Motus Corpus · Play × Systems × Chain

No two rings are linked. All three are inseparable. Pull one out.

Build what people choose, understand and govern.

The Motus Builder Praxis weaves game design, systems thinking and crypto coordination into one way of building. Each field holds the others' blind spot.

Enter the Systems Sanctum Explore the corpus

Three worlds · one light

  1. PLAY · the pull

    A core loop pulls everything toward one lit action. Play knows why people show up.

  2. SYSTEMS · the field

    Out of the waves of behaviour, structure rises. Systems knows where to push.

  3. CHAIN · the ledger

    Every block takes the same light, open to everyone. Chain knows how strangers keep promises at scale.

  4. BOUND · the light

    Where all three overlap, the light rises. Pull one world out and it goes dark.

    All three overlap. The light is up.

01 · THE THESIS

One structure.
Three dialects.

Strip away the vocabulary and an outlier game, a resilient system and a living protocol are the same object: a feedback structure that makes the right behavior the easiest one.

PLAY

knows why people show up

Motivation, feedback, flow, mastery, belonging. Fifty years of learning what makes people choose to stay.

SYSTEMS

knows where to push

Stocks, flows, loops, delays and the leverage ladder. The science of why structures produce the behavior they do.

CHAIN

knows how strangers keep promises at scale

Credible rules, open ledgers, programmable money and new ways for thousands of people to decide together.

Why Borromean: each field holds the others' blind spot

  • − PLAY Correct models and credible rules that nobody chooses to enter.
  • − SYSTEMS Loops nobody can see the limits of: compulsion machines and casinos.
  • − CHAIN Good games and good models that cannot keep promises between strangers at scale.

The deepest layer of all three is the same three rungs of Meadows' ladder: the rules (5), the goal (3) and the paradigm (2). Mechanics are parameters. Meaning lives above them.

02 · THE TWELVE WEAVES

Three rings.
Twelve threads.

Twelve structures that all three fields found on their own, ordered from the shallowest lever to the deepest.

1. The Economy

RUNG 10 · STOCKS & FLOWS

Inflation is a faucet without a sink. A game economy, a firm and a chain all live or die by that one ratio.

PLAY

Faucets & sinks

Rewards flow in, spending flows out. Designers map the plumbing first; Dormans' Machinations models game economies as stocks and flows.

SYSTEMS

Stocks & flows

A stock changes only through its flows. Its level is the running sum of inflow minus outflow: the memory of the system.

CHAIN

Issuance & burn

Block rewards are the faucet; fee burns are the sink. EIP-1559 burns the base fee, so supply responds to how much the chain is used.

THE SHADOWPlay-to-earn economies minted rewards faster than anything consumed them. The token became the score, and the score inflated away.

THE MOTUS MOVEDraw your stocks before your screens. For every faucet, name its sink.

IN THE CORPUSInternal Economy: Faucets, Sinks & Machinations · Stocks and Flows · Dormans' Machinations · EIP-1559 Base-Fee Burn · Play-to-Earn Collapse: Axie Infinity & Ronin

SYSTEMS · structure rises out of the waves

03 · THE LADDER

Climb the
leverage ladder.

Donella Meadows' twelve places to push a system, in all three fields. Most effort lands on rung 12. The climb worth making is 12 → 5.

RUNG 12

Parameters & numbers

PLAYDamage values, drop rates, XP curves: the balance patch.

CHAINFee levels, interest rates, emission numbers.

THE LESSONWhere most effort goes, and where the least leverage lives.

Source: Donella Meadows, Leverage Points: Places to Intervene in a System (1999). Play and chain mappings are this corpus's synthesis.

PLAY · a loop pulls toward one lit action

04 · THE PSYCHOLOGY OF PLAY

Empower
or extract.

The same psychology powers the most loved games and the most predatory ones. The difference is structural: whether a mechanic satisfies a need or exploits one.

The flow channel

Drag the dot (or use the arrow keys). Challenge above skill is anxiety; below it is boredom; the band where they match is flow. A balancing loop keeps people inside it: in games it is difficulty, in Bitcoin it is the difficulty retarget.

In flow: challenge matches skill.

The extraction test

Run any mechanic through seven questions. Each yes closes the ring a little further.

0 of 7: answer each question honestly.

Test synthesized from Self-Determination Theory (Ryan, Rigby & Przybylski 2006), Zagal, Björk & Lewis on dark patterns (2013), and the loot-box literature; sources in the corpus below.

The evidence underneath

The models behind this, each in the corpus below:

05 · LOOP LAB

Structure
generates behavior.

A toy token economy. Move a slider and watch who stays.

PlayersToken value

A toy model, not a forecast. What it shows is structural: in an earn-first economy, earnings are paid for by the growth of new arrivals, so when growth stops, earnings stop. A longer delay pushes the peak later; the fall then lands outside the window you are watching.

CHAIN · every block takes the same light

06 · CHAIN · WHAT CHANGED IN 2026

Rules that
strangers can verify.

In 2026 crypto became plumbing: institutions settle on it, and software agents use it to pay each other.

$314.2B

Total stablecoin circulating supply across all tracked assets, right now

stablecoins.llama.fi/stablecoins?includePrices=true · 2026-10-07PRIMARY

$53B/mo

Combined Kalshi + Polymarket monthly trading volume, most recent clean month

pewresearch.org/short-reads/2026/09/23/prediction-markets-tra… · 2026-07PRIMARY

165M txns

x402 protocol adoption scale (active agents, transactions, cumulative volume)

coinbase.com/developer-platform/discover/launches/x402 · 2026-04

$33.5B

Tokenized real-world-asset value, excluding stablecoins, total across tracked products

app.rwa.xyz/ · 2026-07

Jul 18, 2025

GENIUS Act (S.1582) signed into law by the President, first US federal stablecoin framework

whitehouse.gov/briefings-statements/2025/07/the-president-sig… · 2025-07-18PRIMARY

-99.8%

SLP all-time high vs current price

api.coingecko.com/api/v3/coins/smooth-love-potion · 2026-10-07PRIMARY

The deepest pattern in the chain corpus: nearly every high-leverage mechanism replaces an intermediary's judgment with a balancing loop that watches a number and acts on its own: difficulty, slashing, liquidations, the base fee, a market price. And every shadow in the corpus is that same loop's blind spot.

Market figures come from secondary trackers and press as cited; treat them as dated snapshots, not live data.

The mechanisms

The models behind this, each in the corpus below:

07 · COORDINATION AT SCALE

Many people.
One voice each.

Groups hold together when a vote can't be bought, the reasons are on the record, and leaving is cheap for dissenters and costly for raiders.

$60M+

Gitcoin Grants cumulative public-goods funding distributed to date

gitcoin.co/blog/gitcoin-grants-23-retro · 2025-11

60M+ OP

Cumulative RetroPGF / Retro Funding OP distributed since program launch

optimism.io/blog/retro-funding-2025 · 2025-01

$46M raised

Amount raised and number of contributors, per ConstitutionDAO's own Juicebox campaign page

docs.juicebox.money/blog/constitutiondao-highlight/ · 2021-11PRIMARY

$182M

Beanstalk flash-loan governance exploit, total drained

bloomberg.com/news/articles/2022-04-18/defi-project-beanstalk… · 2022-04

Gini 0.995

Arbitrum DAO voting-power concentration (Gini coefficient)

papers.ssrn.com/sol3/papers.cfm?abstract_id=7103878 · 2025-01

$44.2M

MetaDAO's Backable (ex-Futardio) launchpad total commitments

solanacompass.com/news/metadao-renames-futardio-as-backable-i… · 2026-10

The outlier: the most generous exit ever designed (Nouns DAO's fork) was used by arbitrageurs who bought in to fork out. A safety valve becomes an attack surface when exit pays more than staying. Exit has to be priced for dissent, not for profit.

The mechanisms

The models behind this, each in the corpus below:

THE LIGHT · what none of the three makes alone

08 · CORE VALUES

Values are
the top rungs.

Values set a system's goal (rung 3) and its paradigm (rung 2). Write them down before the rules, and they shape everything below.

FAMILY = EMERGENCE = MOTUS = TRUST = ONE = TOGETHER.

August JD France

Each value becomes a mechanism in all three fields, and a question you can ask of any build. Open one.

ONE PERSON, ONE VOICE

PLAYPlayer councils; every voice counted.

SYSTEMSRequisite variety: a regulator must hear as much variety as it governs.

CHAINProof of personhood; quadratic voting, where voice grows with the square root of stake.

THE GATECan one whale outvote a thousand people?

TRANSPARENCY

PLAYPublished odds, readable rules, patch notes with reasons.

SYSTEMSInformation flows (rung 6).

CHAINTransparent treasuries; every vote on the record.

THE GATECan anyone trace where the value went?

AGENCY

PLAYAutonomy: meaningful choices, many paths.

SYSTEMSSelf-organization (rung 4).

CHAINSelf-custody; exit; rage-quit.

THE GATECan they leave with what is theirs?

MASTERY

PLAYCompetence: skill that compounds.

SYSTEMSLearning loops; feedback that teaches.

CHAINEarned, verifiable credentials.

THE GATEDoes the person get better, or only get more?

TRUST

PLAYFair rules; no retroactive changes to what people paid for.

SYSTEMSCredible feedback; promises kept by structure.

CHAINCredible neutrality; commitments that bind the builder.

THE GATEWould the system behave the same if no one were watching the builder?

EMERGENCE

PLAYSystemic design: few rules, many outcomes.

SYSTEMSSelf-organization and emergence.

CHAINComposability: open building blocks.

THE GATEHas someone done something with it you did not design?

TOGETHER

PLAYRelatedness: co-op, guilds, shared wins.

SYSTEMSCommons governance (Ostrom).

CHAINPublic-goods funding: quadratic, retroactive.

THE GATEDoes one person's win make the next person's easier?

NO DARK PATTERNS

PLAYWhite-hat drives only.

SYSTEMSNo exploiting delays; no shifting the burden.

CHAINNo extraction from your own users: fair launches, MEV protection.

THE GATEWould you show the person the exact mechanic you used on them?

RECORDED RATIONALE

PLAYReplays; patch notes that explain why.

SYSTEMSMemory is a stock; decisions carry their reasons.

CHAINProposals with rationale, on the record.

THE GATECan a newcomer read why it is this way?

09 · THE MOTUS BUILDER LOOP

Truth → Dream
→ Move.

Seven moves from what is true to what you ship. Each carries a gate: a question that could prove it wrong.

  1. TRUTH · MOVE 1

    Name the stock

    What accumulates that a person truly values: mastery, trust, belonging, capital? Build around that stock, not around your metric.

    THE GATEIf the stock can only rise by spending money, it is not theirs.

  2. TRUTH · MOVE 2

    Map the loops

    Draw the reinforcing engine and the balancing governor. Name every faucet's sink and every delay.

    THE GATEEvery faucet has a sink; every engine has a governor.

  3. DREAM · MOVE 3

    Fuel it with needs

    Power the core loop with autonomy, competence and relatedness, not with fear of loss.

    THE GATERemove the extrinsic reward for a week. Does the loop still turn?

  4. DREAM · MOVE 4

    Light the signal

    Feedback within a tenth of a second. Consequences visible to the person who causes them. Rules anyone can read.

    THE GATECan a newcomer state the rules after five minutes?

  5. DREAM · MOVE 5

    Bind the builder

    Encode the rules as commitments that bind you too. Change them only by consent: Motus By Votus.

    THE GATECould the builder change the rules tonight without anyone's consent?

  6. MOVE · MOVE 6

    Hand over the structure

    Give people power over structure: proposals, councils, creation tools, forks and exit.

    THE GATECan a person leave with what they built?

  7. MOVE · MOVE 7

    Write the goal in public

    State the objective, measure it with more than one number, and read it back against what the system actually does.

    THE GATEIf we maximized our top metric, would the people in it be better off?

Motus By Votus. Diagnose the structure, get or confirm the decision, then move. Act freely inside the circle; ask before anything irreversible.

THE LIGHT · now it answers

10 · ASK ALLOUS ASI

Ask the whole praxis.

Ask Allous ASI to fit the praxis to your project. Bring your own OpenRouter key; it stays in your browser and goes only to OpenRouter.

Ask Allous ASI

11 · THE CORPUS

Every model,
with its sources.

Every model with its sources. Open one to see how it works, how it fails and the evidence behind it.

124 models · 192 sourced, dated data points

Open the corpus
PLAY · Frame · RUNG 2

MDA Framework

Mechanics produce dynamics; dynamics produce feeling. Design the chain, not the word 'fun'.

Go deeper
STRUCTURE
A one-way causal chain: designer-authored Mechanics (rules, components) generate run-time Dynamics (emergent play behavior), which produce player-felt Aesthetics (8 named emotional responses) — the designer only ever touches the first link.
WHY IT IS LEVERAGE
It replaces the single word 'fun' with 8 named aesthetics and forces designers to trace cause from rule to feeling, exposing exactly where a design fails instead of vaguely 'not fun'.
THE DRIVER
Need for a shared, falsifiable design vocabulary instead of taste arguments.
THE SHADOW
Used as gospel it reduces design to a checklist of 8 aesthetics, losing the synthesis a strong authorial vision provides.
EXAMPLES
Used across GDC-era design curricula, cited in most modern design postmortems
ANALOGS
Input, process, output model (systems); Protocol layer producing emergent market behavior producing user sentiment (crypto)
DATA
PLAY · Frame · RUNG 2

Theory of Fun: Fun Is Learning

Fun is the brain's signal for successfully building a new pattern; boredom is a mastered one.

