A Brief Guide to Ethical Application Design
Five Steps to Making Better Software
Overview
By nature, I create lists when I work. Interesting links, software tricks, and app ideas get funneled into all kinds of nooks and crannies.
As the industry matures past the wide-eyed, "apps for everything" stage, one list has really taken off: Dos and don'ts for designing good—as in, ethical—software.
Here, I maintain a very abbreviated version. This is not complete, dogmatic, nor even right. Instead, it's written as a thoughtful reference when reviewing projects or business models, summarized as 5 key points:
- Minimal Data for Minimal Time
- Protect the Data You Keep
- Broker a Fair Trade
- Engagement is [Probably] a False Positive
- No Dark Patterns
Partially, this list is a way for me to encapsulate my own thoughts. It's also a response to the community's growing frustration that we may have created a world that's worse than before. After all, it's not every day that Dieter Rams vicariously lambasts Apple and Facebook's design teams: "All this stuff you are creating is terrible!"
Use this list, debate it, reflect on it, ask me about it. Joel Califa's Subverted Design and Thomas Otto's Design Human Data are also excellent resources on the subject.
1. Minimal Data for Minimal Time
Mo Bitar, creator of Standard Notes, makes a fantastic observation that user data isn't an asset, it's a liability. You are the steward of every byte and are responsible for protecting what you collect (ouch).
A trove of data is a trove of risk. Store only what you need; destroy it as soon as possible. Ethical applications reduce their exposure and honor their users by not holding their data in the first place.
2. Protect the Data You Keep
After minimizing, protect what's left. Is it encrypted? Can employees see it? The dev team? Data is asymmetrically expensive to secure, nobody makes the time, and insurance hates paying for negligence.
Data (even anonymized) create surprise collateral damage. Every byte is a delicate game for both sides. Ethical applications minimize the part they play by thoughtfully securing their holdings.
3. Broker a Fair Trade
Sometimes, data is awesome! Spell out why and establish trust and consent with your users. Good UX is key to getting your story across. This takes effort—but that's the point. Their data is theirs. If it's valuable to you, prove that you offer equivalent value in return.
Ethical applications broker trust by respecting their users, playing fair, and spelling out their intent. They give users the opportunity to meaningfully weigh the tradeoffs and make an informed choice.
4. Engagement is [Probably] a False Positive
Engagement metrics—clicks, sessions, etc.—can be misleading. A focus on bumping the numbers hijacks iterative design by giving it a false measure of success. It's easy to confuse better clickbait for value provided.
Never forget that behind each number is a real, live human. Did that new feature enrich lives or train (or addict) them instead? Ethical applications take a nuanced approach to data-driven UX assessments and avoid metrics that encourage stealing the user's attention.
5. No Dark Patterns
Never use design patterns that coerce or manipulate the user. Do not intentionally inject your application into someone's flow of life. Never assume that they understand the implications of their actions.
Our ability to manipulate has vastly outpaced our awareness and maturity. Ethical applications acknowledge that the playing field isn't level and provide utility without driving behavior.