Practical Ethics for the Modern Web Designer

Morten: I wanted to start this talk with a little conversation with you, Rachel.

Rachel: Okay.

Morten: I have a simple question. Do you trust me?

Rachel: Yes.

Morten: Why?

Rachel: Well... because I have experience with you, and you're my friend, and I think I know you enough to trust what you say and what you do.

Morten: You think you know me enough to trust what I say and what I do. This is the core of what I want to talk about, this idea that we can trust one another. Now that I asked that awkward question, I can switch over to the slides.

When I was preparing for this talk, which actually started back in October of last year, I was pondering this idea of whether someone can call themselves a Web professional, or if they're just Web practitioners. I tried to understand what exactly constitutes a professional rather than a practitioner, and can we actually call ourselves professionals?

I came across this definition from this trusted source, Wikipedia, that says, "for a type of work to be called a profession, you need to have a bunch of requirements: an occupation becomes a full-time occupation; the establishment of a training school, a university school, a local association, a national association of professional ethics, and the establishment of state licensing laws."

This is the definition of what a professional is. Once you start talking about professionals, you're looking at a person who, not only went through the proper training to do what they're doing, but a person who is held accountable for their actions, held to their personal ethics.

I want to talk to you about trust. According to Rachel Botsman, at Oxford University, trust is the currency of our interactions. The reason I trust Rachel and Rachel trusts me is that we've exchanged trust currency over time, establishing a barter system, where I do things for her and she sees that I'm doing it the right away, and vice-versa.

When you're looking at family, friends, or close acquaintances or coworkers, it's easy to build up this trust currency. But the people we trust the most in the world are people we don't really know. If you have a vehicle of any kind, you are trusting the people to have built that properly. If you're going through a tunnel or over a bridge, you are trusting engineers to have done the right thing. More importantly, if something goes wrong, like a tunnel or bridge collapses, you know that the people who didn't do their job properly will be held accountable because of the ethics that they adhered to.

This all became important to my wife and me, almost exactly four years, now. Our son was born six weeks premature. Instead of the birth we had planned, this perfect illusion of what things should be, his first few hours involved me running through the hospital with him in an incubator, and a barrage of tests, IVs, and medical interventions to make sure he would stay alive.

My memory of that time is foggy, because it was intense and went on for weeks, but there is a particular memory that sticks with me. The head of the NICU unit said to my wife and I, "Your son is fine. We will do everything to make sure your son will have the capability to live a full life." My wife said, "Do what you think is right. We trust you completely."

We lend our trust to people we don't know because we know that they are packaged — all their behavior is packaged in ethics.

Today, I want to talk about practical ethics for the modern Web designer. Trust is becoming more and more important, and trust is also eroding more and more. The most high-stakes trust interaction you can do as a person is in your interactions with medical professionals. If you go to the hospital and you or your family member is violently ill, you are going to a building with people inside, and saying, "Hello. Do what you think is best. I trust you completely."

A friend had a severe heart murmur, and she asked me to go with her to the hospital. The doctor said the best way to deal with this is to turn off the heart and turn it on again. Imagine the trust you must have.

Today, because of what's happening in the world, these trust interactions are becoming more and more important, because we are surrounded by this pandemic that is hard to understand by those who work with it on a day-to-day basis. The trust of society is eroding, not because they are doing a bad job or because they need to be questioned, but because information is fraught with politicization, and that we no longer know what the information is and whether it is accurate. A lot of that has to do with what we do as Web practitioners.

Why do we trust perfect strangers? When it comes to professionals, we trust them because of ethics. In the professional setting, ethics provides the necessary accountability to say, "I know you will do the right thing. I also know, if you don't, there will be consequences for you." Ethics leads to accountability, leads to trust.

You see this in the medical profession. We talk about the Hippocratic Oath, the "do no harm" clause. The current Hippocratic Oath does not involve the term harm, because it is undefinable. Doctors and other medical professionals adhere to the principle of doing no harm.

That becomes interesting when you talk to doctors about your own work. When we were in the NICU, I think one of the nurses heard me talking to a colleague on the computer. She asked me, "You know a lot about the Internet. How do social-media companies make money?"

Imagine there is a blue monster that says, "Come here, data cow. My algorithms will find all your biases, stoke them to make you addicted to my service, and I will turn that addiction into money."