Go deeper
STRUCTURE
Engagement tracks the rate of novel-pattern acquisition; once a system is fully modeled by the player, the same stimulus stops firing reward and gets relabeled 'boring' — the mechanic hasn't changed, the player's model has.
WHY IT IS LEVERAGE
It explains why identical content fatigues and reframes 'add more content' as the wrong fix — the real lever is new patterns, not more of the old pattern.
THE DRIVER
The brain's intrinsic reward for successful pattern recognition and prediction.
THE SHADOW
Cynically applied, it justifies manufactured fake novelty (reskins, random seeds) that feels new but teaches nothing, burning trust once recognized.
EXAMPLES
Tetris, Baba Is You, Spelunky
ANALOGS
Mental models in a learning organization (Senge, systems); Explaining token incentives as literacy, not just yield (crypto)
DATA
PLAY · Frame · RUNG 2

The Magic Circle

Stepping over a line where ordinary consequences pause and the game's rules are the only ones that matter.

Go deeper
STRUCTURE
Huizinga's 'magic circle' names the bounded space-time where play's artificial rules temporarily supersede ordinary-life rules; Salen and Zimmerman formalize it as the system boundary making a shared ruleset legible and voluntarily binding for as long as players stay inside it.
WHY IT IS LEVERAGE
It's the paradigm frame under every mechanic here: loot boxes, ranked ladders, guild raids only 'work' because players agreed the circle's rules matter — break that consent and the structure of meaning collapses even with mechanics untouched.
THE DRIVER
Voluntary, consensual entry into an alternate ruleset — the basis of play as distinct from ordinary obligation.
THE SHADOW
When money crosses the boundary both ways (buy an advantage, resell it outside the game), the circle stops being magic and becomes just another market.
EXAMPLES
Any competitive ruleset from chess to esports, MMO real-money-trade controversies as a broken-circle case study
ANALOGS
System boundary defining what's inside versus outside a model (systems); Smart-contract-enforced ruleset as a trust boundary, 'code is law' (crypto)
DATA
PLAY · Wellbeing · RUNG 3

WHO Gaming Disorder, Prevalence & the Telemetry Correction

The clinical disorder is real and rare; the 'screen time' panic mostly isn't supported.

Go deeper
STRUCTURE
Two measurement stocks run in parallel — clinical diagnostic criteria estimating disorder prevalence, and objective telemetry of actual play correlated with wellbeing — and the two are frequently conflated in public discourse.
WHY IT IS LEVERAGE
Accurate prevalence data lets clinicians target help at the small group who need it, while telemetry evidence frees most players from unwarranted guilt about ordinary play.
THE DRIVER
Distress and loss of control over gaming, not time spent per se, is the clinical threshold WHO set for disorder.
THE SHADOW
Panic-driven conflation of 'plays a lot' with 'has a disorder' obscures the more common extractive problem: specific manipulative monetization mechanics, not play itself.
EXAMPLES
Plants vs. Zombies: Battle for Neighborville (telemetry study), Animal Crossing: New Horizons (telemetry study), Clinical gaming-disorder treatment services, South Korea
ANALOGS
Two measurement stocks telling different stories about one system (systems); On-chain activity volume mistaken for genuine protocol health (chain)
DATA
PLAY · Social · RUNG 4

Guilds, Raids & Asymmetric Co-op Roles

Content too complex for one mind forces players to self-organize into structure.

Go deeper
STRUCTURE
A raid's mechanic count exceeds individual working memory, forcing role specialization (tank, healer, DPS); the guild is a persistent structure the designer never coded, only made necessary by content no one can clear alone.
WHY IT IS LEVERAGE
The designer never builds the guild — they build content impossible to clear solo, and the social structure assembles itself to meet it: leverage from under-specifying rules, not over-specifying them.
THE DRIVER
Belonging, mutual obligation, and identity bound up in a role the group depends on.
THE SHADOW
Mandatory interdependence becomes a scheduling and attendance tax that pushes out casual players and can curdle into toxicity when a failure is visible to the group.
EXAMPLES
World of Warcraft raiding, Final Fantasy XIV, Deep Rock Galactic
ANALOGS
Self-organization into subsystems (Meadows rung 4); Sub-DAOs or working groups forming around a task too large for one contributor (DAO)
DATA
  • WoW raid sizes forced role specialization, from fixed 40-player raids to flexible scaling: Vanilla World of Warcraft raids such as Molten Core required 40 players; since Wrath of the Lich King (Nov 2008) all new raids shipped in both 10-player and 25-player versions, and Mists of Pandaria's Siege of Orgrimmar later introduced Flexible Raiding scaling anywhere between 10 and 25 warcraft.wiki.gg/wiki/Raid · 2008-11
PLAY · Economy · RUNG 4

EVE Online: Player-Run Economy + Elected Council

A shared server where the developer publishes central-bank statistics and players elect a voice into design.

Go deeper
STRUCTURE
Nearly all matter in New Eden is player-mined and player-destroyed in player wars, a destruction sink dwarfing any NPC sink; CCP publishes Monthly Economic Reports (ISK supply, PLEX price, price indices) and funds an elected player council voicing grievances into design.
WHY IT IS LEVERAGE
It's the only major game handing leverage points 3 (goals), 4 (self-organization) and 6 (information flow) partly to players themselves — proof a game economy can run with real transparency and survive its own player-run bank frauds.
THE DRIVER
Stakes that are real because loss is permanent, plus a legitimate voice in how the rules evolve.
THE SHADOW
The Council of Stellar Management is advisory, not binding — CCP can and does ignore it, so an elected council can function as legitimacy cover for decisions already made.
EXAMPLES
EVE Online
ANALOGS
Central-bank-style published statistics for a closed economy (systems economics); Token-voted elected council / on-chain governance delegates (DAO)
DATA
PLAY · Creation · RUNG 4

UGC Platforms & Creator Payout Economies

Turning the player base into a revenue-shared content studio bigger than the original one.

Go deeper
STRUCTURE
The platform opens its engine or editor to players (Roblox Studio, UEFN, Minecraft marketplace), meters engagement or transactions per creation, and routes a revenue share back — turning a fixed internal content budget into a variable, near-unbounded one funded by players.
WHY IT IS LEVERAGE
Rung-4 leverage at platform scale: instead of building more content, the company changes who holds the power to create structure at all, so the content ceiling stops being headcount-bound.
THE DRIVER
Creators are paid in money and audience; players get near-infinite content variety.
THE SHADOW
Payout formulas are opaque and can be changed unilaterally, leaving creators who built a business on top exposed to rules they never voted on.
EXAMPLES
Roblox, Fortnite (UEFN / Creative), Minecraft
ANALOGS
Commons governed by its contributors (Ostrom); Retroactive public-goods funding and protocol revenue share to builders (crypto)
DATA
PLAY · Ethics · RUNG 4

Octalysis — White-Hat vs Black-Hat Drives

Eight drives, one line: does this make you feel powerful, or just compelled?

Go deeper
STRUCTURE
All gamified mechanics map onto 8 core drives; the 3 White Hat drives (meaning, accomplishment, empowerment) build a loop the player controls, the 3 Black Hat drives (scarcity, unpredictability, loss) build a loop that controls the player.
WHY IT IS LEVERAGE
Designing primarily from White Hat drives produces motivation that feels good in the moment and still feels good afterward.
THE DRIVER
Human motivation decomposes into a small set of recurring drives that any system — game or otherwise — can be built around.
THE SHADOW
Designing primarily from Black Hat drives produces high short-term engagement via urgency and obsession that players often regret immediately after.
EXAMPLES
Habitica (White Hat-leaning), Gacha mobile games (Black Hat-leaning), Duolingo (mixed drives)
ANALOGS
Self-organization toward a rule set the player chose vs. one imposed (systems); Protocol incentives built on real utility vs. pure speculation (chain)
DATA
PLAY · Social · RUNG 4

Bartle Types & the Quantic Foundry Motivation Model

Players don't want the same game — they want different reasons to stay in it.

Go deeper
STRUCTURE
The player population segments into stable motivational clusters; matching content delivery to the dominant cluster sustains engagement, while mismatched content drains it regardless of production quality.
WHY IT IS LEVERAGE
Designing distinct, legitimate paths — exploration, mastery, social, competition — lets one world satisfy genuinely different people without forcing anyone into someone else's loop.
THE DRIVER
Players are drawn to persistent virtual worlds for structurally different reasons: achievement, exploration, socializing, or competing.
THE SHADOW
The same segmentation data gets reused to target each cluster's specific weak point for monetization — Achievers sold completion packs, Socializers sold status gifts.
EXAMPLES
EVE Online, Animal Crossing: New Horizons, World of Warcraft
ANALOGS
Population segmented into stable preference clusters (systems); Airdrop/token targeting segmented by on-chain behavioral cluster (chain)
DATA
  • Bartle's 1996 taxonomy proposed four player types (Achievers, Explorers, Socializers, Killers), later extended empirically by Quantic Foundry's Gamer Motivation Model, built from survey data described in "The Gamer Motivation Profile: What We Learned From 250,000 Gamers": 250,000 gamers surveyed researchgate.net/publication/309020043_The_Gamer_Motivation_P… · 2016-10
PLAY · Ethics · RUNG 4

The Hooked Model & Its Critiques

Trigger, action, variable reward, investment — repeat until it's a habit.

Go deeper
STRUCTURE
A closed loop runs Trigger → Action → Variable Reward → Investment; each Investment (data, content, followers) raises the odds the next internal trigger fires the loop again without external prompting.
WHY IT IS LEVERAGE
Used transparently, the loop describes how any genuinely useful habit — exercise, language learning, saving — gets built on investment that compounds real value for the user.
THE DRIVER
A behavior repeated enough times in response to an internal cue (boredom, loneliness, anxiety) becomes automatic.
THE SHADOW
Tuned for engagement metrics rather than user benefit, the same loop turns a user's own negative emotions into the trigger the product is built to exploit.
EXAMPLES
Social media feeds (cited Hook Model examples), Mobile free-to-play games, Nir Eyal's own "Indistractable" as a stated course-correction
ANALOGS
Reinforcing habit loop with a compounding investment stock (systems); Engage-to-earn loops where the investment is a user's own attention (chain)
DATA
  • Nir Eyal's "Hooked: How to Build Habit-Forming Products" (2014) formalized the Trigger-Action-Variable Reward-Investment loop; his 2019 follow-up "Indistractable" responded to critics about the model's potential for manipulation: 2014 model, 2019 response nirandfar.com/hooked/ · 2014-01
PLAY · Wellbeing · RUNG 4

Prosocial & Cooperative Play Benefits

The same systems that make games sticky can make players kinder, together.

Go deeper
STRUCTURE
A shared, interdependent goal is only completable jointly; helping a teammate succeed raises the whole group's stock, which raises the incentive to keep helping — a cooperative loop instead of a zero-sum one.
WHY IT IS LEVERAGE
Cooperative and prosocial game structures are linked to real-world helping behavior and team trust, especially when success is framed as jointly rather than individually achieved.
THE DRIVER
Structuring success as mutually achieved rather than zero-sum recruits cooperative instead of competitive social instincts.
THE SHADOW
Cooperative framing can be co-opted into obligation — guilds that punish members for under-contributing, or pressure to recruit/monetize friends to keep a shared structure alive.
EXAMPLES
It Takes Two, Overcooked, Animal Crossing: New Horizons island visits
ANALOGS
Cooperative reinforcing loop replacing a zero-sum one (systems); Quadratic/public-goods funding rewarding genuine collective contribution (chain)
DATA
  • A review of experimental and correlational research found cooperative and prosocial game content associated with increased helping behavior relative to competitive or violent content, alongside documented cognitive, motivational, and social benefits of gameplay generally: narrative review synthesis doi.org/10.1037/a0034857 · 2014-01
PLAY · Frame · RUNG 5

A Series of Interesting Decisions

Sid Meier's one-line definition of a game: choices whose outcome you can't predict but can influence.

Go deeper
STRUCTURE
Decision quality scales with viable options times outcome uncertainty times consequence weight. Remove any factor — one dominant option, no uncertainty, no stakes — and the decision stops being 'interesting' regardless of production value.
WHY IT IS LEVERAGE
It's the smallest unit of design leverage: audit any mechanic by asking whether it adds a real decision or just an action. Most added 'content' adds actions, not decisions.
THE DRIVER
Agency — the feeling that a choice, not a die roll or cutscene, caused the outcome.
THE SHADOW
Decays into false choice: illusory options that funnel to one optimal path, which players detect and resent as manipulation.
EXAMPLES
Civilization series, XCOM, Into the Breach
ANALOGS
Decision-tree / expected-value branching (decision theory); Governance proposals as real options with real stakes (DAO)
DATA
PLAY · Emergence · RUNG 5

Emergent / Systemic Design

A small, orthogonal rule set that multiplies combinatorially instead of a list of special cases.

