate tasks so customers would not always have to speak to a live person to get things done. They were created before the Internet became commonly used and before smart phones were invented. Nowadays, many IVRs are used as the “first response” part of a phone call, so that even if the caller ends up speaking with an agent, basic information has been collected (such as a credit card number). For many tasks, even complex ones such as booking a flight, an IVR can do the job. In addition, IVRs are great at routing customers to a variety of different agent pools, so that one phone number can serve many needs. Finally, some users actually prefer using an IVR over speaking with an agent, because they can take their time, and ask for information over and over (such as the early Charles Schwab stock quote IVR) without feeling they’re “bothering” a human agent.

  Although some of the design strategies from the IVR world also apply to mobile VUI design (as well as VUI for devices), mobile VUIs also present a unique set of challenges (and opportunities). This chapter outlines design principles for the more varied and complex world of designing VUI systems.

  One of the challenges of mobile VUIs is whether or not it will have a visual representation, such as an avatar. In addition, when will your VUI allow the user to speak? Will users be able to barge in? Will it use push-to-talk? These challenges are discussed later in the book. One of the opportunities that mobile devices have that IVRs do not is that mobile devices can have a visual component. This can be a big advantage in many ways, from communicating information to the user, to confirming it, even to help the user know when it’s their turn to speak. For example, let’s say you ask Google, “Who are the 10 richest people in the world?” Google could certainly read off a list of people (and their current worth), but that is a heavy cognitive load. It’s much better to display them, as shown below.

  Figure 1 - Google Shows results for voice request

  Taking advantage of the visual capabilities of mobile is essential to creating a rich VUI experience. In addition, this visual component can allow the user to continue at a more leisurely pace. In an IVR, it is rare to be able to pause the system—instead, the user must continually interact.

  Another of the opportunities mobile apps have is that in many instances, they are not replacing a live operator. This means your user has a different set of expectations. If I ask Apple’s Siri when my next meeting is, I don’t expect a live agent anywhere in the background. There are certainly some mobile apps that perform a hand-off to a live agent, but it is much less common.

  Another current common difference between IVRs and VUIS on mobile apps or devices is that they are often used for one-turn tasks. For example, I’ll ask Cortana to set an alarm (Figure 1), or Google what the fastest land animal is, or Amazon Echo’s Alexa to start playing my favorite radio station. These types of interactions are quite contained and do not require the system to maintain a lot of information.

  Figure 2 - Cortana confirming voice-requested alarm

  Although this is quite common now, do not confine your VUI experience to this model. To start thinking more specifically about how to best design VUI for mobile, let’s dive first into the topic of conversational design.


  Imagine you’re having a conversation with a friend. You’re sitting in a coffee shop, catching up after a long absence. Your friend says, “Did you see the new Star Wars movie?” “Yes,” you reply. “Did you like it?” she asks next. You say “I’m sorry, I don’t understand.” No matter how many times she repeats herself, you never answer her question.

  That level of frustration is about where we’re at with many VUI systems today. Despite the many recent advancements of speech recognition technology, we’re still a long way from simulating human conversation. Here’s a real world example from OK Google, illustrating two conversational “turns” (a turn is one interaction between the user and the system):

  USER Ok Google. When’s my next appointment?

  GOOGLE You have a calendar entry tomorrow. The title is ‘Chinatown field trip’. Ok Google. Can you please repeat that? GOOGLE


Figure 3 - Google not being conversational

Google has let down its end of the conversation. It’s like the first part never happened.


Conversational design is becoming a common term, but it is often misused. Many people use it

  to mean any time you have an interaction with a system in which you speak, or text. But many of these “conversations” have only one turn; for example, asking Hound where the nearest coffee shop is located. “Chatbots” are also very trendy, but very few chatbots use conversational design either. (A ‘chatbot’ in this context is an automated system that can respond to simple queries, generally via text and not speech.) In this book, I define conversational design to mean thinking about an interaction with a VUI system beyond one interaction. Humans rarely have conversations that only last one turn. Design beyond that one turn; imagine what users might want to do next. Don’t force them to take another turn, but anticipate and allow it. In addition, it is vital to keep a recent history of what the user has just told you. Having a conversation with a system that can’t remember anything beyond the last interaction makes for a dumb and not very useful experience.

  When designing a VUI, many people only consider one-off tasks, such as answering a search query, setting up a calendar appointment, placing a phone call, playing a song, etc. And sometimes these tasks can be accomplished in one fell swoop. But the best VUI designs also consider what happens next.

  Here’s an example in which Google does a good job of remembering what occurred in previous conversational turns: USER


  Ok Google. Who was the 16 President of the United States?



  Abraham Lincoln was the 16 President of the United States.

  USER How old was he when he died?

  GOOGLE Abraham Lincoln died at the age of 56.

  USER Where was he born?

  GOOGLE Hodgenville, KY

  USER What is the best restaurant there?

  GOOGLE Here is Paula’s Hot Biscuit:


Figure 4 - Best restaurants in the town Abraham Lincoln was born

  It’s not quite the same as talking to a human, but Google successfully carried on the conversation for four turns, knowing the references for “he” and “there”. A good rule of thumb is to let the user decide how long the conversation will be.

  Setting User Expectations

  Good conversational design is not just about crafting nice prompts. As Margaret Urban, Interaction Designer at Google, suggests: don’t ask a question if you won’t be able to understand the answer. She gives the example of a prompt that occurs after the user has finished writing an email: “Do you want to send it or change it?” One response, which you may not initially have planned for, will be “yes”—so build a response in your system to handle it. Although good prosody (where you place the emphasis) can help with this issue, it is often not enough. In a case where you’re seeing a lot of “yes” responses, you may want to reconsider rewording the prompt to something more clear, such as “What would you like to do—send it, or change it?” Urban emphasizes it’s important to set user expectations early on. How does your app introduce voice? You can offer a “tour” to first-time users, and offer educational points along the way.

  “When someone has successfully completed a VUI interaction, it’s a bit of an endorphin boost— the user has a glow of completion and satisfaction. It’s a great time to educate people—“Since you were great at that, how about trying this?’” Be careful about telling users tasks were successful. Urban says “’Setting the alarm’ implies to you go with a user’s mental model, however, you need only let them know if the alarm setting failed. Urban offers a good analogy about designing with breadth. Perhaps you’ve designed a system that allows people to set an alarm—but not given them a way to cancel it. She likens this to giving someone a towel for a shower, but no soap. If you set an expectation you can accomplish a task, think about the corresponding task that goes with it.


Discoverability is another important element of design. How does your user know what and when

  they can speak? I discovered my Android camera app was voice-enabled purely by accident— while taking a picture one day, I naturally said “smile!” and the photo clicked. I quickly discovered I could also say “1.. 2.. 3!” and it would also take a photo. This is a great example of piggybacking off of user’s natural speech.

  When asking the user for information, it’s often better to give examples than instructions. If you’re asking for date of birth, for example, rather than say “Please tell me your date of birth, with the month, day, and year”, use an example, e.g. “Please tell me your date of birth, such as July 22, 1972.” It’s much easier for users to copy an example with their own information than translate the more generic instruction. [Don’t make me Tap! p88] To assist you in creating great conversational designs, let’s talk about tools.


  Sample Dialogs

  One of the best (and cheapest!) ways to start your design process is something called a sample dialog. A sample dialog is a snapshot of a possible interaction between your VUI and your user. It looks like a movie script: dialog back and forth between the two main characters. (The Google examples above are in the form of sample dialogs.) Sample dialogs are not just a way to design what the system will say to the user (or display to the user). They are a key way to design an entire conversation. Designing prompts one at a time often leads to stilted, repetitive, and unnatural-sounding conversations. Pick five of the most common use cases for your VUI, and write out some “blue sky” (best path) sample dialogs for each of the cases. In addition, write a few sample dialogs for when things go wrong: the system did not hear the user, or misunderstood them. Sample dialogs are very low tech, but they are a surprisingly powerful way to determine what the user experience will be like, whether it’s for an IVR, a mobile app, or inside the car. In addition, it’s a great way to get buy-in and understanding from various stakeholders. Sample dialogs are something anyone can grasp, and quickly.

  A great tool for this is the screenwriting software Celtx, but any place you can write text will do. Once you’ve written some sample dialogs, a very useful design exercise is to do a “table read”: read it out loud with another person. Another great use of sample dialogs is to record them, either using voice talents, or text-to-speech (whichever will be used by your system). It is slightly higher cost than simply writing them, but an even more powerful way to know if the design sounds good before investing in more expensive design and development time.

  Visual Mockups

  When designing a mobile app, wireframes / mocks are of course also an important piece of your early design process for a VUI app—they’ll go hand-in-hand with sample dialogs to help visualize the user experience. As this book is VUI-focused, we do not go into detail about best practices


  Once a variety of sample dialogs have been written and reviewed, the next step is to sketch the

  VUI’s flow. “Flows” (referred to as “callflows” in the IVR world) are diagrams that illustrate all the paths that can be taken through your VUI system. The level of detail for this flow depends on the type of system you are designing. For an IVR, or a closed conversation, the flow should include all possible branches the user can go down. (Figure 5) This means that for each turn in the conversation, the flow will list out all the different ways the user can branch to the next state. This could be for simple states, that allow only “yes” and “no” type responses, as well as more complex ones that might have 1000 possible song titles. The diagram needn’t list every phrase someone can say, but should group them appropriately.

  In the case of something more open-ended, such as a virtual assistant, the flow can be grouped into types of interactions. For example: calendar functions, search, calling/texting, etc. In these cases, not all possible interactions can be spelled out, but it helps to group the various intents. (Figure 6)

  Figure 5 - Sample flow with complete branching

  Figure 6 - Flow for system with more complex branching CONFIRMATION

  Once you’ve designed the basic flow and have a set of sample dialogs completed, you can focus on some of the important details, such as confirming input. Making sure users feel understood is an important part of any good VUI design. This also serves the purpose of letting a user know when they were NOT understood. Back in the days of the IVR world, confirmations were sometimes used to an excessive degree. For example:

  IVR TRAVEL SYSTEM Would you like me to book that flight?

  CALLER Yes please.

  IVR SYSTEM I think you said ‘yes’. Is that correct?

  CALLER Yes. Okay, I will book your flight… A funny example of why over-confirmation can sound so unnatural was shown by a Saturday Night Live skit from 2006, called “Julie the Operator Lady”. It was a parody between Rachel Dretch (playing the part of Julie, the Amtrak IVR persona) and Antonio Banderas, talking at a party. Now, the Amtrak system is actually an award-winning IVR—but you can see how the same style of dialog doesn’t quite work in the real world:

  JULIE Will you be needing drinks tonight?

  ANTONIO BANDERAS Sure, thanks.

  JULIE Please listen carefully, as the options have changed—or, if you know the drink you want, you can interrupt me at any time. There’s merlot, chardonnay, various cocktails—

  AB Gin and tonic, on the rocks, lime…

  JULIE Before I get your drink, lemme see if I have that right. That’s one, gin and tonic, on the rocks, with a twist. Did I get that right?


  Your approximate wait time is 1 minute… [Julie walks away] Over-confirming information may ensure accuracy, but it will also drive people (like Antonio Banderas) crazy. When determining the right confirmation strategy for a VUI experience, consider the following:

  • What is the consequence for getting it wrong? (Would the wrong flight get booked?

  Would someone’s money get transferred to the wrong account? Would they not find the right search results?)

  • What channels are available to the system? Auditory only? (IVR) Non-textual visual feedback? (Amazon Echo) A small screen? (iPhone4) A large screen? (iPad)
  • What type of confirmation is most appropriate? (Explicit confirmation? Implicit? A hybrid?) If someone is transferring money from one place to another, confirmation is extremely important. On the other hand, if the app is purely for entertainment, getting one thing wrong may not be a big deal, and in fact may break the user out of the experience of the conversation. Always take advantage of the channels you have available. When you want to talk to the Amazon Echo (by saying “Alexa”), the top rim of the device lights up in blue, letting you know
not light up until it knows you want to initiate a conversation with it. I talk more about how devices show you they’re listening in Chapter 5. When you ask Amazon’s Echo a question, its primary channel for output is audio only. It does have the ring at the top that lights up, but this is generally just to indicate it’s listening (or processing), not to give you actual information. In addition, you can access information you give to the Echo on another device—such as your phone—after the fact (for example if you asked Alexa to add an item to your shopping list). For VUI experiences such as this, where audio feedback is the only real channel to provide feedback, it is important to craft confirmations carefully. One way to do this is using confidence thresholds. A confidence threshold is how the speech recognition engine lets you know how well it thinks it’s performed. For example, it might recognize that you said “YES PLEASE”—but the confidence could vary. Perhaps it’s only 45% confident that is what it heard, or perhaps it’s 85%. Your app can access this information, and handle them in different ways.

  Explicit confirmation: force the user to confirm the information. For example, “I think

  you want to set a reminder to ‘buy insurance before going skydiving next week’. Is that right?’

  Implicit confirmation: let the user know what was understood, but do not ask them to

  confirm. Example: “Okay, setting a reminder to buy insurance…” (In the case of implicit confirmation, it may be necessary to allow the user to cancel or go back a step.)

  Method 1: 3-Tiered confirmation

  In this case, the system will explicitly confirm information between a certain threshold (for example, 45-80), reject anything with a lower confidence, and implicitly confirm anything above the 80% threshold. It’s especially important to explicitly confirm information if the cost of a misrec is high. Here is an example of how this might sound:

  USER Please buy more paper towels.


  [>80% confidence, implicit

confirmation] Okay, ordering more

  paper towels… [45-79% confidence, explicit

  confirmation] I thought you said you’d

  like to order more paper towels. Is that correct?

  [<45% confidence.] I’m sorry, I didn’t

  get that. What would you like to buy?

  Method 2: Implicit confirmation

  Another method is to only confirm things implicitly, without requiring the user to take action. For example, if I ask “What’s the world’s tallest mountain?” a system with implicit confirmation would give me the answer right away: “The world’s tallest mountain is Mount Everest.” It included a piece of the original question, so I know the system recognized me, as well as the answer.

