Quiz: Designing Data Futures

Test your understanding before moving to the next chapter. Target: 70% or higher to proceed.

Section 1: Multiple Choice (1 point each)

1. The "participation deficit" in data governance refers to:

  • A) The lack of available technology for participating in governance
  • B) The gap between the democratic ideal that people should have a say in the rules that govern them and the institutional reality that data governance is designed by experts behind closed doors
  • C) Low voter turnout in elections related to technology policy
  • D) The shortage of qualified data governance professionals
Answer **B)** The gap between the democratic ideal that people should have a say in the rules that govern them and the institutional reality that data governance is designed by experts behind closed doors. *Explanation:* Section 39.1 defines the participation deficit as a specific instance of the power asymmetry theme. Data governance is designed by technical and legal experts — lobbyists, corporate lawyers, government officials, and academic consultants — while the people most affected by governance decisions (data subjects, communities subjected to algorithmic decision-making, workers managed by data systems) are largely absent from the design process.

2. A data cooperative differs from a data trust primarily in:

  • A) A cooperative holds data temporarily while a trust holds it permanently
  • B) A cooperative is governed democratically by its members (one member, one vote), while a trust is governed by a fiduciary who makes decisions on behalf of beneficiaries
  • C) A cooperative is for-profit while a trust is non-profit
  • D) A cooperative operates only at the local level while a trust operates nationally
Answer **B)** A cooperative is governed democratically by its members (one member, one vote), while a trust is governed by a fiduciary who makes decisions on behalf of beneficiaries. *Explanation:* Section 39.2 identifies the key difference between cooperatives and trusts as the governance structure. In a cooperative, members govern themselves — they vote on how data is used, who it is shared with, and how revenue is distributed. In a trust, a trustee exercises fiduciary judgment on behalf of beneficiaries who do not directly participate in governance decisions. The cooperative model is more democratic but requires active participation; the trust model is more scalable but concentrates governance authority.

3. Elinor Ostrom's work on governing commons is relevant to data governance because:

  • A) She proved that all commons should be privatized
  • B) She demonstrated that communities can successfully govern shared resources through self-organization, identifying eight principles for sustainable commons governance
  • C) She developed the first data protection regulation
  • D) She showed that government regulation is always superior to community governance
Answer **B)** She demonstrated that communities can successfully govern shared resources through self-organization, identifying eight principles for sustainable commons governance. *Explanation:* Section 39.2 introduces Ostrom's work (for which she received the Nobel Prize in Economics in 2009) as the theoretical foundation for data commons governance. Ostrom's eight principles — including clearly defined boundaries, collective-choice arrangements, monitoring, graduated sanctions, and conflict resolution — provide a framework for governing data as a shared resource without requiring either privatization or state control.

4. The Citizens' Biometrics Council convened by the Ada Lovelace Institute (UK) is an example of:

  • A) A corporate advisory board
  • B) A citizen assembly specifically focused on a data governance question — biometric technologies
  • C) A government regulatory body
  • D) A technology industry lobby group
Answer **B)** A citizen assembly specifically focused on a data governance question — biometric technologies. *Explanation:* Section 39.3 describes the Citizens' Biometrics Council as a real-world example of a citizen assembly applied to data governance. Fifty randomly selected UK residents spent two months learning about biometric technologies, hearing from experts and affected communities, and deliberating. Their recommendations — including a moratorium on live facial recognition in public spaces — were more protective than existing government policy, demonstrating that informed public deliberation takes data governance risks seriously.

5. In the GovernanceSimulator, the Corporate Centralized model distributes benefits primarily based on:

  • A) Each stakeholder's privacy preference
  • B) Each stakeholder's political influence (0.5 weight) and data contribution (0.3 weight)
  • C) Equal distribution among all stakeholders
  • D) Random allocation
Answer **B)** Each stakeholder's political influence (0.5 weight) and data contribution (0.3 weight). *Explanation:* Section 39.5 explains the Corporate Centralized model's formula: benefits are weighted 50% on political influence, 30% on data contribution, and 20% on technical literacy. This models the real-world dynamic in which corporate governance rewards those with pre-existing power (political influence) and commercially valuable data, while those with less power and less valuable data receive less benefit. Privacy protection under this model is inversely correlated with data value — the more valuable your data, the less privacy you get.