Go deeper
STRUCTURE
Breath of the Wild's Chemistry Engine reduces world interaction to 3 rules — elements change materials, elements change elements, materials never change materials — across 5 elements, tracked apart from physics. N rules by M objects yield far more than N+M scripted outcomes.
WHY IT IS LEVERAGE
Proof that rung-5 leverage (rules) beats rung-12 leverage (content volume): a handful of orthogonal rules outproduces thousands of hand-authored interactions and keeps generating novel solutions years after launch.
THE DRIVER
Discovery and authorship — the player feels they invented the solution, not found the designer's intended one.
THE SHADOW
Systemic depth is expensive to make legible; without strong feedback, emergent systems stay invisible and players never discover the possibility space exists.
EXAMPLES
The Legend of Zelda: Breath of the Wild, Dwarf Fortress, Deus Ex / Dishonored
ANALOGS
Simple rules generating complex behavior (complexity science, cellular automata); Composable DeFi primitives combining into unplanned strategies (crypto)
DATA
PLAY · Feel · RUNG 5

Onboarding: Teach by Play (World 1-1)

The best tutorial is a level that can't be failed without learning, and never says 'tutorial'.

Go deeper
STRUCTURE
Super Mario Bros. World 1-1 sequences forced encounters in fixed order (open run, first Goomba, first pipe gap, first question-block cluster, first staircase) so each rule is demonstrated by a situation the player must resolve, with no UI text at all.
WHY IT IS LEVERAGE
Proof that teaching can be encoded entirely as rule-ordering rather than information overlay — the player builds the mental model through consequence, which measurably beats any other ordering.
THE DRIVER
Learning by doing — a rewarded correct action is retained better than a read instruction.
THE SHADOW
Transplanted into mechanically denser genres, 'no text, just design' onboarding can lose non-core audiences; it works because Mario's rule set is small enough to demonstrate in 60 seconds.
EXAMPLES
Super Mario Bros. (1985), Half-Life 2, Portal
ANALOGS
Teaching via rule sequencing rather than information overlay (rung 5 vs. rung 6); Guided first-transaction, testnet-first onboarding (crypto UX)
DATA
  • Canonical World 1-1 rule-ordering measurably outperforms shuffled orderings: Empirical study found canonical ordering converged fastest, had the highest learning efficiency, and was the only tested condition with zero catastrophic failures arxiv.org/html/2606.29511 · 2026-06
PLAY · Social · RUNG 5

Reciprocity & Social Obligation Mechanics

A free gift from a friend is a debt you didn't ask for.

Go deeper
STRUCTURE
A gift flow between players creates an unbalanced social-debt stock in the recipient; reciprocity norms create a feedback loop pulling the recipient toward a return gift, often routed back through the platform.
WHY IT IS LEVERAGE
Genuine gifting and co-op assist mechanics strengthen real relationships and make cooperation legible and rewarding.
THE DRIVER
Receiving a favor creates a felt social obligation to return it, independent of the favor's objective value.
THE SHADOW
Platforms engineer gift flows that resolve only through continued engagement or spend — a gifted 'life' that expires, pressuring the recipient to open the app or buy in to reciprocate.
EXAMPLES
FarmVille gifting, Candy Crush 'ask for lives', Clash of Clans clan donations
ANALOGS
Social-debt stock cleared only by a return flow (systems); Referral-airdrop obligation chains (chain)
DATA
  • The norm of reciprocity holds that receiving a gift or favor creates a felt social obligation to return it, a mechanism platforms exploit via built-in gifting features: Gouldner, 1960 (foundational norm) en.wikipedia.org/wiki/Reciprocity_(social_psychology) · 2026-09
PLAY · Bias · RUNG 5

FOMO & Limited-Time Events

Scarcity in time is scarcity a designer can manufacture for free.

Go deeper
STRUCTURE
An artificial deadline converts an otherwise-renewable item into a one-shot stock that drains to zero and never refills, spiking urgency-driven action right before expiry.
WHY IT IS LEVERAGE
Real limited-time community events — seasonal content, live shared moments — create genuine cultural moments and a reason to coordinate play with others.
THE DRIVER
Fear of missing a one-time opportunity motivates action more strongly than an equivalent permanently-available opportunity.
THE SHADOW
Manufactured scarcity on items with zero marginal production cost (a reskinned cosmetic on a 48-hour timer) turns ordinary purchases into panic-driven ones with no real content behind the urgency.
EXAMPLES
Fortnite item shop rotations, Mobile gacha limited banners, Animal Crossing seasonal events
ANALOGS
Artificial deadline forcing a stock to a forced zero (systems); Time-boxed mint windows driving panic-buying (chain)
DATA
PLAY · Ethics · RUNG 5

Dark Patterns in the Design of Games

A design choice against the player's interest, built to work without their consent.

Go deeper
STRUCTURE
A taxonomy of specific exploits (grinding walls, monetized rivalries, exploiting the social graph) layered onto normal game loops, each redirecting a flow of time or money toward the publisher without informed buy-in.
WHY IT IS LEVERAGE
A shared, specific vocabulary lets designers and regulators identify and remove exploitative patterns without banning the legitimate mechanics underneath them.
THE DRIVER
The absence of informed consent is the defining line between a hard-but-fair design choice and a dark pattern.
THE SHADOW
Catalogued patterns — grinding walls sized to sell a skip, social-graph spam required to progress — are deployed precisely because they work on players who never agreed to them.
EXAMPLES
Mobile stamina/energy systems, Social games requiring friend-spam to progress, Pay-to-skip grind walls
ANALOGS
A structural exploit layered on an otherwise-legitimate loop (systems); Dark-pattern UX in token-claim flows and hidden slippage (chain)
DATA
  • Zagal, Björk & Lewis (2013) defined dark game design patterns as choices used intentionally to cause negative player experiences against players' interests and likely without consent, cataloguing them across three categories: increasing time spent, increasing money spent, and exploiting the player's social network: presented at Foundations of Digital Games 2013 dl.digra.org/index.php/dl/article/download/889/889/886 · 2013-05
PLAY · Loops · RUNG 6

Skill Atoms & Loops-and-Arcs

The atomic unit of learning in a game: model, action, feedback, updated model.

Go deeper
STRUCTURE
A skill atom chains mental model, action, feedback, and new model; atoms link into loops (repeatable) nested inside arcs (one-time beats) — a mastered atom becomes the input 'verb' for the next unmastered atom.
WHY IT IS LEVERAGE
It gives mastery a mechanical definition: a skill tree is a dependency graph of skill atoms, so you can diagram exactly where a game teaches versus where it just gates.
THE DRIVER
The reward of a mental model being confirmed or corrected — competence-building itself.
THE SHADOW
Over-producing arcs (one-off spectacle) with too few real loops beneath them leaves nothing to master once the content is consumed.
EXAMPLES
Portal, Braid, most tutorial-free indie puzzle games
ANALOGS
Single-loop vs. double-loop learning (Argyris, systems); Protocol literacy ladder: wallet, swap, LP, governance (crypto onboarding)
DATA
PLAY · Feel · RUNG 6

Game Feel / Juice

The non-functional feedback layer that makes a correct input feel correct.

Go deeper
STRUCTURE
Juice adds redundant, multi-sensory confirmation (visual, aural, haptic) on top of an unchanged mechanical outcome — it changes nothing about the rules but raises the information bandwidth of every action's feedback.
WHY IT IS LEVERAGE
Leverage on perception, not mechanics: two builds with identical collision code and win rates are rated completely differently by players depending only on feedback richness.
THE DRIVER
Immediate, visceral confirmation that an action registered and mattered.
THE SHADOW
Can mask a mechanically shallow game for a while, and over-juicing creates visual noise that degrades readability in competitive contexts.
EXAMPLES
Vlambeer's games (Nuclear Throne, Luftrausers), Celeste, Downwell
ANALOGS
Signal-to-noise ratio in a feedback loop (systems); Wallet-confirmation UX trust signals (crypto)
DATA
  • Canonical game-feel/juice talk: title, speakers and venue: 'Juice It or Lose It' by Martin Jonasson and Petri Purho, GDC Europe 2012 — a live demo adding tweens, screenshake and particle effects to a plain Breakout clone with no change to its mechanics or win rate gdcvault.com/play/1016487/juice-it-or-lose · 2012-08
PLAY · Social · RUNG 6

Social Deduction: Hidden Roles + Public Voting

The mechanic is the information itself: who knows what, who claims what, who the vote believes.

Go deeper
STRUCTURE
A minority of players hold hidden, asymmetric information and an opposing goal; the only tools are public claims, private tells, and periodic voting that aggregates imperfect, deliberately corrupted information into one binary group decision.
WHY IT IS LEVERAGE
It isolates information-flow leverage as the entire game, with no combat or economy layer diluting it — proof that controlling who-knows-what-and-when is enough on its own to make a complete, replayable game.
THE DRIVER
Social trust-testing — the thrill of deceiving or detecting deception among real people.
THE SHADOW
Breaks down without a critical mass willing to actually deceive well, and voting-mob dynamics can punish confident innocents as readily as the guilty.
EXAMPLES
Among Us, Werewolf / Mafia (tabletop origin), Secret Hitler
ANALOGS
Information asymmetry and signaling games (game theory); Sybil resistance and reputation-based trust in pseudonymous voting (DAO governance)
DATA
PLAY · Needs · RUNG 6

Self-Determination Theory & PENS

Autonomy, competence, relatedness: the three needs a mechanic feeds or fakes.

Go deeper
STRUCTURE
Three need-stocks fill via loops of real choice, scalable mastery, and genuine bonds; starve one and a balancing loop drags enjoyment and retention down with it.
WHY IT IS LEVERAGE
Real choice, right-sized challenge, and real relationships raise intrinsic motivation and post-play wellbeing — need satisfaction is the renewable resource, not the score.
THE DRIVER
The innate need to feel self-caused, capable, and connected to others.
THE SHADOW
Simulated needs without satisfaction — fake choice, scripted competence, parasocial 'friends' — keep players chasing a satisfaction the mechanic never delivers.
EXAMPLES
The Legend of Zelda: Breath of the Wild, Animal Crossing: New Horizons, World of Warcraft
ANALOGS
Reinforcing competence→engagement loop (systems); Governance tokens granting real vs. cosmetic voice (chain)
DATA
PLAY · Reward · RUNG 6

Dopamine Reward-Prediction Error

The brain doesn't reward the win — it rewards being wrong about the win.

Go deeper
STRUCTURE
Dopamine neurons encode the gap between predicted and received reward; as learning consolidates, the signal shifts from the reward itself to the cue that predicts it, building a loop around cues, not outcomes.
WHY IT IS LEVERAGE
Clear cues that reliably predict meaningful rewards let players build accurate mental models and feel real mastery — the cue becomes a trustworthy signal of progress.
THE DRIVER
Midbrain dopamine neurons fire to the mismatch between predicted and actual reward, not to the reward's raw size.
THE SHADOW
Reward cues deliberately decoupled from real value (slot lights, loot box animations) hijack the prediction-error signal to manufacture craving for cues that track nothing.
EXAMPLES
Loot box opening animations, Slot machine lights and sounds, Mobile game win-streak VFX
ANALOGS
Information signal decoupled from the stock it once tracked (systems); Engagement-farming points decoupled from real protocol value (chain)
DATA
  • Dopaminergic neurons in primate midbrain signal errors in reward prediction, and their response shifts over learning from reward delivery to the predictive cue itself: foundational reward-prediction-error finding its.caltech.edu/~jkenny/nb250c/papers/Schultz-1997.pdf · 1997-03
PLAY · Economy · RUNG 7

Idle / Incremental Games

The genre that removes the player from the loop and sells the pure shape of exponential growth.

Go deeper
STRUCTURE
A number increases on its own, an automated flow into a stock, at a rate that itself rises with the stock (a reinforcing loop with no balancing loop); the only actions left are spending the stock to raise the rate, or prestiging — zeroing the stock to raise a permanent multiplier.
WHY IT IS LEVERAGE
Systems thinking's rung-7 lesson made into entertainment: with no balancing loop at all, the genre is a pure demonstration that unchecked exponential growth either hits a wall or gets deliberately reset — no third option, in a game or any real system.
THE DRIVER
Watching a number grow on its own, plus small bursts of agency spaced across an automated process.
THE SHADOW
Taken as a monetization template rather than a design study, it reduces to selling multiplier upgrades for cash with the decision space shrunk to one repeated purchase.
EXAMPLES
Cookie Clicker, Universal Paperclips, Melvor Idle
ANALOGS
Unchecked reinforcing loop leading to collapse or deliberate reset (Meadows' central warning); Hyperinflationary token emission with no sink, versus a deliberate supply-burn reset (crypto)
DATA
  • Cookie Clicker release date and early viral growth figures: Released August 8, 2013 by Julien 'Orteil' Thiennot; reached 50,000 players within hours of being posted to 4chan, over 200,000 players per day within a month, and peaked at 1.5 million hits in a single day during August 2013 en.wikipedia.org/wiki/Cookie_Clicker · 2013-08
PLAY · Reward · RUNG 7

Operant Conditioning, Variable-Ratio Schedules & the Near-Miss Effect

The unpredictable reward is the most addictive reward there is.