  Some systems provide the answer without confirming the original question, e.g. simply saying “Mount Everest”. This approach can be appropriate when the confidence is very high, and when trying to be more conversational.

  Other examples of implicit confirmation: “The tallest mountain in the world is…”

  • “Okay, I’ve set your appointment for 10 AM tomorrow morning.” • “All right, I’ll tell you a story…”
  • “Cheetahs are the fastest land animal.”
  • Method 3: Handling errors

  In the traditional IVR world, if the user was not heard or understood, the system prompts the user to speak again. This is important because otherwise the user might think the phone call was cut off, or the system isn’t functioning. Another reason is that there is a fixed dialog that is expected; at any given point, the user is required to provide input to move the conversation forward. If the system doesn’t hear anything, it times out, and prompts the user to speak again. If these timeouts are not planned carefully, the user and the system often end up cutting each other off and in awkward back and forth. In the mobile VUI and device world, however, it’s not always necessary to reprompt the user when there is a failure. The Amazon Echo, for example, does nothing if it doesn’t hear you. (If it heard but did not understand, it plays a short sound.) When speaking to a device (especially a device with a name), users are more likely to respond to silence the same way they would with a human—by repeating themselves. In addition, the system is not waiting for the next piece of the conversation, as this is often a one-off command. Therefore, if the system fails to respond, it’s not that the whole transaction fails—just that one particular command. It’s not the same as suddenly ending a conversation in the middle of an important transaction; it has just failed once. Because of the higher fault tolerance, it’s more forgiving for Alexa to ignore you when she did not understand.

  In addition, there is the annoyance factor. If a user has called up an IVR to complete a task, such as get traffic information, or check how much they owe on their credit card, the system needs to do everything it can to make that task successful, even if it means reprompting the user with the often-annoying “I’m sorry, I didn’t get that.” Imagine, however, that your app did that every time it failed to understand your latest command. It would get old very quickly. “I didn’t understand, I didn’t understand, I didn’t understand.” All right, already! A user can quickly get used to the mode that if the device did not understand, just try again.

  Method 4: Non-speech confirmation

  Another type of confirmation relies on completing an action that does not require a spoken response. For example, imagine you are creating an app to turn the lights on and off in your home. The user says: “Turn on the Christmas tree lights.” Is it really necessary to say “Ok, turning on the Christmas lights,” when the lights just turned on? One caveat to this approach: you may want to consider having an audio confirmation that the know it’s going to happen, even though it may take a few seconds. In addition, this lets the user know that if the lights do NOT turn on, it was not that the device did not hear the user. Another type of confirmation that does not use speech but does use audio is an “earcon”: a brief, distinctive sound. In the 511 IVR (which provides traffic & transit information), when the user returns to the main menu, a specific, short audio clip plays. When the user goes to the traffic menu, a short car horn beep is played. The user quickly understands they have been taken to the right place.

  Method 5: Generic confirmation

  In some conversational systems, it may be appropriate to not confirm exactly what the user said—even implicitly. This applies more for conversational systems in which the user may be doing more open-ended chatting. For example, you may have an avatar that asks how someone is feeling, but does not necessarily act on that information.

  AVATAR How are you feeling today, Cathy?

  CATHY Well, pretty good today, I guess. AVATAR


Thank you for sharing that with

me. How did you sleep last night?

  CATHY Not so great. AVATAR I’m sorry to hear that.

  This sort of “generic confirmation” can allow for a richer sharing experience from the user. This type of response will allow a wide variety of input from the user, while still moving the conversation along. Keep in mind that in human-human conversations, we don’t go around confirming exactly what someone says every time. Sometimes we might say “mm hmm” or “tell me more”—and it’s ok for a computer to do that as well. To keep things more interesting, it helps to randomize these types of generic confirmations. Note the confirmation in the second piece, as well: when the user says she did not sleep well, the avatar does not say “So you did not sleep well,” but rather offers up a sympathetic response. This turn in this conversation could have three categories of responses:

  Slept well (“I slept great!”, “Good thanks”)

  • Slept poorly (“Not so great”, “I barely slept”, “I slept terribly”) • System is not sure (“Well I dreamed a lot”, “I was up late”) • There could be a set of appropriate randomized responses for each of these cases.

  Method 6: Visual confirmation

  With mobile devices, visual confirmation is a commonly used method. For example, when asking Google a question, it often provides an audio confirmation, as well as a visual.

  USER Ok Google. When’s my next GOOGLE You have a calendar entry tomorrow. The title is Chinatown field trip.

  Take advantage of the screen! Communicating a list of items is much more effective on a screen. Humans have a limited capacity to remember a string of information; typically, a user can’t remember more than about 7 auditory items at a time. [Ref: The Magical Number Seven, Plus or Minus Two: Some Limits On Our Capacity for Processing Information, Miller 1956] “The ephemeral nature of output in auditory interfaces places significant cognitive demands on the user.” (p6, VUI Design) However, by putting those same items into a visual list on a screen, the cognitive load is lessened. Now the user can examine the list at their leisure, without remembering each little detail. This also gives the user more time to make a decision.

  Another great way to use the screen is for confirming the user’s choice. Imagine a system that allows the user to select a response by either speaking or pressing a button. The system asks “Did you take your antibiotics last night?” and provides a Yes and a No button. The user could say “Yes”, and rather than have the system say “I think you said yes,” simply highlight the button, as if the user had pressed it. The user will know they have been correctly (or incorrectly) understood. [See Figure 7]

  Figure 7 - Users can speak or press buttons to reply to Sensely's avatar

  In addition to thinking about confirmations, an important design decision is when to allow your user to speak.


  Most of the VUI systems out there today are “command and control,” meaning the user must explicitly indicate when they wish to speak. Another type of design, which is becoming more common as systems can be more conversational, uses a more natural turn-taking approach, To know which type is right for your VUI design, answer the following questions:

  Can the user ask the system a question/give a command at any time? (E.g. Siri, Google • Now, Amazon Echo, Hound) Will the user be involved in a closed conversation with an explicit start and end?

  • (Chatbot, game, avatar)


  Many systems use this approach, in which the user must do something explicit to indicate to the system he or she is going to speak. For example, Siri requires the user to press the home button before speaking (or, from the Siri screen, press the microphone soft button (Figure 8)). Ok Google requires either pressing the microphone icon, or saying “Ok Google” (Figure 9). Alexa has a physical button, but typically the user indicates they are about to speak by saying the wake up word, “Alexa”. Even the starship Enterprise required the crew to say “Computer!” before making a request.

  Figure 8 - After holding down home button to activate Siri and then failing, microphone button appears Figure 9 - To speak to Google, say 'Ok Google' or tap microphone button

  In many cars, the user must use ‘push-to-talk’: press a specific physical button on the steering wheel, for example, to indicate they will speak.

  Figure 10 - Toyota Matrix "push to talk" button

  When this happens, the system typically responds with either non-verbal audio (the “bonk” noise), and/or with visual feedback (a wavy line, animated dots, a glowing light on the device). The user then knows the system is listening, and speaks.

  Figure 11 - Listening indicators: Hound, Cortana, Google, Siri

  When the system has decided the user has finished speaking, it indicates in some way, often with more non-verbal audio (the “up” noise from Siri), and then responds. At this point, the conversation may well be over. The system is not expecting the user to speak again, although sometimes it can handle a related follow-up request. But the user must again

  explicitly indicate they are going to speak.

  This works well for situations in which the device has no idea when the user might speak. Imagine you’re in the family room and your spouse is in the kitchen. You haven’t spoken for a while. You will probably do some kind of explicit notification to your spouse to let them know you’re going to speak, such as “Hey honey, can I ask you a question?” or “Hey Chris”. This prepares them to listen. If you’ve had 30 minutes of not speaking and suddenly you said “Where is it?” they’re likely to be confused.

  The time window in which your system should still be listening after the wake word / button push should be carefully chosen. Too short, and you’ll miss users who hesitate briefly before speaking; too long and the system may be listening to a conversation that’s not intended for it. Ten seconds is a good starting point.


  When there will be a longer back-and-forth between the user and the VUI, forcing the user to explicitly indicate they are going to talk is not necessary and can be cumbersome and unnatural. When you are in the middle of a conversation with a real person, you do not have to give these indicators at every turn.

  YOU Long time no see! How are you doing?

  FRIEND I am going to speak now. I’m good, how are you?

  YOU Fine. Where did you go last night?

  FRIEND I will now tell you where I went last night. I went to a meetup about bonobos.

  That’s going to be an awkward conversation. If your user is involved in a conversation with your app, don’t force them to keep indicating they are going to talk. Instead, use natural turn-taking techniques. These include:

  Asking a question •

  • Using eye contact (if you have an avatar or video)

  Pauses • Explicit direction

  • The easiest and most natural technique is asking a question. When the VUI asks the user something, people will naturally respond. Explicit direction is fine as well. For example, “Why don’t you just TELL me the name of the movie you want to see.” Be careful not to force the user into this conversational mode when it’s not appropriate. One virtual assistant I tried would turn the mic back on every time it finished speaking, but it was very confusing.

  VIRTUAL ASSISTANT I’d be happy to look up that info for you. All you have to do is ask. [turns on mic and beeps]

  ME (thinking…. Isn’t that what I just did?)

  Human turn-taking is not always clear-cut. In many cases, a turn may be someone simply making an “mm hmm” noise while the other person continues to talk. As Urban says, a lot of human-human conversations have overlapping turns; when I say ‘mm hmm’ it’s not a signal I want the other person to stop talking and give me the conch. Instead, it’s a “liveness” indicator— I’m still here, I’m still listening.

  Computers are not yet able to handle this more subtle form of turn-taking. In systems that don’t have barge-in, however, this can be workable, since the “mm hmm” will not trigger the recognizer while the system is still talking. With more careful design, a VUI system can handle some of the more common forms of these subtleties—for example, users may say “thank you” at the end of a transaction—you can program the system to either ignore this, or acknowledge with a “you’re welcome” rather than generate an error/reprompt.

  In addition, don’t ask rhetorical questions you can’t understand the answer to. Ben Brown, co- founder of “Howdy”, a chatbot that runs within the office communication tool Slack, had to outlaw the bot from asking rhetorical questions "because people expect to respond to them, even though the bot was just being polite," he says. "You would never just stick a form on your web page and not expect people to type into it." [Fastco design,

  next-phase-of-ux-designing-chatbot-personalities , John Pavlus 1/16]

  Another way you can violate turn-taking is by putting the question before the system has finished speaking. For example, a common IVR structure is “Would you like to hear that again? You can say yes, no, or repeat.” Users often start to speak as soon as the question has finished, which leads to frustration because either they can’t barge-in, or, they barge in just as the system starts the next sentence, stop talking, and have interrupted the flow. With good prompt design and very careful voice coaching it is possible to make this work, but in general it should be avoided by putting the instruction first, and the question at the end. Sometimes, it makes sense to switch between command-and-control and conversational modes, though it’s important to do this only where users will implicitly understand the mode has changed. A good example of this is the Jeopardy! game on the Amazon Echo. To initiate the Jeopardy! game, you must say “Alexa, start Jeopardy”. Once Alexa says the quiz answer, however, you do not need to say “Alexa” again—you may simply say the question, e.g. “Who is Dorothy Hamil?” and Alexa will tell you if you’re wrong or right. There is no “beep” to let you know you can speak, but there is no need for one—it’s a known conversational structure, so users do not have trouble figuring it out.


  Another way to make your VUI more human-like (as well as more engaging) is to use conversational markers. Spot the difference between these conversations:

  VIRTUAL ASSISTANT How many hours of sleep did you get last night?

  USER About 7.

  VIRTUAL ASSISTANT How many servings of fruits and vegetables did you eat yesterday?

  USER Maybe four.

  VIRTUAL ASSISTANT Did you take your medication last night?


  VIRTUAL ASSISTANT I’ll be asking you a few questions about your health. First, how many hours of sleep did you get last night?

  USER About 7.

  VIRTUAL ASSISTANT. Good job. And how many servings of fruits and vegetables did you eat yesterday?

  USER Maybe four.

  VIRTUAL ASSISTANT Got it. Last question—were you able to take your medication last night?

  USER Yes.

  VIRTUAL ASSISTANT All right. That’s it for now. I’ll talk to you again tomorrow. Goodbye.

  Both of these examples collected the same three pieces of information. But which one do you think the user will prefer speaking with, day after day? Conversational markers are an important way to let the user know where they’re at in the conversation, and that they’re being understood. Users will also be more engaged when the system is using basic manners and will respond in kind. They are often the “glue” that keeps the various required pieces of the interaction together. Conversational markers include:

  • Timelines (“First”, “Halfway there”, “Finally”)
  • Acknowledgements (“Thanks”, “Got it”, “All right”, “Sorry about that.”)
  • Positive feedback (“Good job”, “Nice to hear that”) One practical way to see where conversational markers should be added to your dialog is to do a “table read” with someone. Write down a sample dialog (the back and forth between the system and the user) and each read one part. You’ll quickly see where things might be stilted or unnatural, or add to a user’s frustration because they have no idea how long the conversation is going to last. One common concern I have heard from clients is that “computers don’t speak that way” and “people will be put off because it’s a computer, not a person”. Some clients worry the system will sound too informal, or annoy people by pretending to be human. It is important to use conversational markers that are appropriate to your system’s persona, but
Next we’ll discuss a crucial piece of design, that you hope never happens, but is essential nonetheless.