6. Speculative design, as used in data governance, is best described as:

  • A) Designing systems that speculate on financial markets
  • B) A methodology that creates concrete, detailed depictions of alternative futures to expand the governance imagination and make possible worlds tangible and debatable
  • C) Predicting which technologies will succeed in the marketplace
  • D) Writing science fiction about technology
Answer **B)** A methodology that creates concrete, detailed depictions of alternative futures to expand the governance imagination and make possible worlds tangible and debatable. *Explanation:* Section 39.4 describes speculative design as distinct from prediction (which forecasts what will happen) or science fiction (which is primarily literary). Speculative design creates artifacts — mock privacy policies, advertisements, news articles, policy documents — from a speculative future, making that future tangible enough to evaluate and debate. The purpose is not prediction but imagination: expanding the range of governance possibilities beyond incremental reform.

7. Prefigurative governance is the practice of:

  • A) Establishing governance frameworks before a technology is developed (same as anticipatory governance)
  • B) Building the governance structures you want to see in the future in the present — creating data cooperatives, organizing citizen assemblies, and demonstrating alternatives now rather than waiting for institutional change
  • C) Allowing technology companies to set governance standards before regulators intervene
  • D) Studying governance failures from the past to avoid repeating them
Answer **B)** Building the governance structures you want to see in the future in the present — creating data cooperatives, organizing citizen assemblies, and demonstrating alternatives now rather than waiting for institutional change. *Explanation:* Section 39.6 defines prefigurative governance as drawn from the political tradition of prefigurative politics — the idea that the means of change should embody the ends. If you want democratic data governance, you practice it now. Prefigurative governance differs from anticipatory governance (which focuses on building frameworks for future technologies) in that it focuses on building alternative *structures* in the present rather than frameworks for the future.

8. The Participatory Hybrid model in the GovernanceSimulator is distinctive because it:

  • A) Eliminates all coordination costs
  • B) Combines democratic participation (citizen assembly setting principles) with professional stewardship (reducing coordination costs) and explicitly redistributes benefits toward less powerful stakeholders
  • C) Gives all governance authority to government regulators
  • D) Maximizes total benefit regardless of distribution
Answer **B)** Combines democratic participation (citizen assembly setting principles) with professional stewardship (reducing coordination costs) and explicitly redistributes benefits toward less powerful stakeholders. *Explanation:* Section 39.5 explains that the Participatory Hybrid model addresses the limitations of both pure cooperative governance (high coordination costs) and pure regulatory governance (limited voice for affected communities). It uses an inverse-political-influence need weight to direct more benefits to less powerful stakeholders, sets a high privacy floor through citizen assembly, and uses professional stewardship to reduce coordination costs from 15% (cooperative) to 8% (hybrid).

Section 2: True/False with Justification (1 point each)

9. "The GovernanceSimulator's results prove that cooperative governance is objectively superior to corporate governance."

Answer **False.** *Explanation:* Section 39.5.3 explicitly warns against this interpretation. The simulation's results depend on its assumptions — the weight parameters, the coordination cost model, the noise levels, and the definition of "benefit." Different assumptions would produce different results. The simulation's value is not in proving one model superior but in demonstrating that governance structure shapes outcomes — and that debating the assumptions is itself a governance exercise.

10. "Citizen assemblies on data governance are impractical because ordinary citizens lack the technical knowledge to make informed decisions about complex data systems."

Answer **False.** *Explanation:* Section 39.3 directly addresses this objection. Citizen assemblies provide structured learning — expert presentations, Q&A sessions, facilitated discussion — that enables informed deliberation. The Ada Lovelace Institute's Citizens' Biometrics Council demonstrated that randomly selected citizens, given adequate time and information, produced recommendations that were technically informed, ethically nuanced, and more protective than existing government policy. The argument that "ordinary citizens can't understand data governance" is a version of the power asymmetry: it justifies excluding people from governance of systems that affect them.