What happens on the Internet is not covered by the same type of ethics as everything else. I can tell her, "I trust you completely, because you have the education and ethics background to know what is right or wrong and to do the right thing." If something goes wrong, there is a system in place. When you come to me with your private information, you have no guarantee that the people who built the tools to handle that information are going to do the right thing. There is also no accountability built into the system, at all.

The trust we have been coasting on, as Web practitioners, for the past 30 years has been eroding badly. The people who built tools for realized we can't be trusted, or, rather, there is no reason to trust us, because there is no system in place to protect them. So, they go to their elected officials and say, "You need to deal with this," and those elected officials go to lobbyists. Corporations say, "Give us all the power, and we will do the work for you."

We must retain control of our work, and hold ourselves to an ethical standard, so we can be accountable to our own actions and people can trust us.

I want to give you an overview of what ethics is. Here is the Moral Philosophy 101, pandemic education.

There are two terms we use all the time, morals and ethics. Morals are the internal judgments that we have ourselves. Ethics are commonly agreed-upon . . . Morals are personal, internal, individual. Ethics are external, community-based, and societal.

Ethics is misunderstood as rules and regulations, or ways to avoiding risk and legal issues. You can use ethics to avoid these things, but ethics are not universal. Universal are framed and defined by the communities they are defined to, and influenced by the morality of that group.

We see this in masking. Some people see masking as a moral responsibility to our community, some see them as a moral responsibility to safety . . .

As a result, we see different types of ethics establish themselves in different societies. In some regions, masking is recommended, and the vast majority wear masks to project other people.

In other regions, masking is mandatory. The ethics of those regions say that society has an obligation to protect everyone.

In some regions, masking is voluntary. The ethics in those regions put personal freedom first, and see masking of the populace as unethical.

This is why ethics is complicated. Ethics is not this higher-order, "we know what is right or wrong." Ethics is a communal practice reflecting the community. We establish these ethics in our community.

How do we as individual moral actors establish and agree upon a universal framework. How do we be consistent? How do we know what is right and what is right?

The answer is the same today as it was 5,000 years ago. Ethics is a practice. It is not a thing that you can see or do. It's a way of living.

Ethics is also about being and doing.

So, how do we be and do things in an ethical way as the Web community? I think the answers lies in rediscovering what the Web was about, and root those answers in principles.

The Web links information to anyone with an Internet connection. Web design and development is really granting, enabling, and enhancing people's capabilities to share thoughts, creations, and ideas with the world. Web design is capability-centered design, meaning you need to see this in a sort of . . .

The reason for that is, human-centered design sees humans as instruments used by the designer to reach a goal. If the human is able to do that, through heuristics and other things, it's a good design. The problem is that it doesn't consider what happens to the user. Is it useful, practical for the person to do that thing, or does it just benefit the person who designed it?

Human-centered design is design plus ethics. Design is political. Every time I say this, people freak out. They say, "Stop putting politics into everything." Let me be clear, design as a political thing has been discussed for hundreds of years, and it's a well-established truth in the design community. You see it in writings. "Engineers need to think of their work as both a humble contribution to the social order . . . "

Here is Mike Monteiro from 2015.

Here is Mario Bunge from 1975, in "Towards a Technoethics."

And here, in 1962, the First Things First manifesto was published.

Take this statement and apply it to social-media discourse today, and you'll see that, in 1962, designers like us worried about the very things that designers like us built.

As designers, we make decisions for other people. That's what design is, constantly making decisions for other people and deciding what they can and cannot do. So, we need to it sure we do it right.

Take a deep breath. Take everything you know about design and ethics, put it in a box, close the box, and throw it away.

I want to introduce you to capability-centered design. This is a term I came up with. It takes the idea of ethics and splits it into two parts: being, to make ethical reflection part of your daily personal and professional practice; and doing, applying ethical principles to decision-making at every level of your practice, actively rooting every decision you make in ethical principles, and make better solutions that work for everyone.

In moral philosophy, there is a term called human flourishing, the North Star of ethics. If we do things that are ethical, we can ensure people will flourish.

Ethical Web design is work furthering human flourishing through ethical practice and methodology, centered on the rights, capabilities, and agency of the human end-user. That's what we'll work towards.

Being, first. Ethics is not a new science. Ethics has been around for thousands of years. Over time, there are three main theories that have established themselves as good guiding branches for how we think of ethics in the Western world: utilitarianism, virtue ethics, and duty ethics. We use these in conjunction with each other to take out each other's errors, and to make it into a holistic approach.