  “When you talk to a human being, there is never an unrecoverable error state..” - Abi Jones, Design Lead at Google

  We’ve talked a lot about “best path” behavior for a VUI design. But as any good designer knows, you can’t just design for when things work—you have to design for when things go wrong, as well. This is especially true with VUIs, because something will always go wrong. Although speech recognition has improved drastically in the last 10 years (over 90% accuracy given the right conditions), this will in no way ensure your users will have a good experience when you add voice to your design. Think for a moment about human-to-human conversation. We often miss a word (or multiple words) when someone else is talking. Perhaps their voice was muffled. Perhaps you coughed. Perhaps they used a word or the name of a place you’re not familiar with. And those are just things related to hearing the words they used. There are plenty of ways to have a misunderstanding at the interpretation as well. You might think the speaker is being sarcastic when they are not. You might be worried the person is angry with you, which will skew your attention. You think the speaker is talking about Springfield, Illinois when in fact they are referring to Springfield, Pennsylvania. All of these factors play a role in VUI as well. But humans are much better than computers about getting back on track. If I say something to you, and you stare at me quizzically, I’m likely to repeat myself, knowing you did not understand. I can ask you to repeat yourself. I can ask you to clarify. We have many ways to course-correct, because we’re both steeped in years of practice of human conversation. When the VUI does not understand us, things often break down. How you decide to handle these error conditions in your VUI is extremely important. As Pilar Manchon, GM of Voice and Digital Assistants at Intel said, “Every time that you score down because you make a mistake, or you don’t know something, it actually counts a hundred times for every time that you had it right.” [Virtual Assistant Summit, SF, Jan 2016] If you do a good job, error conditions won’t derail the user, and you can get them back on track and have them successfully complete a task. Handle it poorly, and not only will they fail to complete their task, they’re not likely to come back. There are a variety of ways VUIs can make mistakes:

  No speech detected •

  • Speech detected, but nothing recognized

  Something was recognized correctly, but the system does the wrong thing with it • Something was recognized incorrectly •

  No Speech Detected

  Speech recognition engines have a “listening” mode where they attempt to capture a speech signal. If speech is expected but not detected, ASRs (Automated Speech Recognizers) return a “no speech” response. It’s important to keep in mind that this does not necessarily mean the user didn’t say anything. It

  There are two ways to handle the case of “no speech”:

  • Call it out explicitly: Example: “I’m sorry, I didn’t hear anything. What is your account number?” Do nothing. •

  Which one should you use? It depends on your app. For the explicit case, the following should be true: Your system is audio only (such as an IVR). • There is no other way for the user to respond (such as touching buttons on their mobile

  • phone) The system needs a response from the user before it can continue the task/conversation. •

  There are also cases where it’s appropriate to do nothing: The user can move forward another way (such as choosing a response via a button) • The consequence of not acting does not break the conversation

  • There is a visual indicator that the system did not understand, such as an avatar
  • continuing to engage in active listening by looking at the user

  Why not just err on the side of caution, and always prompt the user to speak again? Because it becomes very annoying. Humans have a variety of ways to indicate they did not understand what was said, and one of the most common (and effective) is to say nothing. Instead, you might look quizzically at the person, or smile politely—but this action makes it clear very quickly that the person was either not heard or not understood.

  VUI designers should take advantage of conversational rules that humans are already comfortable with. Rather than continually telling the user they were not understood (and having them lose faith in the system), subtle cues can be just as effective. When conducting user tests at Volio, I have observed cases where this was so well done, users did not even remember afterwards an error had occurred, even when asked. In this case, when the user was not understood, the video showed the actor simply continuing to listen—nothing else occurred. The users naturally repeated themselves, and the system moved on.

  Speech Detected, But Nothing Recognized

  In some cases, the ASR did detect an audio signal, but it was unable to come up with any reliable hypotheses. The strategies for dealing with this case are very similar to the case of “no speech detected:

  • Call it out explicitly: Examples: o

  “I’m sorry, I didn’t get that. What was the name of the restaurant?” o “Sorry, what was that?” Do nothing.

  • Some systems attempt to be clever, or funny. If Alexa knows you were asking a question, but doesn’t have a way to answer it, she replies “I’m sorry, I didn’t understand the question I heard.” Siri, in some instances in which the system does not understand (for example “Siri, tell me the meaning of love,”) it will reply “Who, me?” Be careful with these types of responses. “Who, me?” can be cute the first couple of times, but
Recognized, But Not Handled

  In this case, the ASR did return a perfectly accurate recognition result, but for whatever reason, your VUI did not handle it properly. This could be because you did not program in a response. It could also be that you programmed the wrong response. Examples:

  MEDICAL VIRTUAL ASSISTANT How are you feeling?

  USER I have a cold. MED VA

  How long have you been feeling cold? In this example, the system decided that the key word “cold” meant temperature, and now the conversation has been derailed. A smarter system would look for the concept of “having a cold” vs “being cold”. Another example is when the system recognized it perfectly, but has no programmed response:

  MED VA How are you feeling?

  USER Uh, my arm sort of hurts.. MED VA

  I’m sorry, I didn’t get that. How did you say you’re feeling? This is a case in which whoever designed the types of responses the system could handle did not include one for “arm pain”.

  The strategy for handling these? Better anticipation of all the things a user might say. For tips on how to avoid this issue, see chapter 5, on data collection.

  Recognized, But Incorrectly

  This is the case where the ASR returned the wrong recognition result. This could produce two outcomes: either you don’t do anything with it, because you don’t expect it, or, it incorrectly matches to the wrong behavior.

  MEETING SCHEDULING VA What time would you like to schedule your meeting with Sharon?

  USER Umm… well, I guess I’d like to sort of um I will schedule it for three thirty.

  SCHEDULING VA [ASR returned ‘um well I guess I’d like to sort of um I will sledge it throw forty’] I’m sorry, what time did you say?

  Unfortunately, there is not a lot you can do about the ASR not recognizing things correctly; of real user responses. (An “N-best list” is the list of the top possible recognition results returned by the ASR—for more about N-best lists, see Chapter 4.)

  Escalating Error

  A common strategy for cases when speech is expected (and is the primary mode for communicating with the app) is to use escalating error strategies.

  WEATHER APP I can get the weather for you. What’s the city and state?

  USER Uhhh… it’s Springfield. MOVIE VA

  I’m sorry, I didn’t get that. Please say the city and state.

  USER Oh, it’s Springfield, Minnesota. Escalating error behavior prompts can become more detailed if needed, offering more help, such as where to find an account number. In addition, if after several turns speech has continued to fail, offer another mode of communication, such as pressing buttons or using a drop-down list of options. If you are working on a system in which there is a live person who can provide assistance, set a threshold for number of errors, and transfer the user.

  Don’t Blame the User

  When at all possible, do not blame the user. Blame it on something else, or at the very least blame the system. In a study performed by Clifford Nass & Scott Brave, users performed tasks with a driving simulator, and throughout the tasks, the system voice made comments. Half of the participants heard messages blaming the driver (“You’re driving too fast.”) and half heard messages blaming external factors, such as “The road is easy to handle at slow speeds.” [Nass & Brave, Wired for Speech, p 176] The participants who were blamed by the system rated their driving performance lower, liked the voice less, and most importantly, showed less attention while driving. Error messages may seem like a small thing, but they can impact the user’s perception of the system and even their performance.


  If your users will be using your system on a regular basis, it’s important to include different strategies in the design. For example, a healthcare app may require users to log in every day and take their blood pressure. In the beginning, having the prompts include more details on instructions is useful and necessary. However, once the user has gotten the hang of the app, there is no need to continue lengthy instructions (as well as other prompts).

  AVATAR Let’s take your blood pressure. Please make sure the cuff is turned on. Wrap the cuff around your arm so that the blue arrow is pointing towards your palm. Be sure to sit down and have your feet flat on the floor. When you’re ready, press continue.

  Used the app every day for a week: AVATAR Time to take your blood pressure.

  Please put the cuff on, and press continue. Don’t just count the number of times the app has been used, however. It may be that they’ve used it many times, but only once every month or two. In this case, the novice prompts would still be used. Explanatory prompts can also be shortened. For example, the system may say “I’ll be asking you a series of questions about how you’re feeling. It will take a few minutes, but the questions will help the doctor determine the best course of action. First, did you take your medication yesterday?” After doing this every other day for a week, it can be shortened to “Did you take your medication yesterday?” (Be sure to use “conversational markers” even though the prompts are shorter, to keep the user informed that their answers have been understood, and where they are in the set of questions.) Margaret Urban, Interaction Designer at Google, says it’s important to make sure your goal is not simply to “train” your user. She says rather than beat them over the head with the available commands, adapt to their behavior.


  Remember the example at the top, in which Google was able to continue the conversation about Abraham Lincoln?

  USER Ok Google. Who was the 16


  President of the United States? GOOGLE

  Abraham Lincoln was the 16

  th President of the United States.

  USER How old was he when he died?

  GOOGLE Abraham Lincoln died at the age of 56.

  USER Where was he born? Hodegnville, KY USER

  What is the best restaurant there? GOOGLE

  Here is Paula’s Hot Biscuit What’s good about this example? Google continued the conversation, and remembered the context. It knew “he” referred to Abraham Lincoln. It knew “there” meant Hodgenville, Kentucky. Keeping track of this information is not always easy, but without it, your app won’t be capable of anything but one-turn actions.

  Here’s another example, though fictitious, about an app that can help find movies to watch on TV: USER

  Show me movies with Harrison Ford TV

  [shows list of movies starring Harrison Ford]

  USER Which ones were made before 1990?

  TV [shows new list]

  USER Now show me movies with him and Carrie Fisher.

  TV [shows list] In this case, the app needs to know what “ones” means, as well as “him”.

  It can be difficult to anticipate all the things your system may need to track, but you can start with the basics and build up based on what your users are actually saying (see more in chapter 5). If the user asks a question about a person, store that information. If the person is famous, you can even look up the gender. However, you can also use a simpler method, which is to always store the last person mentioned, and refer to that person whether the user says “he” or “she”. (This won’t always work of course, if the user has mentioned more than one person for example, but it will work in many instances.) If you can’t find out the gender, you can also just use the person’s name, although it can sound robotic to keep using the full name rather than “he” or “she”. You can apply similar processes to remembering the last city or place the user (or the system) mentioned. Systems often struggle with interpreting the user’s references. Here’s a Cortana example that isn’t able to handle the second query in the conversation:

  USER What’s the most expensive dog?


  USER Where can I get one?

  CORTANA As you can see, Cortana did not assign a meaning to “one” and treated the query like a brand- new one. To contrast this, look at what Google does:

  USER What’s the most expensive dog?

  GOOGLE According to PopSugar, the most expensive dog in the global world today is the Tibetan Mastiff.

  USER Where can I get one?

  GOOGLE Here, Google successfully understood the word “one” in the second query, and gave some relevant search results. Two notes on the Google example: when I started writing this chapter, a couple of months ago, Google actually handled this query differently. It replaced “one” with the word “dog” (it actually modifies the search query itself, so you can see it). This was more intelligent than Cortana’s response, but not very useful all the same. It used my original query to replace “one,” where it should have used the answer it gave (as it does now). The example shows, however, how fast these systems are improving. The other note is about the content Google serves up. The description contains a swear word, which some users may find offensive. As a VUI designer, these are the things you still need to be aware of, even though technically it’s out of your domain. Google does have the option at the bottom to give them feedback on what you think of the result.


  When I worked on IVRs, we made sure to include a set of universals at every state: repeat, main menu, help, operator, and goodbye. For mobile apps, this is not a hard and fast rule. Many mobile apps (or connected devices) do not have the concept of a main menu. However, it is still important to make sure users can get help when needed. Supporting the “help” command is useful in many cases, but is traditionally used for context- specific help. For example, imagine an insurance IVR in which a user has been asked to enter their medical record number (MRN). If the user says “help: at this point, it would be useful to

  But what if the user asks for help in an open-ended situation such as a general assistant, like Google, Cortana, or Siri? If the user has just initiated the conversation (by pressing the home button or saying “Ok Google: etc, and then says “help”), you have no context around which to know what type of help the user needs.

  In addition, users are not always accustomed to using these types of universal phrases, despite us IVR designers trying to train them for many years! Therefore, it’s important to think not only about what type of help to provide, but how users might say it. For example, “Alexa, what can you do?” or asking Cortana “What can you do for me?”, or “Ok Google, what can I say to you?” Alexa doesn’t even try to handle that question on her own, but says ““For help with what you can say, take a look at the things to try section in the Alexa app.” Cortana says “Here are some things I can help you do” and provides a visual set of examples: Google says nothing, but provides a screen of examples:

  In addition to allowing users to ask for help, take advantage of visual space if you have it, such as on a mobile app. Imagine designing an adventure game in which the player can talk to various characters and explore different worlds. Having a “help” or “info” button always available in the GUI commands is an obvious way to let the user know assistance is available.

  Another important piece of good IVR design was to include a “goodbye” universal. In the early

  VUI designs, goodbye was not considered important because people could simply hang up to end the conversation. But years of collecting data made designers realize that, because people were used to saying goodbye to people when ending a phone conversation, they often did it with IVR systems as well, and sometimes felt uncomfortable ending conversations by simply hanging up. Therefore, we added dialog to handle this case. Another important lesson was to make sure the system was VERY confident before actually hanging up! If your system will be using a “goodbye” option, be sure to use 3-tiered confidence and confirm that the user really wants to exit the system if confidence is not high.

  An example from the 511 IVR system, when confidence was medium: “I thought I heard you said goodbye. Do you really want to exit the system?” The other night on my way to bed, I passed by my Amazon Echo and said “Alexa, goodnight.” She responded with “Good night.” It may sound like a silly thing to invest development time in, but it was very satisfying. Allowing users to repeat and to go back a step are important for conversational systems as well. task-oriented conversation, go back is very useful. These are commands that are often good candidates for GUI controls, as well.


  Another component that is sometimes missed by designers is latency. It is important to determine as early as possible whether or not your system will have any latency, or delays. Latencies are generally caused by:

  Poor connectivity • System processing • Database access

  • Perhaps your VUI system needs to access patient records, and you know that this will require a database lookup. Find out as early as possible how long this is likely to be, and plan accordingly. If there is a known latency, make sure the system has a way to handle this. This can be done both by the system telling the user about the delay (“One moment while I look up your record…”) as well as non-verbal and visual cues, such as a latency noise (often used on IVRs) and visual (animated loading icon) (figure 12).

  Figure 12 - Sensely avatar says "One moment please", displays the message, and shows the loading GIF There may be cases where the expected latency time could range from 0 seconds to 10 seconds.

