The Use of Configuration Files to Modify and Store Form Fields; The Paradox of Trust: Understanding Continued Social Media Use Amid Data Privacy Concerns

My technical capstone project was targeted at modernizing an outdated Data Manager, a New York-based investment banking firm used for client onboarding. Being slow, rigid, and also maintenance-heavy, the previous system had become reliant on outdated frameworks. The firm's agility was in fact limited in adapting to new client requirements and fields, due to the fact that important developer effort was required for even minor updates to form fields or database structures. I targeted to dynamically scale and to develop a friendly solution that can reduce turnaround time, can improve performance, and allow for smooth future updates.

I addressed these limitations by design and via implementation of a configuration-driven web application. The application uses SpringBoot, Java, and JDBC to construct the backend in addition to React together with TypeScript to create the frontend. The Configuration JSON file, for storing metadata about each form field and about its associated database mappings, was in fact the heart of the system. This file detailed all required information to be able to dynamically render form fields on the front end and dynamically run CRUD SQL operations for the backend.

A transformative outcome of this system was the reduction in development and maintenance time for the application. For each database or field addition, the developers did not have to modify frontend as well as backend logic. Instead, the JSON configuration remained a sole source of truth, thus empowering teams toward a focus upon functionality and performance, rather than change code repetitively. Additionally, performance for the web-page improved as a result of the shift away from legacy frameworks as there was lower latency observed in operations regarding larger datasets.

However, the project can be optimized even more. There are two main areas for future work: automated testing and accessibility enhancements. While the core features are functional and efficient, it is necessary to test the application before deploying it to production. Also, allowing non-technical users to create new JSON configurations directly from the frontend would make the tool even more accessible. In the end, through this project I was able to gain practical experience about software design and real-world implications for improving operational efficiency in an enterprise environment.

For the Prospectus, contrasting the technical focus of my capstone project, my STS paper explores the social, psychological, and structural dimensions of user trust in social media even after knowing about data. In this paper, I explore the following question: why do users continue to engage with platforms like Facebook, Instagram, and TikTok despite widespread awareness of their exploitative data practices?

My motivation behind this paper was due to events such as the Cambridge Analytica scandal, which revealed that Facebook had allowed the unauthorized collection of psychological data from 87 million users. This was an international level scandal, however user behaviour on these platforms showed very little change. The contradiction about the user's knowledge of the scandal, yet continuing to engage in these medias is the focal point of my investigation.

My methods for research included survey analysis, discourse on Reddit, and case studies of regulatory and corporate behavior. Through all of this research, I concluded that users have an illusion of free will on social media and it is shaped by three factors: psychological, social, and structural. From a psychological standpoint, platforms exploit dopamine-driven behavior loops to keep users engaged. On a social level, users fear missing out on the activity that happens on social media, making leaving the platforms much harder. Last, structurally the Big Tech monopoly structure limits the amount of genuine alternatives for users to turn to, as users have to sacrifice functionality or connectivity.

To analyze the research that I did, I used the STS theoretical frameworks of intersectionality and organizational power. Intersectionality specifically helped me analyze and explore how privacy violations are experienced differently across demographics. I learned how marginalized groups feel especially vulnerable, but dependent on these platforms for visibility. Organizational power theory helped me analyze how platform designs are made to control user behavior and that these structures normalize surveillance and the urgency of users to leave the platforms.

The main conclusion that I reached was that continued use of social media platforms does not necessarily indicate user trust, but it reflects the flaws in the system and of people’s resignation, addiction, and lack of alternatives. This raises the question of what should be done? To get justice for this situation, broader questions about democratic regulations, monopolistic behaviors, and the ethical responsibilities of these social media platform designers and engineers need to be asked.

School of Engineering and Applied Science

Bachelor of Science in Computer Science

Technical Advisor: Briana Morrison, Rosanne Vrugtman

STS Advisor: Karina Rider