Preparation Checklist

  • Print 4 team prep sheets (one per team)
  • Print 6 use case scenario cards
  • Print voting ballots (1 per student, minus team members voting on their own work)
  • Print grading rubrics (1 per team = 4 copies)
  • Review fact-check reference below
  • Prepare timer for each phase
  • Optional: Whiteboard/slides for tracking time and scores

Session Timeline (60 Minutes)

0:00-0:10
Team Formation & Prep
  • Divide class into 4 teams, assign mechanisms randomly
  • Distribute prep sheets
  • Teams read and strategize (encourage note-taking)
0:10-0:30
Opening Statements & Cross-Examination
  • Each team: 2-minute opening argument (8 min total)
  • After all openings: Cross-examination round (3 min per team = 12 min)
  • Teams take turns answering questions from opponents
  • Keep time strictly—use timer visible to all
0:30-0:45
Use Case Analysis
  • Randomly distribute 4 of the 6 scenario cards (one per team)
  • Teams have 2 minutes to discuss
  • Each team presents 3-minute argument for why their mechanism fits the use case
  • Brief 1-minute rebuttal period for opposing teams
0:45-0:55
Voting & Tabulation
  • Distribute voting ballots
  • Students vote anonymously (cannot vote for own team)
  • Collect and quickly tally votes
  • Optional: Announce "People's Choice" winner per category
0:55-1:00
Instructor Debrief
  • Key takeaways about tradeoffs
  • Emphasize: "No perfect mechanism—only appropriate choices for specific contexts"
  • Preview next lesson on consensus mechanism details

Moderation Tips

Keep Energy High

  • Encourage respectful rivalry—this is a debate, not a lecture
  • Interject with "Good point!" or "That's a strong counterargument" to validate contributions
  • If a team struggles, prompt with leading questions: "What about energy consumption?" or "How does finality work in your mechanism?"

Handle Dominance/Silence

  • If one student dominates: "Let's hear from [name]—what's your take?"
  • If team is silent: "Team X, you have 30 seconds to respond—what's your defense?"
  • Encourage equal participation in grading (see rubric)

Common Pitfalls to Watch

  • False claims: Gently correct major factual errors (see fact-check section below)
  • Ad hominem attacks: Redirect to technical arguments—"Let's focus on the mechanism, not the team"
  • Oversimplification: Nudge students to acknowledge tradeoffs—"But what's the cost of that benefit?"

Fact-Check Reference

Use this to gently correct major inaccuracies during debates. Don't interrupt flow for minor errors.

Proof of Work (PoW)

Metric Accurate Value Common Misconception
Bitcoin TPS ~7 TPS "Bitcoin is slow" (true, but that's by design for security)
Energy Consumption ~150 TWh/year (2024) Comparable to Argentina; 58.9% renewable energy (2022 data)
51% Attack Cost ~$20B+ hardware + ongoing electricity Often underestimated; requires controlling majority hashrate
Finality Time ~60 min (6 confirmations) Probabilistic, not absolute; exchanges vary (some use 3-6 blocks)
Mining Centralization Top 4 pools: ~55% hashrate Concerning but miners can switch pools instantly

Proof of Stake (PoS)

Metric Accurate Value Common Misconception
Ethereum Energy Reduction 99.95% (post-merge) From ~94 TWh/year (PoW) to ~0.01 TWh/year (PoS)
Validator Count ~900,000 validators (2024) But top staking pools control ~40% of stake (Lido ~30%)
Finality Time ~15 minutes (2 epochs) Faster than PoW but not instant; economic finality earlier
Minimum Stake 32 ETH (~$96k at $3k/ETH) Liquid staking pools enable participation with less
Slashing Minimum 1 ETH, max entire stake Penalties scale with number of simultaneous slashings

Delegated Proof of Stake (DPoS)

Metric Accurate Value Common Misconception
EOS Throughput 4,000 TPS (achieved) Theoretical max higher, but 4k is real-world sustained
EOS Validators 21 active + 49 standby "Only 21" is concerning but standby validators provide backup
Block Time 0.5 seconds (EOS) Much faster than PoW/PoS, enables high TPS
Transaction Fees Zero (EOS model) Users stake tokens for network resources instead
Governance Issues Steem/Hive fork (2020) Example of both vulnerability and resilience—community forked

Byzantine Fault Tolerance (BFT)

Metric Accurate Value Common Misconception
Finality 1-3 seconds, deterministic Instant and provable—no probabilistic waiting
Throughput 1,000-10,000 TPS (Tendermint) Varies by implementation; Hedera claims 10k+
Fault Tolerance Tolerates f Byzantine out of 3f+1 total Requires >2/3 honest validators (33% Byzantine tolerance)
Validator Limit ~100-200 practical O(n²) communication complexity; not infinitely scalable
Cosmos Validators 175 active validators Tendermint (BFT + PoS hybrid); permissionless via governance

Key Takeaways to Emphasize

The Blockchain Trilemma is Real

No mechanism perfectly achieves all three: decentralization, security, and scalability. Each mechanism makes deliberate tradeoffs:

  • PoW: Maximizes security and decentralization, sacrifices scalability and efficiency
  • PoS: Balances all three, but with economic rather than thermodynamic security
  • DPoS: Maximizes scalability and governance, accepts reduced validator decentralization
  • BFT: Maximizes consistency and finality, requires smaller validator sets

Context Determines "Best"

There is no universally superior consensus mechanism. The right choice depends on:

  • Trust assumptions: Permissionless vs. permissioned participants?
  • Performance requirements: High TPS needed or can tolerate low throughput?
  • Finality needs: Is probabilistic finality acceptable or must it be instant?
  • Regulatory/ESG constraints: Energy consumption a concern?
  • Governance model: Who makes decisions about protocol upgrades?

Evolution is Ongoing

Consensus mechanisms continue to evolve:

  • Ethereum's successful merge (2022) validated PoS at scale
  • Layer 2 solutions (Lightning, rollups) address PoW/PoS scalability limits
  • Hybrid mechanisms combine approaches (e.g., Tendermint = BFT + PoS)
  • New variants emerge (e.g., Avalanche's repeated sub-sampled voting, Algorand's Pure PoS)

Grading Guidance

Holistic Assessment

  • Don't penalize teams for their assigned mechanism—grade their defense of it
  • Reward honest acknowledgment of tradeoffs over blind advocacy
  • Look for understanding of why tradeoffs exist, not just what they are
  • Equal participation matters—observe who contributes during prep and presentation

Use Peer Voting as Input

Student votes can inform (but not replace) instructor grading:

  • If team wins peer votes but made technical errors → reduce technical accuracy score
  • If team loses peer votes but was technically excellent → peer engagement may have been weak
  • Peer votes reveal which arguments resonated with audience—valuable signal

© Joerg Osterrieder 2025-2026. All rights reserved.