  If so, pad the latency by a few seconds in the case of no latency, because it sounds broken to the


  There may be times when the user provides some, but not all, the details to take an action. For example, the user may ask for the weather for a location that exists in more than one place: “What’s the weather in Springfield?” If possible, rely on any known information to determine the answer without having to ask the user.

  For example, the Amazon Echo requires the user to set up their location, so asking “What’s the weather?” will produce local information automatically. This can be used when the user asks for somewhere else, as well—choose the “Springfield” close to home rather than across the country. Other contextual clues can also be used. If the user just looked up a restaurant in Springfield, Illinois, and then asks “What’s the weather in Springfield?” you can bet they mean the one they just referenced.

  If no contextual information is available, the system will need to ask the user to clarify.

  USER What’s the weather in Springfield?

  SYSTEM Did you mean the one in Illinois, or Maryland?

  USER Illinois. SYSTEM

  It’s 65 degrees… If the system has high confidence for the word “Springfield,” it can use the reference word “it” rather than explicitly stating it again. Also be sure to allow the user flexibility in their response: the user should be able to say “Springfield, Illinois” OR just “Illinois” or even “the first one” (imagine a list).

  Another example where disambiguation is required is when an action is not clear: USER Call Cindy, please.

  SYSTEM Ok. Cell phone, or home phone?

  USER Cell phone. SYSTEM

  Calling Cindy, cell phone… Note the confirmation of the name came at the end, as an implicit conversation. This would be done if the system had (a) high confidence about the name and (b) there was only one Cindy in the caller’s contact list. I noticed Google recently improved their dialing design. It used to be that when I said “Text Chris Leggetter”, Google responded with “Home or mobile?” and I had to respond. Now it’s smart enough to know I meant mobile, because I can’t text a home phone number.

  Disambiguation may also be required when the user answers with more information that your VUI can handle.

  SYSTEM What is your main symptom?

  USER I’m throwing up and I have a fever. SYSTEM

  Ok. Which one would you say is your primary symptom? USER Uh, my fever I guess. SYSTEM

  Ok, fever… It would be ideal if your system could handle both symptoms at the same time, but given that systems do have underlying constraints, sometimes it is necessary to have the user narrow it down.


  In addition to the previously mentioned sample dialogs and dialog flow document, there are other tangible things that you may need to think about.


  It may be necessary during the design to create lists of prompts. A “prompt” is what the system can say to the user. It may be a complete sentence or sentences (“I can help you with that. Please tell me the make and model of your car.”) or may be smaller snippets, such as numbers, dates, or products.

  Prompt lists serve multiple purposes: A list for voice talent to record • Getting sign off from the client • Input to the TTS engine •

  For a great resource on what prompt lists should look like, and how to create prompt lists for concatenation, see the book Voice User Interface Design.

  Grammars / Key phrases

  In the early IVR days, complete grammars had to be specified for every state in the dialog. For example, if the prompt asked the user “Do you want me to book the flight?” the grammar might look like: Yes: { “yep”, “yeah”, “uh huh”, “sure”, “yes please”, “yes book the flight”} etc. [flesh this out] In addition, it needs filler words, such as “um” and “uh” and pleasantries such as “please” and “thanks”.

  Because of the improvements of speech recognition technology, this is thankfully no longer the case. Instead, a lot of systems can get by be specifying key phrases instead of exact sentences, or using machine learning to map the user’s intent. I’ll get into more details on how to interpret natural language in Chapter 5.


  This chapter introduced many of the key concepts when designing voice user interfaces. Many of them are drawn from the IVR world, with key differences. Basic strategies, such as well-designed error behavior, implicit vs. explicit confirmation, and design artifacts such as sample dialogs and flows, apply in both cases.

  Designing for mobile can be both a richer experience, as well as more complex. You’ll have to make decisions about how to let your users know when and where they can speak, and when and where to use visual feedback. There will be no human, in many cases, to back up your experience.

  Enabling users to speak to their phones and devices opens up a whole world of experiences, whether it’s looking up a piece of trivia during a dinner argument, asking a device to dim the lights, or managing the everyday tasks of your life, VUI can enhance them all.

  O’Reilly Media, Inc.


Chapter 1 Technology and Happiness It’s 2025. You are living in a house. Yes, that house. The pristine, white and glass, ultra-


modern, sustainable, smart home of the future. Objects are minimally designed with a

layer of intelligence embedded deep within, or maybe with a projected layer of

augmented reality. As you wander from one behaviorally and emotionally optimized

choice to the next, occasionally enlisting the help of a chatbot here or a robot companion

there, you never wonder if you are happy. Your chip implant monitors your brain

chemistry, in tandem with a mood bot to give you a happiness boost when needed. You

know, quantifiably, that you are happy.

  In the future, we are all better people. We are happy. And all our technology is positive. Well, maybe it doesn’t look quite like this example above though. What does it mean to design for happiness? We design technologies to make life easier, to make tiresome chores go faster, to add value to people’s lives. We strive for seamless interactions. We obsess over the details.

  We believe we are making the world a better place. Chances are if you are working in technology, you are something of an optimist. Now more than ever before, we want to do good in the world. More than just a motto, we truly want to do no evil. Rightly so. Think about how much of our lives are mediated or augmented by something digital. From the basics of getting from point A to point B or asking your partner to pick up some milk at the store to giving long-ago friendships new relevance and teaching your kids to code, technology is irrevocably mixed up with our everyday existence.

  And yet, we struggle with being human in the digital age. We sleep poorly, we have trouble paying attention, we feel isolated despite being more connected than ever. How can we have such positive intentions and yet see negative outcomes? Why isn’t all this new digital technology making us any happier? Maybe we just need to take a break from technology. We can even design technologies that block our access to technology temporarily. But, the fact that we need to detox from our technology is more than a little depressing. It implies that technology, whether how it’s designed or how we use it or both, is toxic. It presumes that there is a real life out there, separate from our digital life. So, maybe we need to design technology to get out of the way. Not a bad idea. Considering that many of our apps seem to be designed to demand our attention, this could be a move in the right direction. Much better than staring into the phones in our hands for hours at a time.

  O’Reilly Media, Inc.

  9/22/16 Or maybe we can humanize technology. Actually, we are quite good at that already.

  Creating a personality for a website, or a chatbot, or a smartphone, is something we already do. Even if we weren’t designing personalities for our technology, we know people attach to technology anyway. What we are still figuring out is how to use technology to humanize humans.

  The truth is we have never designed technology with well-being, broadly speaking, in mind. We’ve designed for ease, productivity, engagement, and delight. Each has proven inadequate to really fostering well-being. What if we re-frame how we design technology to intentionally focus on happiness with a capital H? Rather than continuing to let our lives be shaped by technology—which will happen anyway—let’s reverse it. Let’s think about the lives we want to lead and develop technology for that. In the past decade, happiness has emerged as a subject of serious scholarship and experimental study. By investigating the everyday behaviors and decisions people make, psychologists and behavioral economists have learned a lot about happiness.

  So far, that knowledge has been applied to urban design, to personal spending, to national economies, but it hasn’t been applied to technology. That’s what this book is about. By paying more attention to the growing body of knowledge about happiness, we can develop technologies that cultivate well-being.

  To understand where we need to go, first let’s look back a bit.

  A Very Brief History of Technology and Happiness

People will live in houses so automatic that push-buttons will be replaced by fingertip

and even voice controls. Some people today can push a button to close a window –

another to start coffee in the kitchen. Tomorrow such chores will be done by the warmth

of your fingertip, as elevators are summoned now in some of the newest office buildings –

or by a mere whisper in the intercom phone.

  This description comes from a Popular Mechanics article circa 1950, but it could just as easily apply (except for the intercom reference) to current day. Even though we might not mention happiness directly, it’s implied. The message is simple. If we can create the right technology and apply it in the right way, we can achieve a happier existence.

  You might say, technology has really always been about happiness. Technology has promised us happiness of one kind or another since the mid twentieth century. What does happiness mean when it comes to technology?

  Automatic for the People

  A good place to start this brief and idiosyncratic history of technology and happiness is in the 1950s. Post World War II, more homes than ever had electricity. Some people had a bit more money to spend. And new technology was focused on making mundane household tasks easier.

  O’Reilly Media, Inc.



The Future of Happy Homes

Figure of the Miracle Kitchen and IoT Ktichen

  Push-Button Manor and a bit later, RCA/Whirlpool’s Miracle Kitchen and the Monsanto/Disney Home of the Future envisioned a home where “the things that women don’t like to do are done automatically.” Okay, I cringe a little at that. We’ve branched out a bit, and now we ‘d say that technology should do the things many people don’t really like to do.

  At the heart of this idea is automation. Automation had already been making markets more efficient and productive. Farming technology made it possible for a small number of farmers to produce food for a lot of people. Automated assembly lines meant that fewer workers could create more consumer goods of higher quality. Making it personal seemed like the next step.


The idea behind automation, whether for business or no, is the same—eliminating

mundane, routinized, repetitive tasks. Smart home products, for instance, take care of

  things we might forget, like turning lights off and thermostats down when we leave. Your Roomba takes care of the vacuuming so you can live your life. Although we worry about what we’ll do when robots take on all the work, part of the thinking behind it is that it will bump up the humanity of the humans. We will be able to engage in more meaningful activities that require empathy, creativity, and critical thinking.

  Tech happiness is automating the mundane to maximize the meaningful.

  The Easy Button

  If we skip ahead to the age of the personal computer in the 1980s, technology’s promise of happiness shifts toward productivity. It’s not exactly automating, but instead making common tasks achievable with less effort. Communicating with colleagues, creating and sharing documents, and many other work-related chores that were time-consuming were streamlined. Like automation, the business goal is the same—productivity and efficiency. On a personal level, it’s competence. Pulling a reference book off a shelf seems misguided when it’s easier to search for just about anything online. On-demand services like Uber and Blue Apron continue to push forward this idea of the easy life. Convenient and frictionless, these services save people time. Ease, as a guiding principle, applies to technology itself. Just as we make interactions with the world around us easier, technology itself is scrutinized and perfected until it becomes as easy as possible to use. Usability is foundational to good design.

  Tech happiness is reducing friction between ourselves and the world around us.

  Superpowers for Everyone

  Another happiness promise that has dominated the last decade of technology is supplementing human capability using technology. Technology has always been about what it enables people to achieve and that’s become ever more personal. Beyond

  O’Reilly Media, Inc.

  9/22/16 automating mundane tasks and making interactions external to ourselves easy, technology aims for our bodies and minds. The smartphone supplements our intelligence through omnipresent Internet access and to supplement our memory with photos or social networks. Wearables already offer improved health, from monitoring chronic conditions to pushing ourselves to get fit. Smart environments promise to boost up abilities for self-care for people who may have once needed a caretaker. Technology certainly amplifies our abilities. Almost as a byproduct it extends our identity. Sure, the technology we buy and how we present it—whether it’s a macbook covered in stickers or an old-school flip phone— says something about us to the world. But it runs deeper than that. What we post, who we follow, how we comment becomes a marker for who we are. We can proactively create avatars or try out different personas. We also spawn multiple identities online inadvertently. Technology extends our notion of self.

  Tech happiness is helping people be better at what they do, and maybe at who they are.


The Extended Self

The concept of an extended self has been around for a while. We know that the

self is not fixed. That’s what makes us human in a way. We evolve. And a lot of

that evolving happens because of the people, the pets, and possessions we bring

into our lives.


More than 100 years ago, William James pointed out that people consider their

possessions to be a part of their self. Mihalyi Csikszentmihalyi (in his pre-flow

work) describes how we attach to products as we create meaning with them.

Sherry Turkle objects are perceived as evocative as they become a companion to

our experience. Consumer behavior researchers, like Russell Belk, describe this

attachment as self-extension. The material things we bring into our lives become

a part of ourselves. And our favorite things connect, remind, and reveal

ourselves to other people in our lives.

So, let’s think about music. Imagine it’s back in the 90s, and we are hanging out

at your house. I’m wearing a thrift shop shift dress and purple Converse high

tops, by the way. I’d definitely scan your CD collection. Not only would this be

a conversation starter, but it would also tell me something about you. In fact,

certain CDs might be core to your identity. After all, you probably chose to

display Nirvana and Beastie Boys and to hide Vanilla Ice.


Today, maybe I’d look at a playlist instead but the idea remains. The music acts

as a cue for me to form an impression about who you are. More than that, it is a way for you to define who you are in the moment, and as you go along.

In some measure, we are what we own. Perhaps this is why Marie Kondo’s The

Life Changing Magic of Tidying Up (2014) hit a nerve. As Kondo succinctly

notes, “The question of what you want to own is actually the question of how

you want to live your life.” It’s not as shallow as acquiring status symbols. Instead it helps us figure out where we fit in and who we want to be. O’Reilly Media, Inc.



Beyond the purchase of the latest iThing, where does technology fit into this

idea of extended self? This idea of extended self seems like it’s about stuff.

Technology is not so much about the object itself, right?

Well, yes and no. We become incredibly attached to our smartphones even if we

don’t think about it as an extension of self (yet). When studies report teens

having anxiety about being separated from their phones or that we all touch our

phones hundreds of times a day, it seems like we are addicts. When fMRI

studies reveal that our phones light up the same centers in our brain as love, this

seems a little foolish. The idea that our phone is an embodiment of who we are

doesn’t come into the discussion. It’s viewed as something outside. That’s how

we use it and that’s how we design it.

In one of my own research studies, listening to over 100 people give me a tour

of the phone, each little icon opens up a big world. As Jamal put it, “my phone is

really about who I am in this moment, and who I was, and who I want to be, it’s

not really about watching videos and looking up facts.” It’s not just re-

embodiment as an avatar, or a virtual possession in a game of Farmville. It’s an

extension of who we are. I wonder if it isn’t a mistake to consider technology

only as something outside ourselves at all.

In Smarter Than You Think (2013), Clive Thompson extends (sorry) the

extended self concept to technology. Technology amplifies our intelligence.