Go deeper
STRUCTURE
Reward is released on a random-ratio schedule instead of a fixed one; near-misses (unrewarded trials landing close to a win) recruit the same dopaminergic response as a real win, steepening the reinforcing loop.
WHY IT IS LEVERAGE
Unpredictable positive feedback — a critical hit, a rare drop — sustains exploration and persistence without requiring constant payout, the same structure that makes discovery fun.
THE DRIVER
Variable-ratio reinforcement produces the highest resistance to extinction of any known reward schedule.
THE SHADOW
Gacha and loot-box pulls tune near-miss frequency deliberately to maximize spend and session length, producing compulsive repetition indistinguishable from gambling.
EXAMPLES
Genshin Impact gacha, Casino-style slot machines, Candy Crush Saga near-win screens
ANALOGS
Reinforcing loop with stochastic gain (systems); Unpredictable airdrop timing driving repeated low-value actions (chain)
DATA
  • Variable-ratio reinforcement schedules produce the highest response rate and greatest resistance to extinction of the basic operant conditioning schedules: classic operant-conditioning finding en.wikipedia.org/wiki/Reinforcement · 2026-09
  • Near-misses in a gambling task activated reward-related brain circuitry similarly to wins, and increased self-reported motivation to continue play: fMRI near-miss study doi.org/10.1016/j.neuron.2008.12.031 · 2009-02
PLAY · Social · RUNG 7

Status & Social Comparison

A leaderboard turns everyone else's progress into your own to-do list.

Go deeper
STRUCTURE
Visible relative rank feeds a reinforcing loop: a drop in standing increases effort to restore it, which can raise others' standing in turn, restarting the cycle.
WHY IT IS LEVERAGE
Transparent, skill-based rankings give players an honest external yardstick and a real reason to improve against real peers.
THE DRIVER
People evaluate their own standing primarily in relation to others, not against an absolute scale.
THE SHADOW
Rankings seeded with whales, bots, or pay-to-rank mechanics turn comparison into a status treadmill escapable only by spending, not by skill.
EXAMPLES
Clash Royale trophy ladder, Peloton leaderboards, Snapchat Snap Score
ANALOGS
Reinforcing status loop (systems); On-chain leaderboards and NFT status signaling (chain)
DATA
  • Social comparison theory holds that people determine their own social and personal worth largely by how they measure up against others, a foundational driver behind competitive game design: Festinger, 1954 en.wikipedia.org/wiki/Social_comparison_theory · 2026-09
PLAY · Feel · RUNG 8

Flow Channel & Dynamic Difficulty Adjustment

A balancing loop that holds challenge just above current skill.

Go deeper
STRUCTURE
DDA samples a player-state proxy (deaths, time-to-kill, accuracy) and feeds it back to adjust enemy stats or hidden aid in real time, holding challenge inside the flow channel instead of letting static difficulty drift as skill rises.
WHY IT IS LEVERAGE
It's a literal thermostat: the deepest lever in any difficulty system is whether a balancing loop exists at all — games that 'feel unfair' usually just lack one.
THE DRIVER
Flow state itself — total absorption from challenge matched to skill.
THE SHADOW
Invisible DDA that over-corrects reads as the game 'cheating' once players notice rubber-banding, destroying trust that challenge is real.
EXAMPLES
Left 4 Dead (AI Director), Resident Evil 4, Mario Kart
ANALOGS
Balancing loop / thermostat (cybernetics, Meadows rung 8); Algorithmic difficulty retarget holding block time constant (crypto)
DATA
  • Flow state as the challenge-skill balance underlying difficulty design: Mihaly Csikszentmihalyi's flow-channel model, widely applied to difficulty design since the 1990s and still cited in design retrospectives gamedeveloper.com/design/revisiting-the-mda-framework · 2021-05
PLAY · Competition · RUNG 8

Ranked Matchmaking & Elo / MMR

A balancing loop that pairs players so the predicted outcome is close to a coin flip.

Go deeper
STRUCTURE
A skill estimate (Elo, Glicko, TrueSkill) updates each match by surprise (result versus predicted odds); matchmaking then pairs opponents to minimize predicted gap — a negative-feedback loop driving every player's experienced win rate toward about 50%, regardless of absolute skill.
WHY IT IS LEVERAGE
The leverage isn't the rating formula, it's that matchmaking converts an open-ended competence ladder into a closed-loop regulator — the game stays equally 'hard' for a beginner and a grandmaster.
THE DRIVER
Perceived fairness and the possibility of winning the very next match, at any skill level.
THE SHADOW
Pure 50%-win-rate targeting can produce 'Elo hell' perception and is exploitable by win-trading or boosting.
EXAMPLES
Chess (FIDE), League of Legends, Rainbow Six Siege
ANALOGS
Negative-feedback stabilization (cybernetics, Meadows rung 8); Reputation-weighted matching for DAO contributor allocation
DATA
  • Elo system origin and adoption into video game matchmaking: Developed by Arpad Elo in the 1950s for US chess ratings; adopted by FIDE in 1970; later extended into Glicko, TrueSkill, and Elo-MMR for multiplayer video games en.wikipedia.org/wiki/Skill-based_matchmaking · 1970-01
PLAY · Needs · RUNG 8

Flow & the Challenge-Skill Balance

The channel between boredom and anxiety where time disappears.

Go deeper
STRUCTURE
A balancing loop holds rising challenge against rising skill in a narrow band; drift out of band flips the system into boredom (challenge too low) or anxiety (too high).
WHY IT IS LEVERAGE
Legible goals, instant feedback, and difficulty tracking ability produce deep, self-rewarding absorption with no external reward needed.
THE DRIVER
The need for immediate feedback on a task matched precisely to current ability.
THE SHADOW
Difficulty retuned to maximize session length or spend instead of flow — engineered frustration spikes that break the channel on purpose, right before a paid shortcut appears.
EXAMPLES
Tetris, Celeste, Candy Crush Saga
ANALOGS
Balancing loop with a narrow stable band (systems); Difficulty/emission curve tuned to retention instead of fit (chain)
DATA
  • Flow is defined by a matched challenge-skill ratio producing intrinsically rewarding absorption and time distortion: foundational construct (Csikszentmihalyi, 1990) en.wikipedia.org/wiki/Flow_(psychology) · 2026-09
PLAY · Ethics · RUNG 8

Loot Boxes, Problem Gambling & the Regulatory Response

When a feedback loop looks enough like a slot machine, regulators treat it like one.

Go deeper
STRUCTURE
A real-money flow buys a randomized reward with no guaranteed value, structurally identical to a wager; without a balancing loop (disclosed odds, spend caps, regulation) the reinforcing loop runs unchecked until an external authority imposes one.
WHY IT IS LEVERAGE
Transparent odds, spend caps, and removing real-money randomization keep the delight of surprise rewards without the gambling-adjacent harm.
THE DRIVER
A randomized reward purchased with real money reproduces the core structure of a game of chance, regardless of what the reward looks like.
THE SHADOW
Undisclosed odds and no spend caps externalize gambling harm onto players, including minors, without necessarily triggering gambling law.
EXAMPLES
FIFA Ultimate Team packs, Overwatch loot boxes (pre-removal), Genshin Impact gacha wishes
ANALOGS
Unchecked reinforcing loop corrected only by an external balancing loop (systems); Regulatory action against token structures with lottery-like payouts (chain)
DATA
PLAY · Economy · RUNG 9

Live Service Seasons & Battle Pass

A fixed-length delay that manufactures urgency and recurring revenue.

Go deeper
STRUCTURE
A season is a hard delay (roughly 10 weeks) on a visible progress-bar stock; the pass is a sunk-cost pre-purchase raising the marginal value of each session (XP 'unlocks' pre-paid rewards) — expiry pressure converts latent players into active ones near season end.
WHY IT IS LEVERAGE
It decouples monetization from any single piece of content: the structure, a timer plus a visible track, generates repeatable revenue and return visits independent of whether that season's content is actually good.
THE DRIVER
Sunk-cost completion drive plus loss aversion on an expiring, already-purchased reward track.
THE SHADOW
The same expiry delay makes it predatory for lapsed players, and stacks across simultaneous passes into 'pass fatigue'.
EXAMPLES
Fortnite, Apex Legends, Call of Duty
ANALOGS
Delay as a deliberate leverage point (Meadows rung 9); DeFi epoch and season reward-emission cycles (crypto)
DATA
PLAY · Bias · RUNG 9

Loss Aversion

Losing a streak hurts about twice as much as gaining one helps.

Go deeper
STRUCTURE
A reference point splits outcomes into gain and loss framing on the same stock; the loss branch of the value function is roughly twice as steep, so 'losing X' pulls harder than 'not yet gaining X.'
WHY IT IS LEVERAGE
Protecting a player's hard-won state — a base, a city, a save — against real threats makes stakes feel meaningful and victories earned.
THE DRIVER
Losses are weighted roughly twice as heavily as equivalent gains in human decisions under risk.
THE SHADOW
Framing ordinary non-purchase as a 'loss' — expiring streaks, decaying resources, countdown timers on owned items — manufactures artificial stakes to drive a defensive purchase.
EXAMPLES
Duolingo streak freezes, Clash of Clans shield decay, Fortnite locker countdowns
ANALOGS
Asymmetric response curve around a setpoint (systems); Impermanent-loss framing used to retain liquidity providers (chain)
DATA
  • Losses loom larger than equivalent gains, with loss aversion coefficients commonly estimated near 2:1 to 2.5:1 in prospect-theory experiments: ~2:1-2.5:1 jstor.org/stable/1914185 · 1979-03
PLAY · Bias · RUNG 9

Endowed Progress Effect & the Goal-Gradient Hypothesis

A head start you didn't earn still makes you run faster to the finish.

Go deeper
STRUCTURE
A visible progress stock toward a fixed goal creates accelerating effort as the remaining gap narrows; artificially inflating the starting level of that stock raises completion even though the true remaining distance is unchanged.
WHY IT IS LEVERAGE
Visible, honest progress toward a real, attainable goal helps people start and finish things they actually value — momentum is a legitimate motivator.
THE DRIVER
Effort and pursuit speed rise as the perceived distance to a goal shrinks, regardless of how the 'already covered' portion was earned.
THE SHADOW
Fake head starts (pre-filled progress bars) and post-reward resets that instantly reopen a new gradient exploit the same pull purely to extend engagement or spend.
EXAMPLES
Starbucks Rewards stars, Battle pass progress tracks, Duolingo unit progress bars
ANALOGS
Accelerating flow as a stock nears its target (systems); Vesting-cliff proximity accelerating token-holder activity (chain)
DATA
PLAY · Bias · RUNG 9

Zeigarnik Effect, Open Loops & Completionism

An unfinished task rents space in your head until the loop closes.

Go deeper
STRUCTURE
An opened-but-incomplete task becomes an intrusive-memory stock that drains only on completion; quest logs and collection sets multiply open loops in parallel, each exerting its own small pull.
WHY IT IS LEVERAGE
Clear quest logs, checklists, and collections give players a legible map of what's left, turning sprawling worlds into achievable, satisfying arcs.
THE DRIVER
Incomplete tasks are remembered better and intrude on attention more than completed ones.
THE SHADOW
Deliberately proliferating open loops — overlapping daily quests, 'collect them all' sets with no real payoff — manufactures chronic low-grade anxiety only a return visit relieves.
EXAMPLES
World of Warcraft quest logs, Pokémon living Pokédex, Mobile game daily-quest boards
ANALOGS
Unresolved stock pulling on a delay until closed (systems); Unclaimed airdrop allocations kept open to pull users back (chain)
DATA
  • Unfinished tasks are retained in memory and recalled significantly better than completed tasks, the original experimental basis for the Zeigarnik effect: Zeigarnik, 1927 en.wikipedia.org/wiki/Zeigarnik_effect · 2026-09
PLAY · Bias · RUNG 9

Streaks & Sunk Cost

The habit becomes the hostage: breaking it costs more than keeping it ever did.

Go deeper
STRUCTURE
A streak counter rises with each consecutive daily action; resistance to breaking it compounds with length, converting a motivational tool into a sunk-cost trap that can outlive the goal it was built to serve.
WHY IT IS LEVERAGE
Honest streaks build the real habit they claim to build — visible consistency is fair positive reinforcement for showing up and compounds into real skill.
THE DRIVER
People continue a behavior to avoid 'wasting' the investment already made, even once the behavior stops serving its original purpose.
THE SHADOW
Streak mechanics monetize the fear of loss directly (paid streak-freezes) and keep people opening an app out of dread rather than desire, long after the habit's real value has decayed.
EXAMPLES
Duolingo streaks, Snapchat Snapstreaks, Wordle streak tracking
ANALOGS
A stock whose accumulated history raises the cost of letting it fall to zero (systems); Staking lock-ups that punish unstaking via forfeited rewards (chain)
DATA
PLAY · Loops · RUNG 10

Core Loop vs. Mastery Loop

The second-to-second action nested inside the hours-long arc that gives it stakes.

Go deeper
STRUCTURE
A fast reinforcing loop (act, feedback, reward, act again) repeats every few seconds, nested inside a slower mastery loop (learn, apply, master, new challenge) that re-arms the fast loop with new stakes every few hours.
WHY IT IS LEVERAGE
Leverage sits in the nesting, not either loop alone: a fast loop with no slow loop above it is a slot machine; a slow loop with no fast loop beneath it is homework. Outliers tune the gear ratio.
THE DRIVER
Competence feedback plus anticipation of the next reward tick.
THE SHADOW
When only the fast loop is tunable (drop rates, timers), it gets optimized for monetization and the slow loop atrophies into a treadmill.
EXAMPLES
Destiny 2, World of Warcraft, Tetris
ANALOGS
Reinforcing loop nested in a slower reinforcing loop (systems); Daily claim loop nested in a season-long token emission arc (crypto)
DATA
  • Loops described as fractal, occurring at multiple nested levels and frequencies: Daniel Cook's 'Loops and Arcs' (Lostgarden) states loops are fractal and occur at multiple levels and frequencies, and that nested, dependent loops yield complex feedback dynamics and unexpected emergent behavior — the structural basis for a fast loop nested inside a slower one lostgarden.com/2012/04/30/loops-and-arcs/ · 2012-04
PLAY · Progression · RUNG 10

Progression: XP, Levels, Skill Trees & Prestige

Visible stock accumulation that converts time-on-task into legible growth.