11. "Data cooperatives and data trusts are theoretical models that have not been implemented in practice."

Answer **False.** *Explanation:* Section 39.2 provides multiple examples of operational implementations: MIDATA (Switzerland — health data cooperative), Driver's Seat Cooperative (US — ride-share driver data), Salus Coop (Barcelona — health data cooperative), and the Open Data Institute's data trust pilots (UK). These are functioning governance structures with real members, real data, and real governance decisions — not merely theoretical proposals.

12. "The Gini coefficient in the GovernanceSimulator is calculated using standard economic methodology and measures the inequality of benefit distribution across stakeholders."

Answer **True.** *Explanation:* The `_calculate_gini` method in the GovernanceSimulator implements the standard Gini coefficient formula, which measures inequality on a scale from 0 (perfect equality — every stakeholder receives the same benefit) to 1 (maximum inequality — one stakeholder receives everything). This is the same metric used in economics to measure income inequality, applied here to the distribution of governance benefits.

13. "Barcelona's data sovereignty strategy demonstrates that city-level data governance can assert control over how urban data is collected and used."

Answer **True.** *Explanation:* Case Study 1 examines Barcelona's data sovereignty strategy, which includes: requiring city contracts to include data sovereignty clauses, building city-owned data infrastructure, supporting data cooperatives (including Salus Coop), and establishing participatory governance processes for urban data. Barcelona's approach demonstrates that meaningful data sovereignty can be asserted at the municipal level — not just the national level — by using procurement power, public investment, and democratic governance.

Section 3: Short Answer (2 points each)

14. Explain why the chapter argues that the "participation deficit" produces worse governance, not just less democratic governance. Identify two specific mechanisms through which participation improves governance quality.

Answer The chapter argues that participation improves governance quality through at least four mechanisms, two of which are: First, local knowledge — people affected by data systems often understand their impacts better than designers. Eli's community understood the effects of predictive policing in ways no algorithm audit could capture. This local knowledge identifies governance gaps that expert-only processes miss. Second, blind spot correction — homogeneous design teams produce governance with homogeneous blind spots. Including diverse perspectives reveals assumptions that insiders cannot see. A governance framework designed only by technologists may not account for the experience of people with low digital literacy; one designed only by lawyers may not account for the technical reality of how data systems operate.

15. In the GovernanceSimulator, explain why the Corporate Centralized model provides the lowest privacy protection to stakeholders with the highest data contribution. What real-world dynamic does this model?

Answer In the Corporate Centralized model, privacy protection is calculated using the formula: `max(10, 100 - data_contribution * 0.6 - (100 - political_influence) * 0.3)`. This means that stakeholders with high data contribution receive less privacy protection because their data is commercially valuable — the corporation has a financial incentive to exploit it rather than protect it. This models a real-world dynamic: companies extract the most data from users who generate the most commercially valuable data (heavy users, users with high purchasing power, users in data-rich environments) while offering those users the least privacy protection. Users with less valuable data (elderly users, users in low-income communities) may paradoxically receive better privacy protection — not because the company values their privacy, but because their data isn't worth the effort to exploit.

16. What does the chapter mean by "hope as a political practice," and how does it differ from optimism?

Answer Section 39.7 distinguishes hope from optimism. Optimism is a prediction — the belief that things will turn out well. It requires no action because it assumes a positive outcome regardless. Hope, by contrast, is an orientation toward the future that combines recognition of the difficulty of the present with commitment to the possibility of change. Hope does not assume things will get better; it commits to working toward a future that is better, even without guarantees. As a political practice, hope means: building governance structures now (prefigurative governance) despite the uncertainty about whether they will scale; participating in democratic processes despite their limitations; investing in alternatives despite the power of incumbent systems. Hope is the motivational infrastructure of participatory governance — without it, the rational response to power asymmetry is resignation, not action.