How do we use this in design and development? The first theory is what we use in technology and design today, utilitarianism, part of consequentialism. It judges the goodness and rightness of an act . . .

We find utilitarianism in our tools and language. Good design is design that benefits the majority.

There's an obvious problem with this, and that is, who defines the majority? And what about everyone else? This is the standard critique of utilitarianism.

We focus on the utility of a design, as in how much benefit the design gives to the people who use it, and then include those outside of the majority group, so everyone is included.

We can have captions, and no facial recognition. Facial recognition is technology used against people all over the world. Black lives are disproportionately attacked by facial recognition.

Next up is duty ethics, or deontology, which judges the goodness and rightness of an act based on whether the actor acted in accordance to the person acted upon and to the community.

"Act only in accordance with that moxie through which you can at the same time will that it become universal law." That means, act in the same way you would want everyone else to act in the same position, kind of like the Golden Principle.

When you make a decision, you set a principle for other people to follow. And, you judge the rightness and goodness on whether or not you can say you want everyone else to do this.

Edward Snowden wrote about this in a book.

This is important, because this tells us, when we do things, we're doing things on behalf of everyone.

The challenge with duty ethics is, there's a problem of ought and can. The fact that we ought to do something doesn't mean that we can. You might work for a company that doesn't care about privacy, and you have no power to change the privacy settings. However, it is your responsibility to make sure you raise this issue.

Duty ethics is based on values, and values are tricky. The guy who published 3D models for guns on the Internet thought he was insisting on something everybody wants him to do. The problem is that his values don't conform with the values of the general society.

We need to focus on our shared principles, which means to figure them out, and then establish best practices for everyone to follow. Is this a decision I would want everyone in the same situation to make? And would it result in everyone flourishing?

Look at the electronic scooters littering the sidewalks. If one person does it, it's fine, but if everyone does it, it's a problem.

No to cryptominers, because they will literally destroy the Internet.

Duty ethics for design is focusing on shared principles.

The last of the three of virtue ethics. Duty ethics look at the character of the person doing the thing. What happens to the person once they do the thing? An action is good and right if it is done in a way that is virtuous. This is infuriatingly circular. We can thank Aristotle for that definition.

A virtue is a property defining and describing a person.

Shannon Vallor, who works at Santa Clara University, has set up new virtues for us: honesty, self-control, humility, justice, courage, empathy, care, civility, flexibility, perspective, magnanimity, technomoral wisdom.

Is a person courageous? Empathic? Do they have care? Are they magnanimous? As long as your actions push you towards a person having these virtues, you're doing the right thing.

But how do we agree on shared virtues as a profession? How do to make sure this virtue approach doesn't lead to a cage of conservatism? Aristotle thought a tree was a perfect example of a tree. We need to figure out how to define virtues. This starts with listing the virtues, then challenging our ideas, ideals, and best practices.

All the oaths you're seeing around technology right now, that is the start of defining virtues. These are value statements based on virtues that people want to become. These are the end result of a conversation we haven't had yet. We need to roll it back. But the work put into the projects is very important.

The other part is that it needs to challenge our ideals and best practices. Why? If we aim to make ethical decisions about Web design, our virtues need to reflect the diversity of our community and audience. This means making everyone first-rate members of our teams, our processes, our work, and everything we do.

Absolutely yes to blocking hateful content online, because hateful speeches have consequences. And, no to surveillance capitalism. Buy this book and read it.

We have the three branches — utilitarianism, virtue ethics, and duty ethics. However, there is a fourth quadrant, the users themselves. This is the capability approach, developed in the 1970s and still being developed. It judges the goodness and rightness of an act that entails two core normative claims.

Etch that last sentence into your head. You want people to do and be what they have reason to value.

Action is right and good if it grants and enables those abilities in people.

Think of a bicycle. It might seem a good idea to give everyone a bicycle, but different people need different tools to achieve the same goal. We have to design solutions that work for the people in their own context. If we just applied the same solution to everyone, we would not help everyone.

In my definition of ethical Web design, I said this. That's what this is. This circular approach here is what I call the four-corners approach, a way of looking at a design decision to question whether it is a good decision.

Start by asking about the capabilities. If we are, ask: what is the utility of this decision on everyone who is affected? Then, go to duty ethics and ask: are we setting the right practices and principles by doing this? Then, ask, is this a decision a person with the virtues I hold would do?