When human cognition is supplemented by machine intelligence, we extend our

abilities. His example is Garry Kasparov. When playing chess against Deep

Blue in 1997, he lost. When augmented by artificial intelligence, Kasparov was

freed up to play a more creative game. We can think about it more in the context

of our daily lives. No longer do we have to learn or remember say historical

facts, we can just ask Google. His point is not so much about identity as about

capability. As we move to wearables and IoT, ambient intelligence is not other.

It’s part of us.

Alex Soojung-Pang, in the Distraction Addiction (2013), encourages readers to

approach digital life as an opportunity for extending our mind. Using technology

with intent amplifies our humanity, while using it to fill in spare moments often

does not. Maybe this doesn’t seem like a big deal. But think of the advice we get

for living with balance in a digital world. Usually it means a separation.

Whether a detox, or an unplug day, or apps that time us out of our tech, the

current approach looks at tech primarily as something outside ourselves.


A fundamental tension between technology of any kind—whether a fitness

tracker, or an app, or an AI assistant—is this line we tread between technology

as me and not me. Is it my sassy sidekick or my deepest self? Is it a means to an

end that exists somewhere offline (whatever that is) or is it a mode of self-

reflection? It’s both. This is fundamental when we design for big picture well-


  Technology as Connective Tissue

  In the aughts, technology really started to underpin our relationships. Emailing was already part of everyday work life. Now texting becomes part of a day-to-day social interactions with friends and family. Social media brought people back into our everyday circles. Our Facebook conversations are a jumping off point for real conversations. Sharing small moments or big accomplishments adds a layer of nuance and depth to

  O’Reilly Media, Inc.

  9/22/16 personal relationships. So much so that tt’s hard to imagine relationships that aren’t supported by technology. At the same time, weak ties became more important. While research shows we only have a certain number of true friends (which I’ll interpret as close ties), we can and do bring people into our lives on a broader and perhaps not quite as close basis. Acquaintances we’ve met once or twice, people whose opinions we value, people who live nearby, groups of people who share our interests—all consequential strangers. These weak ties connect us not just to people, but to ideas and the world. Connection is at the heart of well-being, in study after study. Strong relationships contribute to healthy and happy lives. And technology is the glue holding those connections together.


Tech happiness strengthens relationships and foster the kind of weak ties we take for

granted in other contexts.

  The Best Intentions

  While it may not always be an explicit goal, technology promises happiness by making our lives easier, augmenting our abilities, and, more recently, by connecting us to other people. This may seem obvious. It’s why you and I got into this field in the first place. Technology’s big goal is to make the world a better place and a big part of this is making people happy in a deep way.

  Despite our best efforts, technology has not always or even consistently had a positive effect. In academic research, studies abound on the negative effects of technology on ourselves and our relationships. In popular culture, our smartphones have become a nexus of cultural critique. Letting technology into our lives has had unintended negative consequences.

  Addicted and Conflicted

  As much as technology implicitly strives for well-being or actually contributes to well- being, that isn’t always the result. When we aren’t busy tweeting, checking in, emailing, and ruining tons of pictures with crappy filters, we suspect we might be miserable. We are pretty sure technology is to blame.

  Research makes a convincing case. The damage caused to our bodies and minds by our time online is well-documented. While internet addiction is not included in the latest version of the Diagnostic and Statistical Manual of the American Psychiatric Association, study after study finds strong positive correlation between heavy smartphone use and anxiety, depression, and other symptoms of mental illness. Video games may have some cognitive benefits, but they can also increase social anxiety. Social media is chipping away at self-worth, especially for teens. Staring at our smartphones is damaging our spines and disrupting our sleep.

  Here’s just a sample of recent studies: Excessive internet use predicts school burnout among teens

  Social anxiety among internet addicts O’Reilly Media, Inc.


  Cognitive performance when iPhones are out of reach

  • Text neck as a result of chronic slouching over screens
  • Neuroimaging

  suggestive of behavioral addictions

  • Academic studies focus on the negative effects of technology, and that tends to spiral into negative press. in the media, talk about technology can be a lot of gloom and doom. Say goodbye to your job, it’s going to be automated. Selfie obsession is the surest path to narcissism. Smartphones make us behave rudely to our friends. Your attention span…wait what? We don’t need to read about technology’s ill effects though. We feel it. Maybe we are not so great at estimating how many times we check our phones (it’s actually double what we think according to some research). But we know that we find ourselves spending too much time on YouTube and not enough time with our families. As a parent, I’ve had that moment of reckoning, actually many moments of reckoning, where my daughters call me out on my own tech infatuation. That out of control feeling frames our experience of technology. Technology in popular culture is about unhappiness, from everyday distraction to a deep pathological misery. Mostly, when we talk about our life with tech it’s in terms of addiction.

  Internet Addiction? There’s an App for That [Friend pulls out phone] Ugh, I’m so addicted.

  [At dinner with a group] It’s funny how we are all so addicted. [In the middle of a conversation] It’s hard to pull myself away, I feel like I’m addicted. People used to joke, with a hint of pride, about their crackberries. Now we compulsively refresh our email, all the while longing to achieve inbox zero. We mean to spend a few minutes on Facebook and end up spending an hour. We are a little foggy from binge watching the latest Netflix series. Then there’s notifications.

  Addiction is the story we tell ourselves, and take for granted, about technology. As a


researcher, I hear this more often than most people. It’s actually difficult to remember a

research session where I haven’t heard people speak about technology as an unhealthy

obsession. Internet addiction is a real condition, a matter of debate, and widely accepted

universal state of mind.

  But what are we addicted to anyway? Is it the phone itself? Some studies make the case. We get pangs of anxiety when separated from our phones.

  Other studies suggest that we are less focused when our smartphone is nearby. Our brains flood with oxytocin when we hear our phones — we may literally love our iPhones. Even the sound triggers love and loss. The object, security. Or are we addicted to the content, interactions, the stuff of the app? It’s difficult to untangle the device from the app from the content. Entrepreneurs study addictive behavior, after all, to help create it. Product teams work hard to engage us, hijacking our time.

  O’Reilly Media, Inc.

  9/22/16 Social media, in particular, has been declared addictive in studies and by experts. A friend likes your photo, another comments on your post. Maybe we are addicted to the approval we get? Perhaps we are simply addicted to distraction itself. Like gamblers who get in a “machine-zone”, we are caught up in the rhythm of a repeating process. An alternate reality flow state.


Addiction might be a little strong, but it captures the gravitational pull we feel toward our

tech and away from “real life”. Happiness research shows that the more we stray from the

moment, the less happy we are. It doesn’t seem to matter much whether the task is

enjoyable or not, just that we focus. Our experience of technology seems to take away

that focus, or just shift it toward the wrong thing.


Addiction, especially feeling out of control, is part of the inner narrative we whisper to

ourselves and shout to each other about technology.


Internet Addiction

Internet addiction disorder (IAD) goes by many names. Problematic internet use

(PIU), compulsive internet use (CIU), problematic and even pathological

internet use. Internet addiction is used loosely in pop culture to describe the

negative feelings we have around technology, especially feeling out of control.

After all, addiction is a compulsive behavior that interferes with relationships,

work, and health. Compulsive behavior seems an integral part of our experience

with the internet.

At the same time, internet addiction is being studied as a real condition. Internet

addiction is considered an impulse control disorder, characterized by many

hours spent in non-work technology-related activities. While Internet Addiction

Disorder is not officially recognized in the American Psychiatric Association’s

Diagnostic and Statistical Manual of Mental Disorder V (DSM-V), other

behavioral addictions are included. Gambling addiction, for example, is a recent

addition. Internet Gaming Disorder is flagged for further study.

It takes several decades of extensive research to make the cut. So far, there isn't

enough systematic, longitudinal data about internet addiction. Scientists are

actively studying the negative effects of internet use, though, particularly

adolescents who seem to be most vulnerable.


Pinning down a quantifiable, negative effect of internet use is tricky. With

gambling, you’re losing money and causing harm to yourself and loved ones.

With internet use, the negative effects are less direct. Whether internet addiction

can be teased out from other behavioral addictions is a matter of debate. It may

be that gamers game, shoppers shop, and gamblers gamble no matter the

medium. Preoccupation with the internet, inability to control the amount of time

spent online, withdrawal symptoms when not engaged seem like other behavioral addictions but could be a symptom of depression or anxiety, too. O’Reilly Media, Inc.



Substance and behavioral addictions do seem to trigger the same mesolimbic

dopamine pathway in the brain. That means that behaviors like gambling

activate the brain’s reward circuitry. While there hasn’t been much study of how

internet use affects brain chemistry, behavioral design references the language

of trigger and reward. Designers and developers are tasked with attracting and

keeping as many customers as possible, often leveraging psychology, and

derivatives of Skinner’s methods in particular, and neuroscience.

Cause or symptom? Internet addiction may be classified as a condition or not.

What we know for sure is that we feel an unhealthy pull toward technology. We

often use technology in ways that don’t seem to make us happy. Addiction

has become part of the story we tell ourselves about our relationship with



To me, this seems like an opportunity to turn things around. Despite the rapid

change, we are still in the early years of the internet. There’s still time to change

the story.

  The Eat, Pray, Love Approach to Technology

  When it comes to technology, we greedily feast on the unlimited variety that technology brings. We watch just as much TV as ever, but now we are online for hours each day too. If VR becomes mainstream, there’s no reason to expect that we won’t just add on more hours to our tech consumption. We can’t seem to stop ourselves.

  Then we feel guilty. It’s hard to go through a day without a nagging feeling that we are spending too much time online. We are reminded that rather than liberating us, maybe it’s holding us back. So the advice we hear is to take a break from technology. It seems to make sense on a visceral level. If you want to find the happiest people, that aren’t all from Scandinavia. You can look at the Amish. Depression rates are low, happiness ratings are high. Not


much technology. The comparison is a little simplistic, of course. Even so, the imagined

simplicity of a life without technology is appealing.

  Look to psychology and you find similar answers. The primary way to deal with a behavioral addiction is to remove the triggers. Cancel the credit cards, avoid casinos, steer clear of people who engage in these behaviors with you. So what does this mean when it comes to technology? Shut down your computer? Stay away from your phone? You can’t disconnect and function in the world. Work, classes, parenting, our day-to-day lives all rely on technology. Working in tech, I’ve found it especially difficult to stay away from tech.

  We could use technology to minimize all the technology, calming the symptoms of heavy internet use with more internet use. Apps like Cold Turkey can help us go cold turkey, at least for twenty minutes. Products like Vinaya’s Altruis manage notifications for the fashionable, and affluent, few.

  In the future, maybe technology won’t be all about screens, and that will solve the problem. After all, when technology gets out of our way, or at least out of our hands, we should be able to focus again. But all signs point to no. Whether no screens, or lots of really tiny screens, it’s likely that getting the phone out of our grasp will not solve everything. In truth, more technology is another part of what makes us unhappy.

  O’Reilly Media, Inc.

  9/22/16 A detox suggests that technology, or at least the way we use technology, has become toxic. Even so, we don’t talk about rehab, or interventions. We are a bit stunned by the severity of Chinese internet addiction camps. We rarely hear of friends in therapy for issues of internet addiction. Instead, we detox, like we would after a treat-filled holiday season. We indulge, we cleanse, we hope to restore balance. While there is plenty of evidence that we are a little stressed from our technology, there is not really any evidence of a digital detox having short-term or long-term positive effect. So while I’ll encourage you not to be too involved in your Facebook comments (guilty) or to look up at your children rather than stare at your screen (guilty again), it won’t solve everything.

  Taking a detox isn’t likely to restore a state of well-being with technology.

  The Relentless Pursuit of Next

  We hear a lot about the new life-changing features of the iPhone 6, or excited projections about which company will be the next Uber or Tinder or Medium, or even 10 hacks for extending our battery life. Products, features, cool designs, things to buy or sell—this is a big part of the technology story. Technology's story is the next new thing and how that new thing will make our lives happier. One of the key insights of happiness studies is that people have a very hard time being content with what they have, especially when they know that others have more. So when you buy the latest iThing, you do so knowing that in a few months there’s going to be a better, faster version. And that you’re going to be stuck with the old one. It’s as if disappointment were built in from the very beginning. We metabolize our tech purchases so rapidly, that the happiness last only a short time.

  The pursuit of next keeps us running on a hedonic treadmill, where each purchase promises happiness but ultimately disappoints. It also clogs our basements and landfills with layers of junk we regret. Our smartphone home screens fill up with candy-colored icons in much the same way. Focus on next rather than now encourages a perpetual restlessness.


The rapid pace of technological change encourages us to focus on the next new thing,

rather than enjoying what we have, which is a recipe for unhappiness.

  Dehumanizing Humans

  Oh, the irony. Just the thing we love about technology, that it makes us feel competent, can also be the thing that makes us miserable. Incorporating human skills into a


machine—called “blackboxing” because it makes its inner workings invisible to the

user—allows people to do more with investing less time and effort. With your phone, you

  can make a movie, buy a car, or track your vital signs. All things that used to require a lot of hard-won knowledge. So putting the expertise in the machine lowers the barriers to

  entry. A plus. If those technologies break or disappear, a negative.

So, we depend more on our technologies than ever before, but we can do more than ever

before. Collectively and individually, technology has made us smarter, more capable, and

more productive. On the flip side, we might understand less about what we are doing.

Think about how we use GPS. While I may still enjoy looking at a map and day dreaming

  O’Reilly Media, Inc.



a little, I rarely use a map to navigate. In truth, for all my complaints about GPS and there

are many, I would have trouble relying only on a map to drive from well-known point A

to unfamiliar point B.

Blackboxing translates into a bigger fear, replacing people with machines. A rtificial

intelligence is starting to replace jobs, even “knowledge” jobs in multiple industries.

Expert radiologists are routinely outperformed by pattern-recognition software. Many

aspects of the legal profession or being automated by companies like Rocket Lawyer.

  Companies such as Narrative Science can replicate the work of human journalists. Research tells us that unemployment indeed has a harmful impact on well-being. The experience of unemployment hits people so hard that they don’t always feel as happy as before even after getting a new job. Having a purpose, using and refining skills, and feeling in control are crucial to happiness.