Section 4: Scenario Analysis (3 points each)

17. A national government is considering how to govern agricultural data collected from millions of smallholder farmers by international agribusiness companies. Two proposals are on the table:

Proposal A: Require all agricultural data to be stored within the country (data localization) and establish a government agricultural data authority to manage access.

Proposal B: Support the creation of farmer data cooperatives, provide technical infrastructure for cooperative data management, and require that companies negotiate with cooperatives rather than individual farmers.

Evaluate both proposals using concepts from this chapter. Which better addresses the participation deficit? Which better addresses the power asymmetry? What are the risks of each?

Answer **Proposal A (Government Authority):** Addresses data sovereignty by keeping data within national borders. But it concentrates governance authority in the government — which may or may not represent farmers' interests. It does not address the participation deficit (farmers still have no voice in governance decisions). It risks creating a government data monopoly that could be exploited for surveillance or political purposes. It addresses one power asymmetry (corporation vs. country) while potentially creating another (government vs. farmers). **Proposal B (Farmer Cooperatives):** Directly addresses the participation deficit by giving farmers collective governance authority. Addresses the power asymmetry by pooling bargaining power. But it requires substantial investment in cooperative infrastructure, farmer education, and legal frameworks. It depends on farmers having the time, resources, and institutional support to participate meaningfully. And it may face resistance from both corporations (who prefer negotiating with individual farmers) and government (who may resist sharing governance authority with civil society). **Assessment:** Proposal B better addresses both the participation deficit and the power asymmetry, but it is harder to implement. An optimal approach might combine elements: government support for cooperative infrastructure (Proposal B's structure) with data sovereignty provisions (Proposal A's principle), while ensuring that governance authority is shared between cooperatives and government rather than concentrated in either. The Participatory Hybrid model from the GovernanceSimulator — combining democratic participation with professional stewardship — provides the closest analogue.

18. Using the GovernanceSimulator framework (you may reason through the logic without running code), consider a community with two stakeholders: a technology company (high data contribution: 90, high technical literacy: 95, high political influence: 85, low privacy preference: 20, high benefit threshold: 70) and a community of low-income residents (moderate data contribution: 60, low technical literacy: 25, low political influence: 15, high privacy preference: 85, low benefit threshold: 30).

For each of the four governance models, predict (qualitatively) which stakeholder benefits more and which is left below their benefit threshold. Then explain what the comparison reveals about governance as a distributional choice.

Answer **Corporate Centralized:** The technology company benefits significantly (high political influence receives 0.5 weight, high data contribution receives 0.3 weight), likely exceeding its threshold of 70. The residents receive low benefits (low political influence, moderate data contribution) and are likely below their threshold of 30, though the low threshold makes this less certain. Privacy protection for residents is low (high data contribution, low political influence). **Regulatory Top-Down:** Both stakeholders benefit from the baseline (45), but the technology company benefits more (high technical literacy adds 0.25 bonus). Residents may reach or slightly exceed their threshold of 30 thanks to the baseline. Privacy is more equal due to the regulatory floor (55). But voice remains skewed toward the company (higher political influence and technical literacy). **Cooperative Democratic:** Benefits are roughly equal (equal share ~47 plus small contribution bonuses). The technology company likely falls below its threshold of 70 (the system does not reward power or technical advantage enough). Residents likely meet their threshold of 30. Privacy reflects collective preference (closer to residents' high preference). Voice is formally equal. **Participatory Hybrid:** Benefits are redistributed toward residents (inverse political influence weighting). Residents receive above-average benefits and easily meet their threshold. The technology company receives less than under corporate or regulatory models but may still meet its threshold depending on the specific calculation. Privacy is high for both. Voice is more equal through the assembly mechanism. **The distributional insight:** The same two stakeholders — with the same attributes, needs, and capacities — receive dramatically different outcomes depending on the governance structure chosen. Governance is not a neutral mechanism for allocating benefits; it is a choice about whose interests take priority. The Corporate Centralized model is a choice to prioritize power and commercial value. The Participatory Hybrid is a choice to prioritize equity and participation. Neither choice is "natural" or "inevitable" — both are political.

Solutions

Selected solutions are available in appendices/answers-to-selected.md.