Go deeper
STRUCTURE
XP flows into a Level stock with a rising cost curve (each level needs more than the last, a drag on the loop); skill trees gate new flows behind stock thresholds; prestige zeroes the stock but raises a hidden multiplier, restarting the loop steeper.
WHY IT IS LEVERAGE
It's the cheapest high-leverage retention device in the medium: it needs no new content, only re-labeling existing content as 'not yet mastered', extending the mastery loop's arc almost indefinitely.
THE DRIVER
Sunk-cost investment plus visible, legible growth — a number that only goes up.
THE SHADOW
When the stock becomes the point instead of the play, the game is fully solvable by a spreadsheet and the human stops making real decisions.
EXAMPLES
RuneScape, Diablo III (Paragon levels), Path of Exile
ANALOGS
Stock accumulation with rising outflow cost (stock-and-flow, systems); Vesting and staking tiers, loop-resetting re-stake mechanics (crypto)
DATA
  • Peer-reviewed study jointly modeling XP/level progression against player churn in a freemium RPG: Karmakar, Liu, Mukherjee, Che & Dutta, 'Improved Retention Analysis in Freemium Role-Playing Games by Jointly Modelling Players' Motivation, Progression and Churn' (Journal of the Royal Statistical Society Series A, vol 185, issue 1): dropout hazard rate peaked at 54% at level 14, with engagement rising sharply again after level 30 academic.oup.com/jrsssa/article/185/1/102/7068450 · 2022-01
PLAY · Economy · RUNG 10

Internal Economy: Faucets, Sinks & Machinations

Every in-game resource is a stock; design is choosing its faucets, sinks, and loops.

Go deeper
STRUCTURE
Joris Dormans' Machinations models a game's internal economy as nodes (resource stocks) and rate-controlled edges (faucets add, sinks remove) in a simulatable diagram that exposes runaway loops and starvation states before code is written.
WHY IT IS LEVERAGE
It turns 'balance' from a vibe into a testable structure — you can run the diagram forward and watch a currency inflate, the same way you'd model a real economy.
THE DRIVER
Players want resource decisions (spend now vs. hoard) to matter; designers want no runaway inflation or deadlock.
THE SHADOW
A sink-starved economy (faucets outrun sinks) inflates until the currency is worthless — the most common systemic failure in live-service economies.
EXAMPLES
Diablo's gold economy, EVE Online, any MMO crafting economy
ANALOGS
Stock-and-flow diagram (Forrester / Meadows systems dynamics); Tokenomics emission and burn modeling (crypto)
DATA
PLAY · Progression · RUNG 11

Roguelike Meta-Progression

Losing a run resets the stock that matters moment-to-moment, not the one that matters across runs.

Go deeper
STRUCTURE
Two parallel stocks: an in-run stock (health, gold, temp boons) that fully resets on death, and a persistent meta stock (Hades' Darkness, unlocked weapons, story) that survives death and permanently lowers the difficulty floor — failure becomes a deposit, not a dead end.
WHY IT IS LEVERAGE
The one mechanic that makes repeated failure feel like progress instead of punishment, letting roguelikes break out of a niche built on punishing permadeath into mainstream, narrative-driven hits.
THE DRIVER
Loss aversion defused because death no longer threatens the stock the player is most attached to.
THE SHADOW
If the meta stock grows powerful enough, it trivializes the in-run skill test the genre is built on, turning the run into a grind that plays itself.
EXAMPLES
Hades, Rogue Legacy, Dead Cells
ANALOGS
Persistent buffer absorbing shocks across cycles (Meadows rung 11); Reputation or history that survives a failed venture (DAO / crypto)
DATA
  • Hades sales trajectory around full launch: Surpassed 1 million copies sold worldwide shortly after full launch in September 2020, after about 700,000 units sold during roughly two years of Early Access thegamer.com/hades-sales-1-million-copies/ · 2020-10
PLAY · Economy · RUNG 12

Gacha & Loot Boxes

A randomized-reward sink that monetizes the variance itself, not the item.

Go deeper
STRUCTURE
A published drop-rate parameter (increasingly disclosure-mandated in China, Japan, the EU) converts a currency stock into a weighted random draw; pity timers form a balancing loop capping worst-case variance so 'one more pull' doesn't lose players to bad luck entirely.
WHY IT IS LEVERAGE
The leverage point is one number — the drop rate — hidden behind a probabilistic interface; shift it half a percent and lifetime revenue per payer moves more than any content change could.
THE DRIVER
Variable-ratio reinforcement, the schedule behind slot-machine behavior, stacked on a collection drive.
THE SHADOW
Extracts disproportionately from the heaviest-variance-exposed players and has triggered loot-box regulation in Belgium, the Netherlands, and disclosure laws elsewhere.
EXAMPLES
Genshin Impact, Fate/Grand Order, FIFA Ultimate Team
ANALOGS
Variable-ratio reinforcement schedule (behavioral psychology); Randomized NFT mint / mystery-box mechanics (crypto)
DATA
SYSTEMS · Strategy · RUNG 1

Meadows' Dancing with Systems

You cannot control a system, only dance with it, and that is its own discipline.

Go deeper
STRUCTURE
Practitioner precepts for acting inside systems you cannot fully model: get the beat, expose your mental models, honor information, use language with care, attend to what matters not just what is quantifiable, stay humble, and know you can never fully control, only dance.
WHY IT IS LEVERAGE
It is Meadows' own capstone correction to over-confident modeling, a reminder from the person who built the leverage-points ladder that the ladder is a map, not the system, and humility is itself a leverage move.
THE DRIVER
the modeler's or builder's overconfidence that a sufficiently good model grants control over the system it describes
THE SHADOW
just dance gets used to excuse skipping the rigorous work of loops, stocks, and leverage points, when Meadows intended dancing as a posture on top of that rigor, not instead of it.
EXAMPLES
Donella Meadows, Dancing With Systems, Whole Earth, Winter 2001, excerpted from her unfinished final manuscript, Republished in The Systems Thinker, March 2002, Used as a closing text in Academy for Systems Change curricula
ANALOGS
Game design: live-service designers treat balance as an ongoing dance with player behavior, not a solved equation; DAO: governance participants treat the protocol as a living system to stay humble before, not a machine fully controlled by any single actor
DATA
SYSTEMS · Structure · RUNG 2

Meadows' Twelve Leverage Points

A ranked ladder of where to push on a system, cheapest to most transformative.

Go deeper
STRUCTURE
Meadows ranks 12 intervention points from weakest (parameters, rung 12) to strongest (transcending paradigms, rung 1): buffers, stock-flow structure, delays, loop strength, information flows, rules, self-organization, goals, paradigm.
WHY IT IS LEVERAGE
Builders default to tuning parameters (rung 12) because it is legible and safe; the ladder shows the same effort spent on rules, goals, or paradigm yields far more change.
THE DRIVER
where attention and effort get spent when a system misbehaves
THE SHADOW
used as a cudgel for dismissing low-rung work without doing the harder diagnosis of which rung is actually binding.
EXAMPLES
Inspired by Meadows observing NAFTA's design in the early 1990s, Taught as the entry framework in Academy for Systems Change practitioner training, Opening module of systemsinnovation.network's leverage points course
ANALOGS
Game design: tuning a drop rate (parameter) versus rewriting the victory condition (goal or paradigm); Crypto or DAO: adjusting a fee parameter versus rewriting the governance constitution (rules or paradigm)
DATA
SYSTEMS · Change · RUNG 3

Three Horizons and the Berkana Two Loops

Two views of one transition: overlapping curves, or a dying system and a born one.

Go deeper
STRUCTURE
Sharpe's Three Horizons plots the declining dominant system, disruptive alternatives, and the viable future as overlapping waves; Berkana's Two Loops draws a legacy system declining into compost beside a separate emergent arc, linked only by pioneers, hospicers, and connectors.
WHY IT IS LEVERAGE
Both give builders permission to work on the new system without waiting for the old one's consent or collapse; the new loop does not need the old loop's approval to be real.
THE DRIVER
assuming change must happen by reforming the dominant system from inside rather than building a parallel one
THE SHADOW
emergent-loop status gets used to justify ignoring the legacy system entirely, abandoning resources it still controls that could fund the transition.
EXAMPLES
Bill Sharpe and the International Futures Forum, UK Foresight Intelligent Infrastructure Futures project, Berkana Institute's Using Emergence to Take Social Innovation to Scale, Wheatley and Frieze, Renewable energy build-out proceeding alongside, not replacing, fossil infrastructure mid-transition
ANALOGS
Game design: sunsetting a legacy mode while a new mode is built in parallel, not patched onto the old one; DAO: a legacy governance process left to wind down while a sub-DAO pioneers a different model, rather than a hard-fork fight
DATA
SYSTEMS · Cybernetics · RUNG 4

Viable System Model and POSIWID

Five recursive functions keep an org viable; judged by what it does, not its claims.

Go deeper
STRUCTURE
Beer's Viable System Model specifies five necessary subsystems, operations, coordination, control, intelligence, and policy, recursively nested at every organizational level; POSIWID states the purpose of a system is what it does.
WHY IT IS LEVERAGE
VSM is a diagnostic checklist for which of the five functions is missing when an organization or DAO is failing; POSIWID cuts through stated-intent debates by forcing you to read actual output as the real goal.
THE DRIVER
confusing a system's documented mission with its operating behavior
THE SHADOW
POSIWID gets weaponized as cynicism about what a system does without the VSM diagnostic work of locating which missing subsystem produced the gap.
EXAMPLES
Beer's Chilean Project Cybersyn, 1971 to 1973, an early real-time economic VSM, DAOs with no environment-scanning function blindsided by forks or competitors, The purpose of a system is what it does, applied to engagement-optimized social media
ANALOGS
Game design: a live game's ops, live-balance, community-intel, and design-policy teams mirror the five VSM functions; DAO: judging a protocol's real purpose by its emergent token flows, POSIWID, rather than its whitepaper
DATA
SYSTEMS · Complexity · RUNG 4

Emergence and Self-Organization

Simple local rules, no central director, produce structure no single rule predicted.

Go deeper
STRUCTURE
Agents following local interaction rules with no central controller generate macro-level patterns that cannot be predicted by inspecting any single agent's rule in isolation.
WHY IT IS LEVERAGE
It is the argument for designing rules and incentives rather than outcomes directly: the outcome you want has to emerge from rules you can actually enforce at the agent level.
THE DRIVER
the designer's urge to specify the outcome directly instead of the local rules that would produce it
THE SHADOW
it will emerge gets used to avoid structural design entirely; emergence without enforceable local rules just produces noise, not order.
EXAMPLES
Ant colony foraging trails, a pheromone-rule emergence, Conway's Game of Life, Santa Fe Institute's founding complexity research program, Open-source project governance emerging from contribution rules rather than a fixed roadmap
ANALOGS
Game design: emergent gameplay from simple rule interactions, the central thesis of Dormans' engineering emergence; DAO: emergent price discovery and coordination from simple token-incentive rules, with no central planner
DATA
  • Santa Fe Institute founded explicitly to study emergent complexity across physical, biological, and social systems: founded in Santa Fe, New Mexico santafe.edu/about/history · 1984
SYSTEMS · Change · RUNG 4

Geels' Multi-Level Perspective on Transitions

Transitions happen when a sheltered niche meets cracks opening in the dominant regime.

Go deeper
STRUCTURE
Three nested levels, landscape (slow macro trends), regime (the dominant locked-in configuration), and niche (protected spaces for radical innovation); a transition occurs when niche innovations mature just as landscape pressure destabilizes the regime.
WHY IT IS LEVERAGE
It explains why identical, better technology can fail for a decade and then suddenly win: it was waiting on a regime-destabilizing window, not a product improvement.
THE DRIVER
assuming a superior niche innovation should win on merit alone, independent of regime lock-in and landscape timing
THE SHADOW
the regime is not ready gets used indefinitely as an excuse not to strengthen the niche, instead of building it so it is ready the moment the window opens.
EXAMPLES
Geels 2002, Technological transitions as evolutionary reconfiguration processes, Research Policy 31, The transition from horse-drawn transport to the automobile, a classic MLP case study, Decentralized finance as a niche innovation awaiting a regulatory or banking-regime crack
ANALOGS
Game design: a niche genre breaking into the mainstream when a landscape shift such as streaming or mobile opens a window; DAO: on-chain governance as a niche awaiting a legal or regulatory regime crack to go mainstream
DATA
  • Multi-level perspective formalized across niche, regime, and landscape levels with a case study: Technological transitions as evolutionary reconfiguration processes: a multi-level perspective and a case-study, Research Policy 31(8-9), pages 1257 to 1274 sciencedirect.com/science/article/abs/pii/S0048733302000628 · 2002
SYSTEMS · Archetypes · RUNG 5

Archetype: Tragedy of the Commons

A shared resource rewards every user for over-using it until it collapses.