  Happiness research suggests that challenging ourselves and meaningful work are fundamental for well-being. Does more automating help us achieve a flow state by getting the boring stuff out of the way? Or does it take away work that helps us find meaning and purpose?

  Automating has its downsides, particularly if it takes away meaningful work.

  Social Distortion

  Didn’t I just put connection in the happiness category a few pages back? Just as technology contributes to our happiness, well, it can also contribute to our unhappiness. This is the case with our relationships too. Almost since social media came to be, it’s been studied for negative effects, among them social comparison, fear of missing out, and bullying or worse, much worse. I’ve interviewed teens, who delete more than they post, often dropping their accounts entirely due to bullying. I’ve witnessed bad behavior under the “protection” of anonymity. I’ve listened to stories of harassment that is becoming disturbingly commonplace. Policies have not caught up with practices, and design for well-being alone won’t solve this.

  But there’s also the mundane everyday misery of social comparison to consider. We’ve all felt this from time to time. One thing that’s clear from the latest happiness research is that social comparison makes us unhappy. And social media, with all the perfect vacations and beautiful babies and romantic gestures, can make social comparison a daily routine rather than an occasional rumination. Social media also can encourage thinking about others in terms of endorsements or obligations. Posts to like, comments to make, people to follow define relationships in a tightly circumscribed way. Between three-dot anxiety and the 11-like threshold, technology can reinforce the worst of our social fears.


Social media can end up distorting and diminishing our relationships, as much as

supporting them.

  Fragmented Self

  Technology stretches our experience of identity. We saw how this can be a positive thing, allowing for experimentation and augmenting our abilities. Our experience of our online

  O’Reilly Media, Inc.

  9/22/16 selves is often piecemeal too. Out of concern for our privacy, we adopt different personas in different contexts. We can control what we post and how we present our self to the world to some degree, even if it’s a little burdensome to have to think about this public doppelganger. Some aspects of our online self we can’t control so well. More and more of our experience is created by what we’ve done before—how we’ve browsed, what we clicked, who we’ve followed—and yet it doesn’t feel much like who we are. When the algorithm doesn’t quite sync up with who we are, it’s a problem. “It’s like another me made from according to Matt, who

  clicks and likes. So, it’s half-formed and honestly pretty strange,”

  spoke with me during a recent research study. So far, we don’t have a real way to understand this version of our selves or how to shape it to who we are becoming.


Our sense of self is limited by a certain inflexibility in how we can engage with it online.

  Unintended Consequences

  Just as technology can make lives easier or simplify complex tasks, it can also have negative consequences. Let’s go back to the 1950s again. Think about air conditioners. Yes, they made certain areas of the United States more livable in the summer. But they also meant less time spent sitting in the shade of a porch participating in neighborhood life. Once more people adopted air conditioning, house designs no longer favored porches. So, the unintended consequence of air conditioning for private homes turned out to be more isolation rather than community. With any technology, sometimes the good intentions we begin with don’t play out once the technology is released into a complicated ecosystem. People adopt their own practices and generate their own meaning around technology. This doesn’t mean we shouldn’t aim higher. After all, there is some evidence that internet and computer-related technology is actually contributed positively to well-being.

  The Global State of Tech Happiness

  Well-being seems to be a central goal for technology, even though we don’t necessarily shout it out loud. Ideas about how technology makes us happy are implicit in the technology we create—whether because it makes life easier or or it helps us to be more awesome and more connected. But does technology actually contribute to the good life? We don’t really know for sure. Until the last decade, economists measured well-being by proxy. Measuring wealth at a national level has served as a way to stand in for happiness. If we’re richer, the thinking goes, we have more options and so we must be


better off. That hasn’t turned out to be true. The wealthiest countries are not always

ranked as the happiest. For individuals, happiness levels off at a certain amount of

income. Even so, we still might find it hard to believe that more money means greater



Measuring Well-Being on a Large Scale O’Reilly Media, Inc.


  Money isn’t everything. But for measuring national success, it has pretty much been the only thing (other than sports). Since World War II, the dollar value of a country’s economic output has been the measure of success. While it’s great for measuring short-term fluctuations, it’s limited in a lot of ways. For one, it still misses a lot of things that have economic impact. It measures commodities, but not capabilities. Tim O’Reilly (yes, the eponymous O’Reilly) drew that conclusion too when he applied Stuart Brand’s clothesline paradox to the tech economy. You put your clothes in the dryer and the energy used gets measured. You hang your clothes on the clothesline and it disappears from the economy. Open source software, for instance, creates a lot of value. It may be the foundation of apps, but that value is not captured directly.

  National leaders have been looking to other metrics for success. That’s when happiness started to come into the equation. In the early 1970s, economist Richard Easterlin pointed out that national happiness polls did not correlate with per capita income as expected. The Easterlin Paradox means that beyond a certain level, rises in income over time fail to increase happiness. Around the same time, Bhutan proposed Gross National Happiness (GNH) as a supplement to the GDP. The feeling was that the GDP was not a great proxy for a good life. Instead, looking at subjective well-being in tandem with health, education, good governance, conservation of the natural environment could be a counterpoint to economic growth. The United Nations’ Human Development Index followed suit in the 1990s. The Organization for Economic Co-operation and Development (OECD) has released a well-being index, Better Life Initiative , every year since 2011.

  How can you measure happiness anyway? Mostly, you ask. I know what you’re thinking. In the design world, we are suspicious of asking. People can’t always articulate what they are thinking or feeling. While that may be true when interacting with a prototype, there are lots of valid research methods that rely on people answering questions about themselves. Plus, it turns out you can use magnetic resonance imaging to measure cerebral blood flow, or electromyography to measure the activity of the “smile muscles” in the face but those measures are highly correlated with self-reports for happiness.

  Happiness itself seems a little squishy, but the approach to measuring it has become more and more sophisticated. First, both in the moment and reflective questions are combined to understand subjective well-being. The Cantrill Ladder is one standard question but the Oxford Happiness Inventory and the Panas Scale show the range of questions designed to capture happiness on emotional, behavioral, and cognitive levels.

  Second, happiness indexes take into account multiple measures, sometimes drawn from multiple sources. Think a dashboard of indicators rather than a single number. More than just a feeling, happiness with a capital H takes into account other contributors to well-being including community, environment, health, and a number of other factors (see table x). Using social media as an indicator of community well-being has been tried too. University of Vermont’s Hedonometer assigns a happiness score to certain words and then monitors about 10% of all posts to Twitter. While it’s an imperfect sample, it seems to correlate to many subjective well-being surveys and is more sensitive to changes over a given time period. O’Reilly Media, Inc.



Clearly, the GDP doesn’t account for many of the things that make life

worthwhile. For now, there’s no single well-being or national happiness index

though, no big number like the GDP to tie it all together. That makes it harder to

track and adopt. And that’s why you might see different country rankings

reported at different times of the year.

For our purposes, designing technology for happiness, there aren’t a lot of

directly applicable measures. Or, maybe it’s just difficult to tease them out.

Community, could have an online component even when we think of local

neighborhoods. While I have a strong sense of community in my small Hudson

Valley NY town, it is strengthened and extended by Facebook parent

communities, the local freecycle site, and even the school’s rather

lacking “portal."

Despite the complications, current approaches to measuring personal

and collective values can be a good model for building well-being with

technology. If we accept that what we measure ends up being what we cultivate,

then it seems like a good way to start. I’ll talk about that in greater depth in

chapter 4.

  Greater wealth is generally associated with greater access to technology, too. So, you


would think that happiness and techiness might be somehow connected. Ask your teen

  daughter or your mother or almost anyone else if the latest iPhone will make them happier, and the answer will be a definite yes. Just as we tend to equate more wealth with greater happiness, we also tend to think that more technology can lead to more happiness.

  So far, there is not a lot of evidence to fall back on. Some global surveys like Gallup World Poll measure access to technology, in addition to happiness, healthcare, community, environment, and a number of issues related to overall well-being. In other studies, like the Organization for Economic Cooperation and Development (OECD) Better Life Index, it’s not yet measured separately at all. What we can do instead is to pull together a meta-analysis of global happiness measures and cross-reference it against

  other technology measures for countries.

While happiness, or life satisfaction, or well-being, is measured with multiple indicators

(more on that in the sidebar), let’s just keep it simple for now. Let’s use the self-reported

answers to the Cantrill Ladder prompt, a standard question in well-being surveys: “Please

  imagine a ladder with steps numbered from zero at the bottom to 10 at the top. The top of the ladder represents the best possible life for you and the bottom of the ladder represents the worst possible life for you. On which step of the ladder would you say you personally feel you stand at this time?”


I’ll take the answers to this question from the most recent World Happiness Report and

cross reference that with a few measures of “techiness", like access to broadband and use

of personal technology, reported in the International Telecommunications Union annual

report, Measuring the Information Society .


How Happiness Relates to Tech Access O’Reilly Media, Inc.



Happiness and Techiness Chart

  Looking at measures for global happiness and how they intersect with access to technology above, the correlation is clear. Other researchers, looking at the Gallup World Poll corroborate. Access to technology does have a positive relationship with happiness. Nations that are more technologically advanced, including

  greater broadband access and more citizens with devices in their homes, are happier.

  A rapid digital acceleration may also be linked to happiness. For many Latin American


nations happiness and tech access have simultaneously been on the rise. The rapid

increase in mobile adoption may be contributing to an overall sense of well-being. Where

there was little access to technology before, an increase in access or possibly even a

promise of access may be connected with an increase in reported happiness.


Don’t go out and buy more devices and upgrade all your tech just yet though. Access to

broadband and personal devices like smartphones contributes positively to happiness. As

a basic utility, technology bumps up happiness. But it’s not clear whether more

technology will create more happiness.


Up to a certain point, which seems to be about $75K in the US, money contributes to

happiness. Maybe because money covers those basic needs at the bottom of Maslow’s

hierarchy like food, shelter, safety. Then happiness levels off. More money doesn’t mean

more happiness. It may be that access to technology levels off at a certain point too.


The way a country governs technology also may be tied to happiness.

  Scandinavian countries, among the happiest, tend to have a regulatory authority that protects consumer interests and encourages a competitive market. An approach to technology that is geared toward the collective well-being of a society may have an impact on personal well-being for individuals in that society.

  So, one way to increase well-being with technology is to improve access. Another is to cultivate trust and transparency in how technology is regulated. No small order.


More than just a silicon and cables and code, when it comes to personal well-being, it is

about how technology becomes a part of our lives. Technology enables communication,

socialization, learning, growth. It’s an extension of self. Happiness is in the experience of

technology. That’s something we assume to be true, and can guide us as we try to

develop it further.

  Trending Toward Happiness

  Positive intentions aside, there’s some evidence that technology is actually aligned with our well-being. There’s also a lot of evidence that it is chipping away at our well-being. How do we create technologies that are more apt to cultivate well-being? It’s a tricky question. And it may mean a shift in the current focus on problem-solving.

  The Problem Problem

  Competence and autonomy, the hallmarks of our current vision of tech happiness, are at the core of all our methods. In truth, it’s a kind of mantra. “Design must solve problems,” we reaffirm. Design thinking, now wholeheartedly adopted by organizations of all kinds,

  O’Reilly Media, Inc.

  9/22/16 is adopted as a problem-driven approach. Organizations tend to practice design thinking by starting at an undesirable situation and then imagining how to resolve it in an ideal future. Our customer journeys aspire to move people from pain to pleasure. Focusing on solving problems, what Evgeny Morozov calls “solutionism” when taken to extremes, does not always lead us to positive and worthwhile experiences. Making something difficult easier doesn’t always make us happy. Actually challenges often make us happy, whether we get in a flow state through satisfying work or helping us develop our own meaning from the experience. By removing all friction, we remove moments for


personal growth, serendipity, and self-reflection. At scale, these may start skew our lives

towards intolerance and impatience, a lack of resilience, and an inability to navigate


  Lately, we’ve started to confront a bigger question, too. Problem-solving for whom? A while back, this quote by Startup L. Jackson was making the rounds, "The Uberfication of everything is turning San Francisco into an assisted living community for the young.” Uncomfortably funny. All this problem solving may be designing growth out of everyday experiences and focus on problems that aren’t really that big after all. How will the new


economy engage and support the least visible among us? How can we move from sharing

economy to caring economy? Tap, Scroll, Repeat

  The problem-solving approach, in some contexts, introduces another complication. It can be directed at filling empty moments that might be a little uncomfortable or just boring. Texting for a minute feels better than standing in line. Listening to a podcast feels better than a walk in silence. So rather than a vitamin, we learn how to design a pain killer.

  Using Skinner’s model of trigger and variable reward, websites and apps encourage repeated use. In response to so much pressure to grow we measure success in clicks and page views. Design blogs do a great job of documenting techniques for making experiences frictionless and getting attention. Endless scroll, auto-play, default settings, and even menu choices increase engagement but maybe at the expense of personal well- being.

  Now, we may be so good at this type of behavioral design that experiences are starting to elicit compulsive behavior. So much so that Nir Eyal, who popularized this model in his book Hooked, has cautioned technologists to consider the consequences. We may be solving problems a little too well, as it happens.

  A Wellness-Based Approach

  We know how to get people to click, we know how to make technology easy to the point of disappearing into the background, but do we know how we can maximize well-being? It turns out that technology contribute to our happiness. The next wave of


  technology, grounded in the latest knowledge about the science of happiness, can cultivate our best possible selves. The disconnect between positive intentions and not-always-positive outcomes is prompting new approaches to designing technology. Solve CoLab and Games for Change create technology that is more socially relevant and purposeful. Time Well Spent is a

  O’Reilly Media, Inc.

  9/22/16 movement toward minimizing distraction caused by technology. Value by design and slow design encourage an intentional focus on long-term meaning over short-term interaction.


Convergence of Design Trends

Positive Design Trends


More and more people who work in technology are beginning to take well-being in to

account. Sometimes directly. Certainly there are apps for physical health as well as plenty

of options for psychological well-being. Apps , like Happify and Talkspace, employ a

range of interventions drawing on cognitive behavioral therapy and positive psychology.