That's the being part. And yes, this stuff is really heavy. That's what I was saying. Ethics is a practice, in the same way that meditation is a practice. You have to practice this for a long time, internalize this, question everything around you in these terms, to be able to move forward.

It's important because we work in a difficult environment.

People always say, "I can't quit my job because it is ruining society. I have a family."

Marcotte suggests we create a union.

It's also how we build trust. We need to build trust in ourselves for other people to interact with our products. Like I said, we have work to do.

You see all that and you go, that's heavy, theoretical, and weird. I need something more practical. The doing part is practical.

I want to introduce you to the core capability framework. It breaks down into how you do everything in your regular work life. If you think of capabilities, you actually have four layers in your work. You have the core capability, what you want the user to be able to do after interacting with the design. Then you have the material capability, lent by the tool.

What happened? I lost control over my slides.

Then you have the meta capabilities. Finally, you have the environmental capabilities.

To see how this works, I've come up with a practical example. Say someone puts this on your table. You all work for higher ed. The university needs to communicate information about safety for the fall. The information must be accessible and understandable by all. This is a life-or-death situation.

You must develop a Web portal that might mean the difference between life and death.

How do you approach this in an ethical work? Start by making sure the capability we intend to grant or enable in the user through our design . . . This is the true north of your design. You must successfully transfer the capabilities to your user.

The core capability asks: what new capability does the user have after interacting with the design? Here, the information must be accessible and understandable by all students and faculty. Then, we can look at how we're doing this, and we can measure the outcomes based on these two things. Do people understand the information, and can they access it?

Then there is the material capability. Paper has material capabilities. You can fold it, transport it, write on it, give it to other people, tear it. Are these capabilities going to help us further the core capabilities? If so, use them.

An HTML document gives information to capability of being sent over the Internet, of being parsed, which means auto-translation and all of this stuff. We can use HTML to enhance the core capabilities.

In the case of our project, you send out paper mail, e-mail PDFs, and caption videos, to cover all bases.

Then there are meta capabilities, afforded by the material. They are latent, and we can add them or hook them into the project. In HTML, this might mean hyperlinks. Everything HTML tag is a meta capability, because they add information to the data, but they don't have to be there. Alt-text is an example of a meta capability. Does adding this information make the information more accessible and understandable? Captions are another example.

Meta capabilities must be appropriate, and they must not be dependencies. The fact that you used them does not mean people must have access to them.

Meta capabilities might be correct links, captions, stuff like that.

Lastly, there are environmental capabilities, provided by the users' environments. These are out of the control of the designer. These include browser features, anything the user uses that changes the content you published.

We don't know whether the user has these capabilities, or if they know about them if they're available.

We need to think about progressive enhancement. Lay attributes to the documents, so auto-translation kicks in. You can look at capabilities added by other apps, like social-media sharing, screenshots, and reposts.

You can say that we need to provide easily sharable and Instagrammable images with info. Create social-media videos with burn-in captions, because people like these over a boring website.

That is the core capability framework. This is directly applicable to what you do. You can use this framework to argue for accessibility and to inform more core practices. All of this goes down to the capabilities in the end-user. Every decision we make as designers must anchor to those. Then we can ask about the ethics of those capabilities, whether they are the capabilities that people need to be and do what they reason to value.

Being is the four-corners approach. Doing is the core-capability framework. These work together to allow you to have a toolkit for tackling ethics smells, a jokey term to talk about those things where you think there's an ethical problem but you don't know what they are. You will quickly discover the ethics smells, and be able to work toward solutions.

Ethics leads to accountability, which is what allows people to trust us as practitioners and, eventually, professionals. "Trust is the currency of interactions." I know this because I have a kid. I can see how, when I interact with him, and the way I interact with him, directly influences his trust in me. The more carefully I interact with him, and the more I show he can trust me, the more he is willing to do with me, like use me as a climbing structure.

Someone says, "But, Morten, there is always someone more bullied, enterprising, etc., enough to do the right thing. Even if I don't do it, someone else will." This is true. This is a problem of evil, a problem that no one can solve. But we can make the problem of evil less relevant, by being humble, accountable, self-aware, and graceful to ourselves and our peers.

Ethics is a practice. Ethics is what allows us to hold ourselves accountable to our desire to shape the world to our vision. Because, with every design decision, we build the future for our users and for ourselves. Build the future. That's the work.

Let's believe we are compassionate, caring designers and developers who care about the people we work for and the work we do, and let's build a better environment for a better practice. Thank you.