Exploring Gray Patterns: How can we tell if design is truly good?
Our class at AC4D has spent the past few weeks studying ethical quandaries in design. I’ve become particularly interested in dark patterns, which can be defined as tricks used in websites and apps that make you do things that you didn’t mean to, like buying or signing up for something (courtesy darkpatterns.org). There are many types of dark patterns: straightforward ones like Hidden Costs and Disguised Ads as well as more humorous ones like the Roach Motel (when you’re trapped in a web of pages designed to get you to buy something) and Privacy Zuckering (when you’re tricked into sharing information publicly about yourself). The Internet is awash in dark patterns; one example can be found below.
If we are to be proactive in creating good design, we must do more than simply avoid using dark patterns—we must create light patterns. We can sum up the difference between dark and light patterns as follows:
- Operate insidiously
- Remove agency
- Encourage negative behavior
- Operate transparently
- Promote agency
- Encourage positive behavior
It is easy to define light patterns by simply reversing the definition of dark patterns. However, it is much harder to identify light patterns than dark patterns. For example, consider this request I received from the language-study app Duolingo:
On first glance, I would characterize this as a light pattern. But does it meet the criteria identified above? It’s promoting positive behavior, but it’s not exactly transparent—how does the notification system work? Am I agreeing to set up a regimented reminder or allowing Duolingo to send me unscheduled nudges and notifications? Do I want that kind of needling from an app? And what’s motivating this request—genuine interest in my education or a desire for more user minutes and more ad revenue? Does it matter?
How can we tell if design is truly good?
I don’t ask this question wondering whether design is effective, easy to navigate, or aesthetically pleasing. I ask this in an ethical sense. If we want to create design for the social good (and I do), we must understand what the social good entails. We must also understand the ways that design functions and its ramifications on both individual and social levels.
It is hard to identify light patterns in design because the above criteria are ill-defined. Consequently, when we set out to create light patterns, we often create what I identify as gray patterns. Gray patterns demonstrate the following characteristics:
- Can be transparent or hidden
- Aim to improve your experience
- Promote ongoing behavior
Unlike dark patterns, gray patterns aim to improve your experience. They can be transparent or hidden (the example above is a bit of both), and they often promote further engagement with whatever program or service you are using. This can be good or bad, depending on each individual user.
Consider YouTube’s endless playlists. I often use YouTube to listen to music that I cannot find on Spotify, and YouTube’s algorithm has become very adept at finding music that I am likely to enjoy. I am glad that I can put a song on and let it play forever, never worrying about choosing the next song. This saves me time and energy and introduces me to music I would not otherwise discover.
YouTube’s “Endless” playlists are great for discovering new music, but can lead to mental and social disengagement.
However, this mechanism can have negative ramifications. If I’m on break from work and I put on a short comedy clip, the endless play of videos will tempt me into watching longer, distracting me from my job. It can promote binge-watching, eating up long periods of time and encouraging users to mentally disengage. This mechanism is effective at getting users to spend more time on YouTube. But are the effects good for users, or for society?
Gray patterns result when designers aim to improve a user’s experience without thinking through the ramifications. To help determine the difference between dark, light, and gray patterns, I’ve created the following schematic:
Design Pattern Observation Process
Again, differentiating between gray and light patterns requires an understanding of the term good. This conversation requires serious discussion, as there are many viewpoints that should be considered. A quick attempt at identifying “good” design criteria might look like this:
Criteria to consider when designing for the social good.
I am sure there are many more criteria that I am leaving out. And even if we were to agree on a broad set of criteria to consider, implementing design to achieve these outcomes will be a significant challenge. It will require us to continually test our ideas for unintended impacts, as it is impossible to predict how the implementation of new technology will affect individual and collective behavior.
However, identifying the problem will allow us to iterate toward a solution. With this burgeoning framework in hand, I hope that I will learn to better differentiate between light and gray patterns and better design for social good.