  Likewise, Headspace, Smiling Mind, and the growing number of mindfulness apps encourage people to be reflective. But we need a new model — a wellness model. Psychology (and medicine, in general) has shifted focus from surviving to thriving in the past decade. Researchers have begun to study how flow, gratitude, and compassion can get people beyond a neutral state to well- being. Here is groundwork for understanding how our relationship to technology can move beyond a neutral state. In designing technology, we still need to address pain points, but should devote more thought to the well-being of the whole person on the other side the screen. Not just as users of technology, but as individuals living complicated lives and striving toward better health, creative outlets, and fulfilling relationships. Not just having, or doing, or even feeling in the moment, but becoming. Designing for happiness is not restricted to material, short-lived pleasures but also to other elements of personal and collective well-being. So how do we design differently? So far, there is a lot of encouraging movement in the design world. In the following chapters, we will look at how build on what we already do well to create a new model.

  Field Note: Tech Highs and Lows

  Because misery and addiction seemed to dominate our way of thinking about technology, I was determined to study what might make us happy about technology. One way I tried to understand this was by documenting highs and lows.

  Really, at first, I had maybe naively wanted people submit their peak moments with technology. As it turns out though, it was difficult to get below the surface of likes and lolcats. Once people were asked to think about highs and lows, something shifted. Just as we know that a positive emotion like pride is connected with a negative emotion like shame, so it seemed that technology’s best moments were connected somehow with the worst. Maybe it’s like a vacation, where some of your worst moments might end up being your best memories. Over the course of a month, my research team had diverse group of 50 people—diverse in abilities and baseline happiness level as well as ethnically and geographically diverse—keep an online diary of highs and lows. Each person would submit a picture, a description, and answer some questions about the emotions associated with the moment and tell us why they felt that way.

  O’Reilly Media, Inc.

  9/22/16 Some of those moments are what you might expect. A funny app with talking animals, an uplifting story, a picture of a grandbaby. More often, it went much deeper than that. Moments of stolen intimacy with a loved one, moments of meaningful connection, sharing mundane everyday details and small personal triumphs were universals. And there were unexpected moments too. Creativity felt in Snapchat crowns and Spotify playlists. Personal growth, ranging from Timehop to Duolingo, was clearly important.


Examples of Peak Happy

Diary snapshots of peaks

  While people did feel happy when they were entertained, people felt happiest when they were creative, connected, and challenged. Rather than passively consuming, these were moment of active participation.


The Mix of Happy Emotions

Chart of emotions

  The lows centered on ways that technology distorts what we feel to be real, particularly about our sense of self and our relationships. Like our positive moments, we also tagged the negative moments, breaking it out into the categories below.

3 Ways Technology Makes Us Unhappy

  Abilities Relationships Time Tethered Comparative FOMO Stressed Obligated Distracted Automatic Performative Fragmented Incomplete Voyeuristic Shallow

  Obsessed Disconnected

  Rather than cultivating a deep participation in their own personal growth, a relationship, or whatever each might define as a good life, these lows left people feeling off-course. The more intense the participation, the more meaningful the experience. And the more people felt a sense of deep, and sometimes complicated, well-being.

  As researchers, designers, and developers, the implications of this research steer us toward a different approach to happiness. One that is not entirely frictionless or delightful, but that strives toward meaningful engagement in the long and short-term.

  Briefly Happiness is an implicit goal of technology, but our definition of what

  • constitutes tech happiness continues to evolve. Creating happiness with technology, so far, centers on making people feel more
  • competent and connected. It is likely that the goals for how technology can play a role in happiness will broaden as our use of technology does.
O’Reilly Media, Inc.


  While technology implicitly, and sometimes explicitly, aims to create the good

  • life, those good intentions don’t always play out. The relationship we have with technology is conflicted. Whether or not Internet addiction is an official condition, we feel an unhealthy pull toward technology. This is becoming part of the inner narrative we have about tech use. The tech community is beginning to acknowledge the broad and deep impact
  • that technology has on people’s lives. Because technology is so deeply embedded in our lives, we can’t not have an impact on well-being. For now, we do know that access to technology does have a quantifiable impact
  • of collective and personal well-being. Yet, we still don’t intentionally choose happiness as a goal for designing
  • technology. Instead we design to solve problems, which is only part of the picture. Taking away negatives doesn’t always lead to big-picture happiness. A convergence of methods in the design world, like Value by Design, Slow • Design, and Time Well Spent, propose alternative models that take into account broader goals.

  Further Read Clive Thompson’s Smarter Than You Think: How Technology is

  • Changing Our Minds for the Better (2013), a nice counterpoint to Nicholas

    Carr’s The Shallows: What the Internet Is Doing to Our Brains (2011)

    While Neil Postman’s point is about how television degrades us individually and
  • collectively, Amusing Ourselves to Death: Public Discourse in the Age of Show Business (2005) it is often applied in the context of technology. I found Virginia Heffernan’s Magic and Loss: The Internet as Art (2016) to be a refreshing take on the positive potential of technology.
  • For how design itself may be leading us to addiction, I enjoyed Natasha Dow

    Schull’s Addiction by Design: Machine Gambling in Las Vegas (2013).

    Sherry Turkle’s Along Together: Why We Expect More from Technology and
  • Less from Each Other (2012) read alongside the only semi-positive exploration of social media that has made the mainstream, danah boyd’s It’s Complicated: The Social Lives of Networked Teens (2013).
  • Maybe you haven’t read Badass: Making Users Awesome by Kathy Sierra (2015)? You should.

  O’Reilly Media, Inc.


  6 An Architectural Lens Layout, Surfaces, Structure and Change

  This chapter presents an ‘Architectural Lens’ on design and behavior, with patterns drawing on techniques used in architecture, urban planning and related disciplines such as traffic management and crime prevention through environmental design. While most of the techniques have been developed in the built environment, many of the ideas can also be applied in interaction and product design, even in software or services (often metaphorically): they are effectively about using the architecture, or structure, of systems


  to influence people’s actions. Some of the patterns, such as , DELETION and HIDING are really fundamental to all kinds of design. Probably the most fundamental design patterns of all, around human behavior, relate to the layout of entities in relation to one another—physical objects, elements on a screen or other interface, and so on. These are the most obvious examples of how design influences people’s actions physically, whether intentionally or not. The four patterns here,


  , & , & and &


  , are perhaps intuitive enough for most designers that they need no explanation, but sometimes it is worth surfacing these ‘obvious’ techniques, and exploring their use in different contexts. In Chapter 10, on the Perceptual Lens, we will revisit aspects of layout in more depth, through looking at patterns deriving from Gestalt psychology (see Chapter X). The ways in which people experience the surface of entities (physical, but potentially digital as well) is important in architecture and product design, and can be used to affect how people act in a wider range of context. Here we look specifically at patterns around


  , ( ) , and of surfaces. Beyond the layout of specific design elements, we can also think at a more ‘system architecture’ level about how things are structured, and how that structure influences people’s actions.


  Patterns covered here are , , ,


FEATURE DELETION , and, very fundamentally, SIMPLICITY . Finally, we look at where


  architecture changes in relation to people’s actions, with two patterns, and PAVING THE COWPATHS . The latter of these represents a particularly distinctive stance towards behavior, as we have explored in more depth in Chapter X.

  O’Reilly Media, Inc.


  Positioning What would happen if you rearrange things so people interact with them in the locations you

want them to?

  Figure 6-1: Pedestrian crossing push-button units can be positioned so that pedestrians waiting to cross are more likely to turn to face— and notice—oncoming traffic. This example is in Brighton, England.

  Figure 6-2: Where crossings are designed with a central ‘island’, the lights and barriers can be positioned so that pedestrians are again turned to face oncoming traffic, and potentially make eye contact with drivers.

  From planning a dinner party to designing an interface, much design is about strategically


  elements so that people interact with them in locations that serve the needs they, and/or you, have for the situation. We often notice this pattern by its absence— when things are positioned badly. This could be that elements are difficult to find (e.g. on a web page), or it could be that they are positioned such that certain other actions take place or don’t take place as a consequence.


  is mainly about enabling or

  constraining action. It mainly assumes a pinball or shortcut model of people’s

  actions. Using


  does not require the designer to know much about how different people are experiencing, or thinking about, a situation. Unless the placement of elements adapts to situations, it is usually ‘one size fits all’, by its very nature. There could be consequences—for example, the height position of a screen or interface could make it easier or more difficult for wheelchair users to use. Figure 6-3: This poster for a Japan Association event, by placing the (small) text lower down, causes some visitors to ‘bow’. Designed by Grey Group India. O’Reilly Media, Inc.


  What would happen if you channel people so they come together (or split Converging & up)? diverging

  Figure 6-4: Mediaeval towns in Europe often Figure 6-5: Tollbooths, as part of larger had gated entrances, such as here at Rye, East toll plazas, both converge vehicles to pay Sussex, England, which channeled visitors the toll, and often also diverge them (merchants, potential attackers) through a depending on choice of payment method. narrow gateway. This example is in central France.


  and people’s routes through a system can enable a designer to control, or at least shape, people’s movement or progress, either physically or metaphorically. This might be in order to carry out some kind of temporary action or treatment that needs people to be observed or processed one at a time, or could be about

  TAILORING splitting people into groups which can then enable (see Chapter 8).


  & are mainly about constraining action.


  & mainly assume a pinball model of people’s actions. From the perspective of the person being ‘converged’ or ‘diverged’, the experience may not always be pleasant, and may not even be voluntary. The approach necessarily imposes a power structure; the ‘sorting’ aspect of diverging may also enable discrimination against, or unwanted

  Figure 6-6: The ‘conversion funnel’ idea in

  SEPARATION of, a group. Use carefully.

  marketing has parallels. For example, there are many routes into finding something on eBay, but ultimately all customers need to converge on buying the item.

  O’Reilly Media, Inc.


  What would happen if you get people to follow a path you want them to, on the Mazes & paths way to reaching something they want?

  Figure 6-7: Some store layouts channel Figure 6-8: PATHS can be marked on the shoppers in a MAZE past ‘impulse floor of factories, airports, hospitals, and purchase’ items—often snacks—on their stations, to route pedestrians. Here at a way to the checkouts. E.g. IKEA has nuclear research facility, safety is the main traditionally routed people through a kind reason for the path markings. of physical version of its catalogue.


  and can, as with other patterns here, route people in a particular way through an environment or a system. Generally, the aim is to ‘prescribe’ a particular sequence through the environment (as in some museum exhibits, for instance), so that people pay attention to particular elements, don’t deviate from a set path, or remain within the environment for more (e.g. in a shopping mall) or less time.

  MAZES & PATHS are mainly about MAZES PATHS

  & mainly assume a pinball constraining action. or shortcut model of people’s actions.


  If users are forced in a particular way through a , they can feel manipulated, or try


  to avoid the space entirely. However, suggested and clearly marked can help people with wayfinding in unfamiliar and potentially stressful environments such as hospitals. The extent to which museum or gallery visitors’ routes should be intentionally curated, and how much this embeds politics into the experience, has been a matter of debate, e.g. in Fred Turner’s book The Democratic Surround (U. Chicago Press, 2013).

  O’Reilly Media, Inc.


  Roadblocks & speed bumps What would happen if you put things in people’s way, so they adjust their speed, or take an alternative route?

  Figure 6-9: ‘Chicanes’ can slow down drivers, pedestrians and cyclists; this chicane on a narrow path is supposed to ‘enforce’ pedestrian priority.

  Figure 6-10: Speed bumps, humps or cushions are nicknamed ‘sleeping policemen’ in the UK—the idea of the object preventing, or punishing, speeding.

  ROADBLOCKS and SPEED HUMPS temporarily impede people’s progress through an environment or a system. Reasons could be related to safety, to enforce priority (e.g.

  drivers yielding to others), or to limit the number of people entering an area (e.g. a website or a nightclub), or to enable a procedure to be carried out which involves the people who are slowed down or diverted. The aim is usually not to prevent access to an area entirely, just to slow it.

  ROADBLOCKS & SPEED BUMPS are mainly about constraining action. ROADBLOCKS & SPEED BUMPS mainly assume a pinball model of people.

  From the perspective of people whose journey is impeded, the experience may not be pleasant. In other circumstances it may simply be a temporary ‘bump’ on the way to something better. As with other Architectural Lens patterns, the approach imposes a power structure, in which the designer (or whoever operates the system) has substantial (temporary) influence over people’s actions. Figure 6-11: At times of very high demand, some retailers’ websites are configured to operate a ‘queueing’ system, to block new visitors from entering. This example is from the UK catalog store Argos. O’Reilly Media, Inc.


  What would happen if you use the properties of different materials to make Material properties some actions easier or more comfortable than others?

  Figure 6-12: Differently textured paving Figure 6-13: [WILL REPLACE]. Design can act as a subtle barrier between cycle researcher Julia Johnson ran workshops using and pedestrian tracks, such as this the non-slip material Dycem to enable people example in Oulu, Finland: stray over the who struggle with grip or mobility to line on a bike and you’ll feel it. customize products and environments to make it easier to take actions they wanted.


  can be used to make an action more or less pleasant, or simply to draw attention to the difference between two areas or parts of something. In some cases a material property can be used to prevent something happening entirely, such as police flashlights or plastic prison cutlery that can be designed to break if used as a weapon. In product design, textures can be used to increase the likelihood that people hold or grip something in a particular way, or even simply to encourage certain interactions through providing a perceptually appealing or intriguing texture.


  can be about mainly assume a

  enabling or constraining action, or pinball or shortcut model of people’s

  potentially—if appealing enough—about actions.

  motivating certain actions.

  While the use of particular properties or textures can empower people, as seen above, there is also a history of using MATERIAL PROPERTIES in ‘disciplinary architecture’ (see Chapter 3), for example to make it difficult or unpleasant for protestors, or homeless people, to congregate in certain areas. Unpleasant Design by Selena Savić and Gordan Savičić (Gloria, 2003) chronicles other examples such as a sandpaper lacquer on a bridge rail to make it unpleasant to climb, and a rubber texture on lampposts to prevent posters being affixed.

