Student work

Challenge

The rise of sensor-based technology and Internet of Things (IoT) has resulted in increased public interest in personal data security and privacy. Companies now have a greater moral obligation to educate consumers about the implications of using such technologies.

Bosch tasked my team to explore how their proposed data management model might interface with their growing smart home infrastructure — all while letting users retain full control over their personal information.

Process

Back in 2015, smart home technologies weren’t as mainstream as today. Our research strategy aimed to:

1. explore people’s understanding of the emerging home automation landscape.
2. assess comprehension of how personal data is used and stored.

From left to right: Eric, Jess, and Jeel facilitating card sorting x business origami research sessions

Key takeaways

Desire to see and manage data through physical means

Participants often expressed the need to be able to “hold” and manipulate data with their hands. They wanted something tangible that they could interact with, such as smartphone or tablet UI. It was these interactions with physical devices that reinforced their perceived sense of control.

Fear of system replacing humans

A common concern was that as sensors become more sophisticated, they have the harmful side effect of influencing how humans interact with one another. Participants didn't want their smart home to undermine their interpersonal relationships nor diminish the quality, human-to-human connections they experienced on a regular basis.

Access to an "off" button

The idea of living with a context-aware smart home system alarmed some participants. One individual exclaimed, "If I am living in a smart home, I would like a dark space where I know there is no sensing." They wanted the ability to turn off all data collection functionality in an instant.

Android app prototype

With Kin, users can:

• manage devices and data.
• configure routines.
• monitor their smart home remotely.
• share access and permissions with others.

Kin facilitates communication between a smart home and its devices, allowing users to set up automations called "Routines". A routine consists of at least one trigger and a corresponding response action. For example, a user might create a routine to send a push notification to their phone when the smart fridge senses that its water filter needs to be replaced.

Moreover, Kin lets users to monitor home activity and manage access from afar. Homeowners can create guest profiles with expiring user permissions.  

Process

To start, we explored several other problem spaces involving vehicles before making the aging population our focus. As smart cars increase in popularity coupled with a growing senior population, we saw an opportunity to investigate the relationship between the two.

To gain a better understanding of this domain, we spoke with caretakers who had experience trying to dissuade their parents from driving. A common theme that emerged was the importance of preserving an elderly person’s mental health after revoking their driving privileges. Participants emphasized how important the physical car keys meant to the seniors in their care — these were literally the “keys” to their independence.

Storyboard scenarios

Our conversations with caretakers inspired storyboards centered around four themes that were evaluated via a speed dating research exercise. We ran these storyboards by mix of caretakers and individuals who had close relationships with elderly people to get feedback on our proposed directions.

Generally, the idea of owning a smart car that monitored driving habits, provided continuous constructive feedback, and assumed control in unsafe conditions was well-received.

Participants also suggested extending this functionality beyond the senior population. They wanted a semi-autonomous car that could evaluate a driver’s performance sooner — before their abilities deteriorated. Starting early has the added benefit of tracking patterns and trends over a longer period of time and this data could be used to create a personalized transition plan later in life.

User enactment and outcomes

We used our collected feedback to devise three futuristic smart car scenarios:

1. Unsafe driving episodes
2. Trek mobile app x having to take away car keys
3. Trek taking control of the vehicle in unsafe driving conditions

I sat in the passenger's seat while Maggie (bottom) provided HUD notifications and Joyce (right) simulated haptic feedback.

Our user enactment setup consisted of a mock smart car outfitted with a fabricated steering wheel, front seats, and windshield. Participants assumed the role of an elderly driver while my teammates simulated the car’s functions (e.g. HUD) and acted as the senior’s caretaker. I moderated each session and walked participants through a couple of scenarios.