Go deeper
STRUCTURE
Multiple users share a regenerating resource; each user's local payoff rises with their own use, so aggregate use overshoots the resource's regeneration rate and the stock collapses.
WHY IT IS LEVERAGE
It is the default failure mode of any unmetered shared resource a builder creates, from compute to attention to a subsidy pool, unless access or feedback is structurally bounded.
THE DRIVER
individually rational use with no feedback loop connecting one user's draw-down to another's loss
THE SHADOW
education or goodwill campaigns get proposed in place of the structural fix (bounded access, metering, Ostrom-style governance), and the commons collapses anyway.
EXAMPLES
Hardin's overgrazed pasture framing, 1968, Overfishing of shared fisheries, Unmetered API limits exhausted by a few heavy actors, Gas-fee auctions and MEV extraction on public blockchains
ANALOGS
Game design: unbounded shared loot pools or server resources drained by griefing; DAO: a shared treasury drained by uncapped proposal spending
DATA
SYSTEMS · Archetypes · RUNG 5

Archetype: Success to the Successful

Whoever wins the first allocation round gets the resources to win every round after.

Go deeper
STRUCTURE
Two parties compete for a shared, limited resource; the better performer receives more of it, which further improves their relative performance, a reinforcing loop that starves the loser.
WHY IT IS LEVERAGE
It explains winner-take-all outcomes builders did not design for; the fix is decoupling the allocation rule from the outcome it rewards, not blaming the loser.
THE DRIVER
allocating resources by past performance instead of by need or potential
THE SHADOW
a merit-based framing hides that the allocation rule itself is the reinforcing loop, so no one audits the mechanism as the actual cause.
EXAMPLES
Internal capital or headcount allocated to whichever business unit already hit its numbers, Algorithmic feeds amplifying already-popular content, Staking rewards compounding to the largest existing stake
ANALOGS
Game design: matchmaking or rewards that compound a winning streak into rich-get-richer loot; Crypto: proof-of-stake reward compounding to whales with no caps
DATA
SYSTEMS · Archetypes · RUNG 5

Meadows' System Traps and Their Ways Out

Named, recurring structural traps, each with a structural exit, not a willpower fix.

Go deeper
STRUCTURE
Thinking in Systems catalogs traps including policy resistance, tragedy of the commons, drift to low performance (eroding goals), escalation, success to the successful, rule beating, and seeking the wrong goal, each a specific loop pathology with a specific fix.
WHY IT IS LEVERAGE
Naming the exact trap tells you the exact class of fix, for example escalation needs a ceiling or unilateral exit, not more negotiation, instead of generic try-harder advice.
THE DRIVER
treating a systemic failure as a one-off instead of a named, recurring structural pattern with a known exit
THE SHADOW
a trap's way out gets cited as a slogan without the structural work it requires, such as calling for better communication instead of changing the measured goal.
EXAMPLES
Policy resistance: competing actors pulling the same stock toward conflicting goals, Escalation: arms races, flame wars, gas-fee bidding wars, Drift to low performance: goals quietly ratcheted down to match eroding results, Seeking the wrong goal: optimizing a proxy metric instead of the real target
ANALOGS
Game design: PvP arms races needing a hard power ceiling, the escalation trap; DAO: governance participation metrics drifting down to match low turnout, drift to low performance
DATA
SYSTEMS · Commons · RUNG 5

Ostrom's Eight Design Principles for Governing the Commons

Shared resources survive without privatizing or centralizing when 8 conditions hold.

Go deeper
STRUCTURE
Clear boundaries, proportional costs and benefits, collective-choice rule-making, monitoring, graduated sanctions, fast and accessible conflict resolution, a recognized right to self-organize, and nested governance across scales.
WHY IT IS LEVERAGE
It is the most directly portable systems-thinking model for DAO governance design; every one of the eight maps to a concrete, auditable governance mechanism.
THE DRIVER
the false binary that shared resources must be either privatized or centrally regulated to avoid collapse
THE SHADOW
principles get cherry-picked, such as adding monitoring without graduated sanctions or accessible conflict resolution, producing surveillance without legitimacy.
EXAMPLES
Swiss alpine grazing commons, centuries old, Maine lobster fishing territories, Elinor Ostrom, Governing the Commons, 1990, Cited in early Ethereum and DAO governance design discourse
ANALOGS
Game design: player-run guild economies need the same eight conditions to avoid griefing collapse; DAO: graduated sanctions and nested sub-DAO governance directly implement principles five and eight
DATA
SYSTEMS · Structure · RUNG 5

Goodhart's Law and Campbell's Law

Once a measure becomes the target, people optimize the measure, not the goal.

Go deeper
STRUCTURE
Any observed statistical regularity breaks down once pressure is placed on it for control purposes, Goodhart; the more a quantitative indicator is used for social decision-making, the more it is corrupted and distorts the process it was meant to monitor, Campbell.
WHY IT IS LEVERAGE
It is the mechanism-design warning every builder of incentive systems needs before choosing a KPI, a token-emission trigger, or a governance voting-weight metric: the metric will be gamed in proportion to the weight placed on it.
THE DRIVER
treating a convenient proxy metric as if it were the actual goal, under optimization pressure
THE SHADOW
you cannot measure anything then becomes the excuse to abandon measurement entirely, instead of using multiple, hard-to-game, outcome-linked measures with real stakes attached.
EXAMPLES
Charles Goodhart's UK monetary policy critique, Donald T. Campbell's social-indicator corruption thesis, Standardized-test teaching to the test, Token-airdrop farming that optimizes the snapshot metric instead of real usage
ANALOGS
Game design: players optimize a visible score or achievement metric in ways that break the intended play experience; DAO: governance-token distribution farmed via wash trading once trading volume becomes the airdrop target
DATA
  • Goodhart's original framing of control-induced regularity breakdown: any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes en.wikipedia.org/wiki/Goodhart%27s_law · 1975
  • Campbell's Law on corruption of quantitative social indicators under decision pressure: formulated independently, generally dated to the 1970s en.wikipedia.org/wiki/Goodhart's_law · 1979
SYSTEMS · Strategy · RUNG 5

Mechanism Design

Work backward from the outcome, and engineer rules so self-interest produces it.

Go deeper
STRUCTURE
Mechanism design, reverse game theory, starts from a desired outcome and designs the game form, rules, information structure, and payoffs, such that rational, self-interested, privately informed agents produce that outcome in equilibrium.
WHY IT IS LEVERAGE
It is the formal bridge from systems thinking's change-the-rules leverage point to actual incentive engineering, the discipline underneath every well-built auction, voting system, or token mechanism.
THE DRIVER
designing an institution around how people should behave rather than how self-interested, privately informed people will actually behave under its rules
THE SHADOW
a mechanism proven correct in theory, incentive-compatible and individually rational, fails in practice because real participants collude or have information the model never assumed.
EXAMPLES
Leonid Hurwicz, Eric Maskin, Roger Myerson, Nobel Memorial Prize in Economic Sciences, 2007, Myerson's revelation principle, Vickrey second-price auctions, EIP-1559's fee mechanism and quadratic voting in DAO governance
ANALOGS
Game design: auction-house and matchmaking rules designed so self-interested play produces balanced, fun outcomes; Crypto or DAO: bonding curves, EIP-1559 burn, and quadratic funding are literal applied mechanism design
DATA
SYSTEMS · Cybernetics · RUNG 6

Ashby's Law of Requisite Variety

Only variety can destroy variety: a controller needs disturbance-matching range.

Go deeper
STRUCTURE
For a regulator to hold a system within a target set of states against disturbance, the regulator's repertoire of possible responses must be at least as large as the variety of disturbances it faces.
WHY IT IS LEVERAGE
It is the precise, falsifiable reason rigid rulebooks fail against messy reality, and the argument for building adaptive response capacity instead of writing more rules.
THE DRIVER
the urge to control complexity with a bigger fixed rulebook instead of growing the system's own capacity to respond
THE SHADOW
cited to justify unlimited discretionary power, skipping Ashby and Beer's actual point that variety must be deliberately matched and attenuated, not simply maximized.
EXAMPLES
Ashby's homeostat machine demonstrations, 1948 to 1952, Fixed customer-service scripts failing against novel complaints, Decentralized incident response outperforming one fixed runbook
ANALOGS
Game design: AI opponents need response variety matching player strategy variety or the game feels solved; DAO: governance needs enough decision-mechanism variety (signaling, delegation, sub-DAOs) to match the variety of proposals it faces
DATA
SYSTEMS · Strategy · RUNG 6

Cynefin Framework

Clear, complicated, complex, chaotic: different domains demand different decisions.

Go deeper
STRUCTURE
Cynefin sorts situations into five domains, clear (best practice), complicated (good practice, expert analysis), complex (emergent, probe-sense-respond), chaotic (act-sense-respond), and confused, each requiring a different decision approach.
WHY IT IS LEVERAGE
Most strategic failure is domain misdiagnosis, applying clear-domain best practice to a complex-domain problem that actually needs safe-to-fail probes.
THE DRIVER
defaulting to the one decision style a leader or org is trained in regardless of which domain the situation is actually in
THE SHADOW
calling a problem complex rather than complicated gets used to excuse skipping analysis that would have worked, or to dodge accountability for a decision.
EXAMPLES
Snowden and Boone, A Leader's Framework for Decision Making, Harvard Business Review, November 2007, Used across NATO and UK government policy design, Agile probe-sense-respond framing for product discovery in the complex domain
ANALOGS
Game design: tutorial and onboarding (clear) versus late-game emergent strategy space (complex) needing different design treatment; DAO: routine treasury operations (complicated, expert process) versus exploit response (chaotic, act-sense-respond)
DATA
SYSTEMS · Strategy · RUNG 6

OODA Loop

Observe, orient, decide, act: win by cycling faster than your rival makes sense of it.

Go deeper
STRUCTURE
A continuous decision cycle, observe (gather data), orient (make sense of it via mental models), decide (choose a course), act (execute, generating new data to observe), where advantage comes from completing the loop faster than an adversary's loop.
WHY IT IS LEVERAGE
It reframes competitive strategy around tempo and the orient step rather than raw resources; a builder who reorients faster beats a better-funded rival stuck on stale assumptions.
THE DRIVER
the belief that superior resources or a better plan win, rather than superior tempo of sensing and reorienting
THE SHADOW
move fast gets cited while skipping orient entirely; pure speed without updated mental models is just acting on stale orientation, faster.
EXAMPLES
Colonel John Boyd, US Air Force fighter-pilot tactics, 1970s, Used widely in business strategy and cybersecurity incident response, Agile and lean-startup build-measure-learn as a civilian OODA variant
ANALOGS
Game design: real-time strategy games directly score players on OODA tempo through micro and macro decision cycles; DAO: a protocol that reorients governance faster after an exploit than attackers can chain the next one
DATA
  • Boyd fused thermodynamics, cybernetics, and evolutionary epistemology into a four-stage competitive decision loop emphasizing tempo over raw power: OODA loop, Patterns of Conflict briefings en.wikipedia.org/wiki/OODA_loop · 1976
SYSTEMS · Structure · RUNG 6

Conway's Law

Your product's architecture copies your org chart, whether you planned it or not.

Go deeper
STRUCTURE
Any organization that designs a system will produce a design whose structure mirrors the organization's own communication structure, because components can only be built compatible by teams that actually talk to each other.
WHY IT IS LEVERAGE
It predicts your technical or governance architecture from your team topology before a line of it is written, and says the fastest way to change the architecture is to change who talks to whom.
THE DRIVER
assuming architecture is a purely technical choice, independent of who communicates with whom while building it
THE SHADOW
cited to justify reorganizing teams to mirror a desired architecture, the inverse Conway maneuver, without accounting for the real cost and disruption of the reorg itself.
EXAMPLES
Melvin Conway, How Do Committees Invent, Datamation, April 1968, Amazon's two-pizza team and API-mandate org design, A DAO's siloed working groups producing siloed, incompatible sub-protocols
ANALOGS
Game design: a game's level or feature seams mirror which studio teams do not talk to each other; DAO: governance module boundaries mirror which working groups actually coordinate versus operate in silos
DATA
  • Verbatim core statement of the law: any organization that designs a system, defined broadly, will produce a design whose structure is a copy of the organization's communication structure melconway.com/research/committees.html · 1968
SYSTEMS · Loops · RUNG 7

Reinforcing and Balancing Feedback Loops

Only two loop shapes exist: one amplifies, one corrects.

Go deeper
STRUCTURE
A reinforcing loop compounds a change in the same direction; a balancing loop opposes a change and seeks a goal. All complex system behavior composes from networks of these two loop types.
WHY IT IS LEVERAGE
Naming which loop drives a symptom tells you whether to interrupt it (reinforcing, runaway) or find its set point (balancing, stuck) instead of treating the symptom as a one-off event.
THE DRIVER
the compulsion to explain behavior by an external event instead of the loop structure producing it
THE SHADOW
mistaking a reinforcing loop for linear growth until it is exponential and unmanageable, or fighting a balancing loop's goal instead of changing the goal itself.
EXAMPLES
Viral referral loops (reinforcing), Thermostat and PID control (balancing), Bank runs and compound interest (reinforcing), Predator-prey population cycles (balancing with delay)
ANALOGS
Game design: XP compounding (reinforcing) versus rubber-banding or catch-up mechanics (balancing); DAO: token-price-to-speculation flywheels (reinforcing) versus algorithmic stablecoin pegs (balancing)
DATA
SYSTEMS · Complexity · RUNG 7

Network Effects and Preferential Attachment

Nodes with more connections get connected to faster: rich-get-richer network growth.