  O’Reilly Media, Inc.


  Hiding What would happen if you hide, temporarily or otherwise, functions or elements you’d prefer people didn’t use or see?

  Figure 6-14: Quora, a question-and- answer site, ‘hides’ the answers until users sign in, although the text is tantalizingly shadowed rather than fully hidden.

  Figure 6-15: These church hall heating controls have been hidden (leaving only the timer accessible) to reduce errors by users unfamiliar with them.


HIDING things strategically can be used to make some actions more difficult, or perceived

  as more difficult (or even impossible), or, crucially, not perceived at all, but can also enable users to focus on whatever is considered most important in a situation by removing extraneous elements. If partially hidden, it can provoke and entice.

  HIDING is mainly about constraining action, sometimes enabling. HIDING mainly assumes a pinball or shortcut model of people’s actions.

  There may be a different experience of

  HIDING depending on whether users are

  aware that things are being hidden from them. If you don’t know that something is being concealed from you, it may not cause you concern, but if you do—and cannot find it or get access to it—the experience can be frustrating. As with most of these Architectural Lens patterns, some power lies very much in the hands of the designer. Figure 6-16: Medium’s editing interface for posts hides almost all functions, simplifying the interaction visually and making for reduced distraction while writing. O’Reilly Media, Inc.


  What would happen if you angle things Angling so some actions are easier than others?

  Figure 6-17: Slanting the top surface of these lockers makes it impossible to put things on top—important in some schools, workplaces, and shared settings such railway stations.


  surfaces can be used to ‘present’ something to users more directly (such as a screen), but is also often used to make some actions more difficult. Bins often feature slanted top surfaces, to make it less likely that litter is left on top; a sound technician told me once that he added angled tops to speaker cabinets used at large concerts to prevent audience members putting beer glasses on top, which would inevitably have been spilled.

  Figure 6-18: This old fire bucket outside a London pub has a convex bottom, making it less useful as a normal bucket and so unlikely to be ‘borrowed’. It also means it is less likely to be left on the ground, as it will not be stable.

  ANGLING is mainly about constraining action, sometimes enabling. It mainly assumes a pinball or shortcut model of people’s actions.

  A relatively subtle way of influencing people’s behavior,

  ANGLING is perhaps most likely to be experienced as a minor

  inconvenience, if it is noticed at all. Social implications start to come when, for example, angled surfaces (e.g. on benches) are used as anti-homeless or anti-loitering measures, to prevent people sitting or lying (see Chapter 3). O’Reilly Media, Inc.


  What would happen if you make (Un)evenness something more or less even or regular in its structure or surface?

  Figure 6-19: Deliberately uneven staircases Figure 6-20: Uneven spacing of rumble have been used—with one or more steps a strips (getting progressively closer) on different height—to catch out potential approach to road junctions can lead drivers 1 intruders and make them fall. There’s no to perceive they are travelling faster than suggestion the varying step heights here at they are, so they slow down more quickly. a restaurant are deliberate, but they do cause stumbles from customers.


  ( ) can be manifested through the ways in which parts of a system, or the surface of something, are spaced so that they are experienced in a particular way. Even, regular spacing can lead to someone more easily falling into a rhythm—for example, performing a repetitive task on a production line, keeping synchronized with others (e.g. drummers on a dragon boat), or even the use of a metronome. Uneven spacing can slow


  people down by forcing them to take care, or to pay attention; as part of

  PROPERTIES , it can also manifest in ‘disciplinary architecture’ (see Chapter 3).

  ( UN ) EVENNESS is about constraining or enabling action. EVENNESS probably mainly assumes a pinball or shortcut model of people’s actions, but UNEVENNESS can also be used with an assumption that people are thoughtful, or can be prompted to become so, by getting them to pay attention. The diversity of possible ways in which ( UN ) EVENNESS can be manifested means that it is difficult to generalize about the experience. However,

  EVENNESS or UNEVENNESS THAT APPEARS INTENTIONAL is likely to be experienced 1 differently to something that is assumed to be accidental or an error.


Henry Petroski notes the case of William Grey’s house in Edinburgh, built in 1662 (Small Things Considered , Vintage, 2003). See also Harry Brignull’s blog post, ‘Dangerous by design’: burglary-staircase/ O’Reilly Media, Inc.


  Segmentation What would happen if you divide your system up into parts, so people only use one part at a time?

  Figure 6-21: These individual seats on the Paris Métro are spaced so that someone cannot lie down or occupy more than one.

  Figure 6-22: The jokes on some ice lolly or popsicle sticks are broken in half, with the punchline revealed through eating.


SEGMENTATION involves breaking something up into two or more parts, so as to make it

  hard to use more than one part at a time, to make something more ‘digestible’, to direct people’s attention to use parts differently, to encourage following a sequence, or to suggest that ‘categories’ exist. The technique may involve spacing between the parts such that they cannot be used, or seen, together, or are experienced in sequence. This use of the term is distinct from segmentation in marketing, which we mention in Chapter 2.


  can be about constraining action, or enabling in the sense of making ‘sorting’ or sharing easier. It mainly embodies pinball or shortcut models.

  The designer has significant power in


  through structuring how users encounter or perceive a situation. It can be used with political intent (e.g. segmented anti-homeless benches), commercial intent (e.g. pagination of articles), or with other intentions. Figure 6-23: Pagination of articles can increase advertising revenue, potentially at the risk of frustrating users. O’Reilly Media, Inc.


  What happens if you structure things so Strategic juxtaposition people come into contact with elements, ideas or people they wouldn’t otherwise?

  Figure 6-24: The London headquarters of the aid charity CAFOD was designed by Black Architecture to promote interaction and knowledge-sharing between different departments, through one side of the building with shared social spaces such as kitchens being ‘offset’ by half a floor from the other side, with visibility between them.


  is a form of , but with the specific intention of introducing people to other people, design elements, or particular content that they would not otherwise encounter. This can have a direct short-term behavior change motivation, or be something intended to change attitudes over the longer term, e.g. through familiarizing people with new ideas or other cultures.

  Figure 6-25: Socratic Search, created by Ted Hunt, could be used to ‘engineer’ serendipity in online search behavior, breaking people out of 2 their ‘filter bubble’ by presenting opposing views, juxtaposing alternative positions on news stories and concepts—filtered by clarity,

  implications , evidence, alternatives and assumptions .


  is mainly about

  enabling actions. It potentially transitions

  people from a pinball or shortcut model to a more thoughtful one. ‘New’ elements might be ignorable, but it is clearly also possible to design to force confrontations. Equally, the pleasure of serendipitous meeting or idea juxtaposition can be something actively sought.

2 Eli Pariser, 2011, The Filter Bubble: What The Internet is Hiding From You. Penguin.

  O’Reilly Media, Inc.


  What would happen if you separate a Separation system into sections targeted at specific uses or categories of user?

  Figure 6-26: Cycle paths can be Figure 6-27: Schools and workplaces often segregated from other traffic and had separate entrances for categories of pedestrians, using painted lines (as here people (e.g. here at a former school converted in Nice) or physically protected with into the headquarters of Zaha Hadid’s curbs that offer more tangible separation practice in London). See also the recent trend and reassurance. for ‘poor doors’ (Chapter 3).


  can be as simple as dividing a system into areas or sections which are labeled or obviously presented as being ‘for’ particular activities. For example, creating an area of a shared office which suggests a more relaxed atmosphere (lower chairs, coffee table) can potentially lead to its use in that way. An area can be made more appealing, or desirable, in contrast to others; or enforced division or restricted access based on certain criteria (see the Security Lens, chapter X) can influence behavior in very different ways.


  can be about enabling, motivating or constraining actions.


  mainly assumes a pinball or

  shortcut model of people’s actions; sometimes maybe thoughtful.


  Clearly, in the form of segregation has a darker history and connotations, particularly in conjunction with related practices such as redlining. The experience as a ‘user’ of a system that is separated can vary considerably; the very act of being able to choose (or not) which area to access can itself reinforce or challenge power

  Figure 6-28: Reddit’s ‘Lounge’ subreddit st structures. The notion of classes (e.g. 1 or is a special ‘VIP’ area reserved for users business class travel) can be a price who have been given, or have bought,

  discrimination strategy (see Machiavellian ‘Reddit Gold’.

  Lens, chapter Y). O’Reilly Media, Inc.


  What happens if you structure things Simplicity simply, to make it easier for people to do what you’d like them to do?

  Figure 6-29: Sign-up processes often Figure 6-30: EcoButton puts a Windows PC

  SIMPLICITY provide examples of being into a low-power state with just one press.

  used to influence actions. Here, Pinterest Note it does not replace other functions


  claims “15 seconds to sign up”. The rise (compare )—it just makes in social media integration and use of a very simple, easy interface for a series of OAuth has made this faster. actions.


SIMPLICITY is often fetishized in design, although some good treatments provide a more


  nuanced approach. In the context of influencing people’s actions, simplifying things is often a reduction in the number of steps needed—the minimum viable action in Stephen Wendel’s term—to turn an otherwise complex behavior into something easier to do.


  is about both enabling and

  constraining actions. It can assume either a shortcut or pinball model of people.


  , in common with


  , and , takes a stance where designers have significant power to influence the way someone experiences a situation. But, arguably, thoughtful reduction can

  Figure 6-31: On this emergency coastguard empower people to act who might phone, in a remote area on the Isle of otherwise be put off by perceived Wight, dialing any number will put the complexity. See also the discussion of caller through to emergency services. We reductionism in Chapter X. might question why multiple buttons are 3 needed at all.


Giles Colborne, Simple and Usable, New Riders, 2011; John Maeda, The Laws of Simplicity, MIT Press, 2006. O’Reilly Media, Inc.


  What would happen if you take away Feature deletion features you don’t want people to use?

  Figure 6-32: This hotel taxi phone features Figure 6-33: Various politicians have a single speed-dial button to connect the proposed removing standby buttons from caller to a local taxi company electronic products, in an attempt to reduce (appropriately enough, Streamline Taxis). energy use by forcing a ‘hard’ on/off. The other buttons are hidden under the white insert. Compare Figure 6-31.


  is related to notions such as (Chapter X) and may be


  presented as an extreme form of (this chapter). In the sense of removing affordances, it can be used to prevent someone doing something, believing that it is even possible, or make it much more likely that he or she chooses to use the features that are available. Motivations might be commercial (e.g. driving consumers to use a more expensive feature) or for social benefit (e.g. safety). Perhaps the best social benefit example here is from 1854, when the physician John Snow asked the local Board of Guardians to remove the handle from the water pump in Broad Street, Soho, London, which he had identified as being the center of a cholera epidemic.


  is about constraining actions. It generally assumes a pinball model of people.


  is one of the ‘strongest’ patterns in the Architectural Lens: if a feature is simply not there, people cannot use it. Designers’ power is very evident

  Figure 6-34: Apple has progressively here. Yet people are adaptable: they will removed ‘legacy’ features such as optical find ways round constraints which prevent drives and ethernet sockets from its them doing what they want to. products (with 3.5mm headphone jacks on iPhones next on the hitlist at time of writing), driving adoption of its own standards (and peripherals).

  O’Reilly Media, Inc.


  What happens if you bring a feature to people, Conveyor belts or move people to where you want them to be?

  Figure 6-35: Moving pop-up reminders or Figure 6-36: Moving walkways or travelators in nag screens that ‘follow’ the reader as he or airports and railway stations help travellers move she scrolls are used on some sites to try to more quickly, prevent people blocking corridors, influence sign-up or subscription, as here especially in groups, and potentially reduce on Times Higher Education. perceived distance. The idea of CONVEYOR BELTS as a pattern is that by making elements move or change their position, people’s interaction with those elements, or other parts of the system can be influenced. This might include a literal application of ‘meeting users where they are’ (see Chapter X) or moving people, automatically or otherwise, to somewhere else. The original sense of a conveyor belt on a factory assembly line, which brings items to each worker in turn, can be applied in contexts as diverse as museums or galleries (making sure that exhibits are experienced in a certain order) and restaurants (e.g. conveyor belt sushi) to present diners with dishes from which to select (encouraging impulse choices).


  can both constrain or enable particular actions. They mainly assume a

  pinball model of people’s actions. From the perspective of the person being ‘conveyed’ or

  having something conveyed to him or her, the lack of control may mean the experience is essentially not voluntary, particularly where the intention is strategically to limit the person’s time with something (e.g. viewing the Crown Jewels in the Tower of London).

  O’Reilly Media, Inc.


  What happens if you recognize how some Paving the cowpaths users act, and codify them into your system for others to follow too?

  Figure 6-37: A classic built environment Figure 6-38: Some cities ban barbecues example of PAVING THE COWPATHS is the entirely in public spaces because of problems ‘formalization’ of desire paths (see with litter, fire risk and damage, but others

  Chapter 3) worn across a park or (e.g. Stockholm, here) design ways which campus, by turning them into ‘real’ make it easier to behave ‘better’ when paths, channeling pedestrians and barbecuing, with built-in stands and litter bins reducing further wear. provided.


  We have explored the ideas behind , stigmergy, and the notion of ‘understanding what people are trying to do and helping them do it better’ more thoroughly in Chapter 3. In practical applications in both the built environment and in systems more widely, the pattern may be used to suggest particular actions to users who might not otherwise consider them or know what to do; to make it easier for all users to benefit from the ways in which some users have found to use a system (e.g. redesigning menu hierarchy to match common workflows); or to make it faster or easier (and so more likely) for certain behaviors to occur.


  is about enabling actions. It generally assumes a shortcut model of people’s actions. The experience of a ‘paved cowpath’ is often one of low friction (Chapter X) from a user’s perspective, to the extent that the ease with which one is channeled into particular behaviors may not even be

  Figure 6-39: Twitter formalized emergent perceived. There is a risk that this practices of using @replies, #hashtags, and frictionlessness can lead to the ‘RT’ for retweets through incorporating them unthinking normalization of certain into its interface in low(er)-friction ways, to actions uncritically, with other encourage ongoing and widespread adoption. consequences.

Dokumen baru