Abstract
University of Virginia students and Charlottesville residents regularly need help with short-term tasks—moving furniture, running errands, pet sitting, assembling items—yet no trusted platform exists to connect those who need help with those willing to provide it. Current solutions remain scattered across Facebook groups, listservs, and word of mouth, none of which offer verified identities, secure payments, or accountability. National data shows that millions of adults earn income through short-term tasks and that a majority of undergraduates work during college, suggesting both strong demand and a willing labor supply. Yet platforms like Facebook Marketplace exclude local pickup transactions from purchase protections, leaving users to arrange informal work without recourse. To address this gap, my technical team developed Hoos Helping, a web application designed for the UVA and Charlottesville community that functions as a hyperlocal task marketplace. Users authenticate through passwordless email login, create requester or helper profiles, and post tasks with structured details including descriptions, budgets, deadlines, and locations. The system incorporates affiliation-aware trust signals by distinguishing UVA-verified accounts from non-university accounts and supporting optional background verification for helpers. Core features include task browsing with filtering, a task acceptance and completion workflow, ratings and reviews, and integrated payment processing. An administrative dashboard enables platform oversight for safety and quality assurance. Beta testing with UVA students validated that the core task lifecycle functioned correctly and that the platform felt relevant and useful, with affiliation indicators and structured workflows increasing confidence compared to informal alternatives.
While Hoos Helping deliberately requires identity verification to establish trust, not all digital platforms benefit from such transparency. Anonymity tools like Tor—The Onion Router, an open-source network that encrypts and routes internet traffic through thousands of volunteer-operated servers to make user activities extremely difficult to trace—serve fundamentally different purposes by protecting journalists in authoritarian regimes, activists organizing under surveillance, and ordinary citizens seeking privacy from commercial tracking. These two technologies occupy opposite ends of a critical spectrum: systems where verified identity enables safe transactions versus systems where anonymity protects vulnerable populations. Understanding how technical architecture negotiates this tension reveals that design decisions are never purely technical—they carry political implications that shape what becomes possible for users and institutions alike. My research paper, grounded in Science and Technology Studies (STS)—a field examining how social, political, and cultural values shape and are shaped by technological systems—investigates how Tor's design decisions function as political interventions that actively construct the boundaries between legitimate privacy rights and state security interests. The theoretical framework draws on Langdon Winner's (1980) concept of technological politics, the argument that artifacts can embody specific forms of power through design choices among viable alternatives, and on Collier and Stewart's (2022) empirical research documenting competing "privacy worlds" within Tor's development community. Their work identified three worldviews shaping Tor's design: engineers who view privacy as a technical problem solvable through cryptography, maintainers who understand privacy as charitable service delivery requiring sustainable operations, and activists who frame privacy as a political struggle against surveillance. These worldviews create persistent tensions around protocol decisions that cannot be resolved through purely technical analysis. To investigate how these dynamics shaped Tor's evolution, I conducted document analysis of primary sources across three phases: hidden services emergence enabling anonymous marketplaces (2004–2013), post-Snowden security hardening against state-level adversaries (2013–2017), and the censorship circumvention era where Tor became critical infrastructure for activists facing government blocking (2017–present). Sources included developer mailing list archives where engineers debated value tradeoffs in specific technical proposals, government policy documents framing anonymity as an investigative obstacle, and advocacy reports documenting civil liberties perspectives. This analysis reveals a recursive, bidirectional relationship between technical design and political contestation—design decisions do not merely respond to privacy threats but actively construct what privacy means and which threats matter most, provoking state responses that motivate further technical adaptation. Considered together, these projects illuminate a broader principle for engineers and policymakers: context determines whether transparency or anonymity better serves human autonomy. Hoos Helping demonstrates that users willingly trade some privacy for accountability when the platform provides sufficient trust infrastructure, while Tor's history shows that the same architectural features protecting political dissidents simultaneously enable illicit activity, with no technical modification capable of satisfying all stakeholders. Engineers can use findings from both projects to make more deliberate decisions about when verified identity versus anonymity best serves their users, and policymakers can develop more nuanced approaches than blanket encryption restrictions, preserving privacy for vulnerable populations while addressing legitimate security concerns.
