Our Digital Many Selves

Intention

This project started with the questions like "What types of data do digital platforms acquire from us?" and "What kinds of assumptions do those platforms make about their users?" 

I was curious about the unexpected or unrevealed self that gets presented through our personal data on different digital platforms, and how these data traces portray ourselves and lead our digital experiences. I started by looking into my own data from Instagram, Twitter and Amazon, some information is somewhat accurate, but some is definitely way off.

This interface attempts to reveal the contrast in a disturbing way. It invites people to reflect upon the unseen traces and consider how data has affected their digital experiences.

Outcome/Prototype

Borrowing from Twitter's mobile interface, I created a fake Twitter feed page with my ads preference data from Instagram, Twitter and Amazon. The interface is displayed on a P5.js canvas. The displayed Twitter texts are pulled real time through the API working with Codebird.js. The searched keywords are based on different keyword groups of the three platforms. The page can be regenerated by pressing the "n" key to restart the search event.

Here is the link to the prototype. (Press "n" for regenerating new feed. The library may be a bit slow so please allow it to process for a while)

Precedents

My idea for this project started with the concept of a "digital doppelgänger", a digital duplicate based on data about our preferences and behaviors. It is our personal data collected by digital platforms for making our experiences more convenient and personalized. This term and the impact of our data traces lead me to a project called Data Portraits created by Giorgia Lupi for TED 2017. Presented in the form of abstract drawings, it is an innovative way of connecting personal data with the conference attendees. This project shows the humanistic side of data, and invites users to construct their portraits by themselves through the guiding questions. As an opposite approach, I attempted to present the data the other way around, from the perspectives of the platforms' observations.

Process

Initially I wanted to present the data in a tangible form and imagine it to be in the form of a device. But in the process of researching about precedents and the technology to implement it I switched to the webpage form. The first iteration only focused on generating and designing different user profile images, to add more detail and insights to the data I added in the fake Twitter feed later on.

Open Questions and Next Steps

Currently the interface relies on my personal data requested from the selected platforms and manually downloaded and coded to the files. In the future if this project is to be put out as an interactive work for multiple users, I want to add in features that can automatically pull data from the platforms through the APIs. The interface will also be redesigned with more instructions guiding the users to navigate around.

The "spookiness" of this project mainly comes from the unsettling feeling of being constantly surveilled and the direct confrontation with personal data through this interactive form. I am curious about how other people will react towards their data, and whether knowing about the insights will cause any changes in their behaviors on digital platforms.

Personal Reflection

Reflect on making this project. What did you learn? What would you do differently? Did you get where you wanted to? If not, why not? What do you need to get there, etc?

Reflecting on the making process, I think I would add in more interaction to the display and refine the design to have a more consistent aesthetic. The form is different from my initial expectation, but I still learned about using the Twitter API with P5.JS.  

Attribution and References

JS libraries: P5.js and Codebird.js