TDS
42 flashcards · Other · by John Terry
communication
Alice sends Bob an E-mail or makes a conference call.
association
a set of users form an online discussion group or forum.
cooperation
users work together to write or design something together.
Curation
users finding, filtering, vetting, organizing relevant content.
Deliberation
users employ formal processes to debate issues or policy.
Decision
users employ social choice (e.g., voting) to decide collectively.
Organization
users form organizations/societies, possibly decentralized.
Governance
users employ processes to govern membership, behavior.
Incentives
digital payments, cryptocurrencies, finance and investment.
Evolution
collectively finding and introducing better ways to self-organize.
Promises of computing and networking technology as a “democratizing” force
Inclusiveness, Support for individual freedoms, Resistance to control/censorship by entrenched authorities
Abstraction flaws and challenge examples
Unscalable communication or human attention costs, Unreliable or misbehaving human participants, Unintended effects of algorithms (polarization, radicalization), Bias, both real and perceived, from self-selecting online groups, Overcompression of noisy, unreliable, manipulable information, The competence- and expertise-identification problem; Funding/incentivizing the production, curation of information, Use and misuse of AI in democratic societies
Implementation flaws and challenges
Network security, availability, resilience; “routing around” failures. Identity and personhood problem Software bugs, exploitable flaws. Transparency issues Fairness challenges Inclusion challenges
Inclusiveness
Borderless communication with anyone anywhere on Earth Access to information for “everyone” Ability to collaborate productively with anyone globally Ability to associate with like-minded people anywhere
What might our long-term end goals be?
Technology enabling and incentivizing civil bottom-up self-organization. Ensure freedoms of expression and association but with accountability. Enable attention-limited users to obtain the (good) information they need to participate fully as time and interests permit, avoiding polarization. Give communities not just voting but effective and scalable deliberation, agenda-setting, and long-term governance evolution processes.
Robert Dahl's criteria
Effective Participation, Voting Equality at the Decisive Stage, Enlightened Understanding, Control of the Agenda, Inclusiveness
Basic premises of democracy
No one is perfect or trustworthy enough to be guardian of all the rest Knowledge and perceptions are imperfect and noisy Aggregation of opinions is essential to get “good” collective intelligence It’s hard to tell - and dangerous to tell - who is most trustworthy
What properties might we “like” a voting system to have?
Majority rule: if a majority supports a particular choice, that choice wins Universality: the voting system always comes to some decision Determinism: the system always decides the given for the same inputs Condorcet criterion: if a candidate A would beat all others in two-choice pairwise elections, then A should win Non-manipulability: encourages people to express their true opinions instead of being tempted to vote strategically Simplicity: Understand the voting system. Monotonicity: Adding v…
Single-winner, two-choice
majority voting: Incentive compatible; Universal, deterministic (apart from exact tie)
Single-winner, multi-choice
Plurality Runoff elections Instant runoff voting Approval voting
Multi-winner - explain both representations
Proportional versus district representation: District: reduces to set of simple winner-take-all elections Weakness: exclusion of spread-out minorities Weakness: gerrymandering Proportional: representative sample over wider area Weakness: much more complex, many systems
Party-list proportional
Widely used in Europe, ensures proportional representation in parliament Threshold requirements often imposed, for better or worse Centralizes power on party leadership; voters can’t choose candidates
Single transferable vote (STV)
Multi-winner version of instant runoff voting Takes ranked-choice ballots from voters Runs multiple rounds just as computation to decide results
Blockchain basics
Concept of maintaining a shared, public, immutable ledger Bitcoin: blocks, chain, consensus, transactions Smart contracts: Bitcoin pay-to-script, Ethereum
DAO
Goal: fully on-chain organizations Purposes: many in principle DAO example: community investment On-chain funding Often: anyone can join, contribute On-chain governance Proposals, voting (see E-voting next week)
Delegative or liquid democracy
A balance or midpoint between direct & representative democracy Key idea: give people more free, dynamic choice of representation Participate and vote directly when you want and have time Delegate to a chosen representative, but revoke at any time Free choice of representative: can be trusted friend or neighbor Different choices of delegation (or not) on different topics
Some prototypes supporting delegative democracy
LiquidFeedback Adhocracy Democracy.OS
Voter-delegate interaction
Ideal: shift from distant “celebrity” representative to social advisor Does delegate serve in mandate or representative role? Mandate: carry out the voter’s stated will or preference Representative: learn situation and decide for the voter Delegate acquires more voting weight, but must “answer for” it Threat of revocation always present, but only the blunt instrument
Delegation design space
“One-hop” or transitive delegation? Single-choice or multi-choice? How to split a multi-choice delegation? What about delegation cycles? Further issues and challenges Vote privacy versus accountability
Prediction markets
Basic idea: let people make predictions, but put money where mouth is
Crowdfunding
Basic idea/model: announce project, collect small investment commitments
Quadratic voting
Basic idea: buy votes for/against a position, but pay the square of number of votes
Random sampling and mini-public deliberation
Basic principles: the statistical representative property of a “good” sample
Why be anonymous ?
Avoid repercussions or reprisals tied to your real identity Separation of roles (private, professional, etc...) General privacy protection (digital uranium) Creation of multiple identities Freedom Prejudice based on reputation
dark side of anonymity
Be more candid, forget that other is a person + escalation Organized mass attacks / Mass gossip (Anonymous Sybil)
Mitigate, “balance” anonymity vs. identity / accountability
One to one mapping pseudonym Advantages : Counters Sybil attacks Makes pseudonyms non-discardable: creates scarcity Deanonymization on legal requirement? Challenges: IP address, geolocation Biometrics: privacy, false positives/negatives, invasive Phone number (insecure SS7, burner phones) Identity theft
Technologies for online anonymity
Anonymous communication tools Naive: anonymity through obscurity, “IP addresses look anonymous” Trusted third party: anonymous remailers, VPN services Decentralized systems: MIX nets, onion routing (Tor), DC-nets, researc Cryptographic tools for anonymity and pseudonymity Weak pseudonymity: e.g., 4chan tags, public keys, Bitcoin wallets Single-use pseudonyms: e.g., per-transaction keypairs, wallets Anonymous credentials, group or ring signatures 1-to-1 mappings of anonymous to real identities B…
Potential mitigation factors:
Labeling of information origin, provenance Helps people take detached/skeptical perspective, evaluate sources Diversity of perspective opinion: make groups “divers” (how?) Random sampling-based groups, e.g., deliberative polls, juries Increase breadth, diversity of information diet (Not just from friends but from larger perspectives)
Potential technological/algorithmic approaches
Basic: diversity through random sampling Advanced: balancing local and global with compact graph summarization
Motivations for remote voting and E-voting
Short-term, pragmatic considerations Convenience Usability, accessibility, inclusion: e.g., people with disabilities Boost voter turnout, participation? Longer-term: enabler for continuous-participation democracy? Liquid democracy Online deliberative polls
E-Voting stages
Forming/marking ballots Casting ballots Anonymizing ballots Tallying ballots
Challenges of e-voting
Authentication: ensure that only authorized voters can vote Inclusion: ensure that all authorized voters can vote Integrity: ensure correctness of result Privacy: keep secret how you voted (and sometimes even whether you voted) Coercion-resistance: vote privacy even if voter is “cooperating” Transparency: ability to convince voters that the vote is correct