Go deeper
STRUCTURE
In a growing network, new nodes attach preferentially to already well-connected nodes, producing a scale-free degree distribution, a few hubs and a long tail, rather than a random or even one.
WHY IT IS LEVERAGE
It predicts, before launch, that any growing network you build will concentrate around a small number of hubs; design for that concentration instead of being surprised by it.
THE DRIVER
treating network growth as uniform or random when the real attachment rule is winner-reinforcing
THE SHADOW
network effects get invoked as inevitable to excuse not building the specific mechanism, referral, interoperability, liquidity depth, that would actually generate the preferential attachment.
EXAMPLES
World Wide Web link structure, Barabasi and Albert 1999, Social graph hub concentration among a few high-follower accounts, Liquidity concentration in a handful of decentralized exchange pools
ANALOGS
Game design: a handful of discovered meta builds dominating a crafting economy via attachment to the meta; Crypto: stake or liquidity concentrating in a few validators or pools despite nominal decentralization
DATA
  • Scale-free degree distributions arise from network growth plus preferential attachment to already well-connected nodes: Emergence of Scaling in Random Networks arxiv.org/abs/cond-mat/9910332 · 1999
SYSTEMS · Archetypes · RUNG 8

Archetype: Limits to Growth

A reinforcing engine hits a balancing limit it never modeled, and growth stalls.

Go deeper
STRUCTURE
A reinforcing loop drives accelerating growth until it feeds a balancing loop, a limit such as capacity or saturation, that was invisible at small scale; growth slows, flattens, or collapses.
WHY IT IS LEVERAGE
The limit is almost always discoverable in advance by asking what stock this growth consumes that is not being replenished, before the stall happens rather than after.
THE DRIVER
optimism bias that extrapolates the visible growth curve past an invisible constraint
THE SHADOW
teams respond to the slowdown by pushing the growth engine harder instead of addressing the limit, accelerating the eventual collapse.
EXAMPLES
Limits to Growth world model: population versus finite resources, Meadows et al. 1972, Viral apps hitting server capacity or moderation limits, Yield-farming TVL outrunning available real yield
ANALOGS
Game design: exponential resource-gathering loops capped by inventory or server limits; DAO: token emissions growth capped by real protocol revenue, a tell for ponzi-like dynamics
DATA
SYSTEMS · Archetypes · RUNG 8

Archetype: Shifting the Burden

The quick fix relieves the symptom and atrophies the capacity for the real cure.

Go deeper
STRUCTURE
A symptom is addressed by two competing balancing loops, a fast symptomatic fix and a slower fundamental fix; repeated use of the fast fix erodes the capacity or motivation to apply the slow one.
WHY IT IS LEVERAGE
Every temporary workaround that becomes permanent is this archetype; spotting the atrophying side loop catches addiction to the fix before it becomes structural.
THE DRIVER
the preference for relief now over capability later
THE SHADOW
the fix's side effect, dependency, is mistaken for a separate new problem, triggering another quick fix stacked on top of it.
EXAMPLES
Painkillers masking injury instead of physical therapy, Senge's canonical example, Hiring contractors instead of building internal capability, Subsidizing liquidity instead of fixing product-market fit
ANALOGS
Game design: buying power-ups instead of improving skill, eroding the mastery loop; DAO: mercenary liquidity mining instead of organic protocol demand
DATA
SYSTEMS · Loops · RUNG 9

Delays and Oscillation: the Beer Game

A gap between action and feedback turns a stable loop into a swing.

Go deeper
STRUCTURE
Delays in a balancing loop cause overcorrection; the longer the delay relative to the loop's natural response time, the larger and more persistent the resulting oscillation, known as the bullwhip effect.
WHY IT IS LEVERAGE
Builders act, see no effect yet, and act again before the first action has landed, stacking corrections that produce the exact boom-bust cycle they were trying to prevent.
THE DRIVER
impatience with lag: acting again before the last action has shown its effect
THE SHADOW
blaming irrational actors for oscillation that is actually a structural property of the delay, so the real fix (shorten the delay, expose the pipeline) never gets built.
EXAMPLES
MIT Beer Distribution Game, Sterman 1989, Real-world bullwhip effect in supply chains (e.g. Procter & Gamble diapers), Crypto liquidation cascades driven by oracle-price delay
ANALOGS
Game design: input lag or netcode delay breaking the feel of a combat loop; DAO: governance vote delay versus treasury drain rate producing late, oversized corrections
DATA
  • Beer Game players systematically underweight the supply line already in motion, amplifying orders into large oscillations: demonstrated across repeated MIT System Dynamics Group experiments web.mit.edu/jsterman/www/Order_Stability_070505.pdf · 1989
  • Original formalization of the misperception-of-feedback model behind the oscillation: Modeling managerial behavior: misperceptions of feedback in a dynamic decision making experiment, Management Science 35(3), 321-339 pubsonline.informs.org/doi/10.1287/mnsc.35.3.321 · 1989
SYSTEMS · Archetypes · RUNG 9

Archetype: Fixes that Fail

The fix works now and makes the problem worse on a delay you will not see yet.

Go deeper
STRUCTURE
A fix produces an immediate correction to a symptom but also triggers a delayed, unintended reinforcing side effect that regrows the original problem, often past its starting severity.
WHY IT IS LEVERAGE
The delay is what defeats you: by the time the problem returns, the fix has earned credibility and gets applied again, harder.
THE DRIVER
evaluating a fix only against its immediate, visible effect
THE SHADOW
the recurrence is read as the problem being worse than thought, calling for more of the same fix, instead of recognizing the fix itself has a side effect.
EXAMPLES
Antibiotic overuse breeding resistant bacteria, Bot rate-limiting that trains bots to evade detection, Emergency token buybacks that train sell-the-news behavior
ANALOGS
Game design: nerfing a strategy that only makes the next-best exploit dominant; DAO: emergency parameter patches that create new attack surface
DATA
  • Formalized as the fixes that fail archetype: a quick fix with an unintended, delayed reinforcing loop: Kim, Systems Archetypes I, 1992; Senge, The Fifth Discipline en.wikipedia.org/wiki/Fixes_that_fail · 1992
SYSTEMS · Structure · RUNG 10

Stocks and Flows

A system is a tub filling and draining; behavior is the plumbing, not the water.

Go deeper
STRUCTURE
A stock is an accumulation changed only by inflows and outflows; you can never change a stock directly, only change a rate feeding or draining it.
WHY IT IS LEVERAGE
Builders chase visible levels (price, active users, treasury balance) instead of the rates feeding them; naming the stock and its flows is the fastest diagnostic move available.
THE DRIVER
the instinct to manage the number you can see instead of the rate that produces it
THE SHADOW
dashboards show stock levels with no flow context, so interventions top up the stock instead of fixing the flow that is actually broken.
EXAMPLES
World3 model's population and capital stocks, Limits to Growth, 1972, Corporate cash balance versus monthly burn rate, A DAO treasury as stock, grants as outflow, protocol fees as inflow
ANALOGS
Game design: mana or HP pools filled and drained by regen and ability costs (a Machinations resource node); Crypto: circulating token supply as stock, emissions and burns (e.g. EIP-1559) as the flows
DATA
SYSTEMS · Strategy · RUNG 10

Theory of Constraints

A chain is as strong as its weakest link; find the one binding constraint first.

Go deeper
STRUCTURE
Any system has one, or very few, binding constraints limiting its throughput; the five focusing steps are identify the constraint, exploit it, subordinate everything else to it, elevate it, and repeat on the next constraint.
WHY IT IS LEVERAGE
It stops builders from optimizing non-bottleneck parts of a pipeline, which produces zero throughput gain, and forces identification of the one place where effort actually converts to output.
THE DRIVER
the instinct to improve everything evenly rather than find and fix the one binding limit
THE SHADOW
the constraint gets fixed once and never re-identified after it moves, so optimization effort keeps targeting a constraint that already shifted elsewhere.
EXAMPLES
Goldratt, The Goal, 1984, Drum-buffer-rope scheduling in manufacturing, A DAO's single multisig signer as the real throughput constraint on every treasury action
ANALOGS
Game design: the one under-tuned mechanic that caps a game's entire pacing is the real bottleneck to balance first; DAO: the single slowest approval step caps every proposal's cycle time regardless of how fast the other steps run
DATA
  • The Goal introduced the five focusing steps via a manufacturing-plant narrative and was named one of the 25 most influential business management books: Time magazine list, 2011 tocinstitute.org/the-goal-summary.html · 1984
SYSTEMS · Structure · RUNG 10

Gall's Law

A working complex system was always grown from a simple one that worked.

Go deeper
STRUCTURE
Working complex systems are only ever reached by incremental evolution from a simpler working system; systems designed complex from the start lack the intermediate working states needed to debug them, and no amount of patching recovers that.
WHY IT IS LEVERAGE
It is the strongest argument for shipping the smallest working version of anything, a DAO's governance or a game's economy, and evolving it, rather than designing the final complex version up front.
THE DRIVER
the designer's confidence that enough upfront analysis can substitute for the debugging that only comes from running a simpler version
THE SHADOW
used to justify never attempting necessary complexity at all, when the law only says complexity must be grown, not that it must be avoided.
EXAMPLES
John Gall, Systemantics, 1975, later The Systems Bible, 2002, The World Wide Web's incremental growth versus CORBA's complex-from-scratch failure, Agile and iterative software development's core justification
ANALOGS
Game design: shipping a one-mechanic prototype and growing the economy, instead of designing the full endgame economy pre-launch; DAO: launching with a simple multisig before evolving into token-weighted on-chain voting
DATA
  • Verbatim statement of the law: a complex system that works is invariably found to have evolved from a simple system that worked; a complex system designed from scratch never works and cannot be patched up to make it work; you have to start over, beginning with a working simple system en.wikiquote.org/wiki/John_Gall · 1975
SYSTEMS · Strategy · RUNG 10

Dormans' Machinations

A runnable notation for a game's economy: stock-flow diagrams that simulate play.

Go deeper
STRUCTURE
Machinations represents a game's internal economy as a diagram of resource pools connected by flow edges with rates, gates, and triggers, which can be run as a live simulation to reveal emergent balance and progression dynamics before any code is written.
WHY IT IS LEVERAGE
It is the direct, named bridge between system-dynamics stock-flow modeling and game economy design, the formal link anyone working both disciplines in this corpus is relying on.
THE DRIVER
designing a game economy by intuition or spreadsheet instead of simulating its stock-flow structure before building it
THE SHADOW
a Machinations diagram that balances in simulation still fails against real players who find off-diagram exploits, social dynamics or meta-strategies the formal model never captured.
EXAMPLES
Joris Dormans, Engineering Emergence: Applied Theory for Game Design, PhD thesis, Adams and Dormans, Game Mechanics: Advanced Game Design, 2012, machinations.io used in commercial game economy design
ANALOGS
Game design: this is the game-design-native form of stock-flow modeling itself; DAO: Machinations-style diagrams increasingly used to simulate token-economy stock-flows before mainnet launch
DATA
SYSTEMS · Complexity · RUNG 11

Antifragility

Some systems do not just survive shocks, they need disorder to grow stronger.

Go deeper
STRUCTURE
Taleb places systems on a spectrum from fragile (harmed by volatility) through robust or resilient (unchanged by volatility) to antifragile (improved by volatility within limits, via optionality, redundancy, and hormesis).
WHY IT IS LEVERAGE
It redirects builders from preventing all shocks, which is impossible and fragilizing, to designing so shocks make the system stronger: convex exposure, small reversible bets, option-rich positions.
THE DRIVER
the planner's illusion that more prediction and prevention of volatility is always better
THE SHADOW
antifragile gets used to glamorize recklessness, taking large undiversified risks and calling any survival antifragility after the fact rather than designing real convexity upfront.
EXAMPLES
Taleb, Antifragile: Things That Gain from Disorder, 2012, Immune systems strengthened by controlled pathogen exposure, hormesis, A DAO bug-bounty program converting attack attempts into hardening, versus a fragile single-audit security model
ANALOGS
Game design: roguelike permadeath loops make the player, not the character, stronger through repeated failure; DAO: protocols that harden their own code via public, ongoing exploit attempts rather than relying on prediction alone
DATA
  • Antifragile defined as beyond resilient, not just resisting shocks but improving because of them: the resilient resists shocks and stays the same; the antifragile gets better en.wikipedia.org/wiki/Antifragile_(book) · 2012
CHAIN · Agents · RUNG 1

AI Agents + Crypto-Native Payments

Stablecoins give software its own wallet, so agents pay each other with no human in the loop.

Go deeper
STRUCTURE
New rule layer: x402 reuses the dormant HTTP 402 status code so a server quotes a price, an agent signs a stablecoin transfer, and the resource unlocks on settlement, no accounts or invoices. ERC-8004 adds onchain identity, reputation, and validation registries for agents.
WHY IT IS LEVERAGE
Once software can hold and move value without a human approving every payment, agents stop being tools that ask a human to click pay and become economic actors that negotiate and settle in a loop measured in milliseconds, which is a new class of participant in the economy.
THE DRIVER
AI agent builders need machine-speed micropayments for APIs, data, and compute; platforms want a neutral, bank-free settlement rail.
THE SHADOW
Autonomous agents holding wallets with no human in the loop are a fast-moving attack surface, such as prompt-injection-to-drain-wallet; reputation registries can be sybil-gamed just like airdrops.
EXAMPLES
Coinbase x402, ERC-8004, Cloudflare Agents SDK
ANALOGS
Transcending-paradigm leverage point, Meadows rung 1, a new class of agent enters the system (systems); NPCs granted their own inventory and trade rights instead of scripted drops (game design)
DATA
CHAIN · Cautionary · RUNG 1

Meditations on Moloch

Not every disaster has a villain — some are just what competition optimizes for.

Go deeper
STRUCTURE
Describes 'multipolar traps': situations where every actor rationally optimizing for local incentives (an arms race, a status competition) produces a collective outcome none of them wanted and none can unilaterally exit — the archetype underneath every coordination failure this corpus studies.
WHY IT IS LEVERAGE
Supplies the paradigm-level diagnosis for why DAOs, public-goods funding, and governance design matter at all: without a structural answer to 'why would any individual defect from the group optimum,' every mechanism above is cosmetic.
THE DRIVER
the dread of watching a system grind toward an outcome everyone inside it hates and no one inside it can stop
THE SHADOW
Has no mechanism of its own — it is a diagnosis, not a cure, and is sometimes misused to wave away any specific fixable problem as 'just Moloch,' producing fatalism instead of design.
EXAMPLES
Scott Alexander / Slate Star Codex (2014), cited across effective-altruism, AI-safety, and DAO-design writing since
ANALOGS
Every guild racing to out-gear rivals until raid content requires gear no one enjoys grinding for anymore (play); Transcending the paradigm: seeing the game being played, not just playing it better (systems, Meadows rung 1)
DATA
CHAIN · Consensus · RUNG 2

Credible Neutrality

A rule the losers still believe was fair, because it never looked at who they were.

Go deeper
STRUCTURE
A design rule applied to protocol mechanics: do not optimize for particular outcomes or players, validate only against objective, pre-committed criteria. Same base fee, same mempool rules, same gas cost for anyone's bytes.
WHY IT IS LEVERAGE
Neutrality is what lets strangers with zero trust in each other, or in the protocol's authors, build on one base layer without asking permission. It is the load-bearing assumption under composability, stablecoins, and every market above it.
THE DRIVER
Builders need a foundation they can trust will not be rewritten against them after they have committed capital.
THE SHADOW
Neutral code still has non-neutral deployment: client and pool concentration, VC-weighted tokens, and core-dev social layers can re-inject favoritism the base layer formally lacks.
EXAMPLES
Ethereum L1, Bitcoin L1, EIP-1559 fee market
ANALOGS
Paradigm leverage point, Meadows rung 2 (systems); A ruleset and referee with no favorites (game design)
DATA
CHAIN · Money · RUNG 2

Stablecoins

Dollar-denominated value that settles like a token, crypto's actual killer app by volume.

Go deeper
STRUCTURE
A pegged stock of reserves, mostly short-duration Treasuries, backs a circulating flow of tokens redeemable 1:1. Arbitrage between peg price and reserve NAV is the balancing loop holding the price near one dollar.
WHY IT IS LEVERAGE
Stablecoins are the settlement rail under almost every other model here, AMMs, lending, prediction markets, agent payments. They let a volatile, permissionless chain carry a non-volatile unit of account, bridging crypto rails to the real economy.
THE DRIVER
Users want dollars that move at internet speed 24/7; issuers want float yield on reserves; governments now want a regulated onramp for dollar dominance onchain.
THE SHADOW
Depegs under reserve-quality or bank-run stress, as with Terra/UST in 2022; concentration in two issuers is itself systemic risk.
EXAMPLES
USDT (Tether), USDC (Circle)
ANALOGS
Paradigm shift in what counts as money, Meadows rung 2 (systems); A stable in-game currency pegged to real value anchoring a volatile economy (game design)
DATA
CHAIN · Money · RUNG 2

Sovereign & Corporate Bitcoin Treasuries

Nation-states and public companies now hold BTC as a reserve asset, not a trader's bet.

Go deeper
STRUCTURE
A new stock category: BTC moves from trading inventory to reserve asset. The US order locks forfeited BTC into a non-sellable stock; corporate treasuries use debt and equity issuance as the flow adding to BTC held indefinitely.
WHY IT IS LEVERAGE
Re-prices Bitcoin's buyer base: once a G7 government and public companies treat it as a reserve asset, pension and sovereign wealth funds must at least model being underweight, a paradigm change in who holds it and why.
THE DRIVER
The US wants a strategic asset funded via forfeiture at zero budget cost; treasuries want a debasement hedge that drives their equity story.
THE SHADOW
Concentrates BTC in a few leveraged balance sheets whose forced-selling risk in a drawdown could hit the market the reserve was meant to stabilize; politicizes a once-neutral asset.
EXAMPLES
US Strategic Bitcoin Reserve (EO 14233), Strategy/MicroStrategy, Metaplanet
ANALOGS
Paradigm leverage point, what counts as a reserve asset, Meadows rung 2 (systems); End-game resource hoarding by the largest guild shifting a whole server's economy (game design)
DATA
CHAIN · Voting · RUNG 2

Futarchy: Vote on Values, Bet on Beliefs

Markets pick the policy; voters only pick what success means.

Go deeper
STRUCTURE
A measurable welfare metric is set by vote; competing policies are each priced by a conditional prediction market on their expected effect on that metric, and whichever market trades highest is automatically enacted — decision rights inverted from ballot to market price.
WHY IT IS LEVERAGE
Separates values (what we want) from beliefs (what works), routing around the fact that voters forecast poorly while markets aggregate dispersed information efficiently — a genuine paradigm swap in how collective choice gets made.
THE DRIVER
distrust that a popularity contest is the same thing as being right
THE SHADOW
Needs deep, well-capitalized markets to resist manipulation; a thin market lets a wealthy actor move the 'belief' price and so dictate the 'decision' anyway.
EXAMPLES
MetaDAO (Solana), Robin Hanson's original proposal
ANALOGS
EVE Online's player-run markets setting prices no council could dictate (play); Paradigm shift: decision criterion moved from votes to prices (systems, Meadows rung 2)
DATA
CHAIN · Deliberation · RUNG 2

Plurality (Weyl & Tang)

Technology for bridging difference, not erasing it onto one scale.

Go deeper
STRUCTURE
A framework and open book arguing collaborative technology should let diverse groups coordinate across their differences (plural, overlapping identities and group structures) rather than forcing everyone onto one universal ranking or one isolated individual choice.
WHY IT IS LEVERAGE
Names the paradigm underneath most of this cluster — QV, QF, bridging-based ranking, sortition, and DAOs are all instances of 'plurality' as a design stance, making this the closest thing to a unifying theory of the chain cluster.
THE DRIVER
the sense that both pure individualism and forced consensus betray what a real community feels like
THE SHADOW
As a framework rather than a protocol it has no direct failure mode of its own; the risk is 'plurality' becoming a slogan any project claims without implementing any of its actual mechanisms.
EXAMPLES
Plurality (book, 2024), Taiwan's digital governance under Audrey Tang, RadicalxChange, Plurality Institute
ANALOGS
A guild alliance where member-guilds keep distinct identities instead of merging into one mega-guild (play); A paradigm choice about what 'coordination' is even for (systems, Meadows rung 2)
DATA
CHAIN · Cautionary · RUNG 3

Memecoins & pump.fun as Attention Markets

A one-click casino where the token's only utility is the story, and the house keeps its edge.

Go deeper
STRUCTURE
A bonding-curve launchpad turns attention into the tradeable asset: anyone mints a token in seconds, price follows a curve until a liquidity threshold graduates it to a DEX. New-launch flow vastly exceeds the stock of attention able to sustain any one token.
WHY IT IS LEVERAGE
Strips crypto speculation to its purest form, no roadmap or team promise, just reflexive attention-as-price, which makes it the clearest evidence of where this design space's empowering story, permissionless issuance, tips into its shadow, permissionless extraction.
THE DRIVER
Issuers chase fast liquidity from launch and trading fees; traders chase asymmetric lottery-style upside on attention momentum.
THE SHADOW
A large majority of traders lose money while the platform earns fees regardless of outcome, a structurally extractive casino dressed as a market, with rampant wash-trading and insider pre-mines.
EXAMPLES
pump.fun, PUMP token
ANALOGS
Goals leverage point, the system's real goal versus its stated goal, Meadows rung 3 (systems); A slot-machine loot box dressed as a trading game (game design)
DATA
CHAIN · Deliberation · RUNG 3

Collective Intelligence Project — Alignment Assemblies

Ask the public what 'aligned AI' should mean before a lab decides for them.

Go deeper
STRUCTURE
CIP convenes structured 'Global Dialogues' — recurring, large, cross-national deliberations combining surveys, Pol.is-style clustering, and expert framing — feeding directly into AI labs' constitution-writing and policy processes, a goal-setting loop routed through the public rather than a lab's internal values team.
WHY IT IS LEVERAGE
Targets the goal layer directly: instead of debating how to reach a given AI-alignment target, it asks who gets to set the target in the first place — the deepest lever in this cluster, since every mechanism below it inherits whatever goal this layer picks.
THE DRIVER
not wanting the values baked into increasingly powerful AI systems decided by a handful of labs alone
THE SHADOW
'The public' convened is still a sampled, self-selecting, lab-funded slice; a lab can treat the output as legitimizing cover while still deciding unilaterally what to implement.
EXAMPLES
Collective Intelligence Project (CIP), Anthropic's Collective Constitutional AI (with CIP), CIP Global Dialogues
ANALOGS
A server's player council petitioning developers on what 'balanced' should even mean before the next patch (play); Intervening at the goal-setting rung itself, above any individual rule (systems, Meadows rung 3)
DATA
  • CIP x Anthropic Collective Constitutional AI partnership: CIP partnered with Anthropic to crowdsource public input directly into an AI model's constitution cip.org/blog/ccai · 2024-10
  • CIP Global Dialogues scale: roughly 1,000 participants from over 70 countries convened every two months on AI's role in society, launched publicly at the AI Summit in Paris cip.org/blog/alignment-assemblies-nine-months-in · 2025-02
CHAIN · Infrastructure · RUNG 4

Smart Contracts, Composability & Money Legos

Permissionless calls turn every deployed contract into everyone else's building block.

Go deeper
STRUCTURE
Rule layer: contracts are callable by any other contract with no API key or approval process. Reinforcing loop: each new primitive, token, AMM, lending market, becomes raw material for the next, compounding the design space available.
WHY IT IS LEVERAGE
This is the generativity engine of the whole ecosystem. It collapses the cost of recombination near zero, so innovation compounds instead of resetting with each team, the way closed corporate APIs never allowed.
THE DRIVER
Builders want to ship fast by reusing audited, liquid primitives instead of rebuilding a bank from scratch.
THE SHADOW
Composability composes risk too: one exploited primitive cascades through every protocol that integrated it, and audited pieces can still interact in unaudited ways.
EXAMPLES
MakerDAO, Uniswap, Yearn, Curve
ANALOGS
Self-organization leverage point, Meadows rung 4 (systems); Modular crafting systems where player-built items recombine freely (game design)
DATA
  • Total DeFi TVL across composable protocols has fallen with the broader market in 2026: ~$96.3B, down from ~$115B in Jan 2026 defillama.com/ · 2026-10
CHAIN · Infrastructure · RUNG 4

DePIN: Decentralized Physical Infrastructure

Token rewards bootstrap real-world hardware networks before the paying customers exist.

Go deeper
STRUCTURE
Reinforcing loop: token emissions pay early hardware operators, wireless hotspots, GPU nodes, sensors, ahead of paying customers, bootstrapping physical supply. As real usage and revenue arrive they are meant to replace emissions as the reward flow.
WHY IT IS LEVERAGE
Lets a protocol crowdfund and crowd-operate capital-intensive physical infrastructure that would otherwise need one company to raise and deploy huge capex. The token coordinates strangers to deploy hardware in exactly the gaps a centralized rollout would skip.
THE DRIVER
Operators want yield on hardware they already own or will buy; networks want real-world coverage or supply without owning a single device themselves.
THE SHADOW
Token emissions can badly overpay for infrastructure nobody uses yet; hardware can be spoofed or clustered to farm rewards without providing real service.
EXAMPLES
Helium, io.net, Render Network
ANALOGS
Self-organization leverage point, Meadows rung 4 (systems); Player-run supply chains building a game's world infrastructure (game design)
DATA
CHAIN · Commons · RUNG 4

Ostrom's Commons Principles, On-Chain

Eight design rules for shared resources, now encoded as protocol logic.

Go deeper
STRUCTURE
Smart contracts formalize Ostrom's 8 principles (boundaries, proportional costs/benefits, collective-choice, monitoring, graduated sanctions, conflict resolution, autonomy, nested tiers) as enforced rules and loops.
WHY IT IS LEVERAGE
Ostrom proved self-governed commons can outperform both state and market without tragedy; blockchain can make her soft social rules hard, legible, and auditable at planetary scale.
THE DRIVER
the instinct to protect a shared resource one didn't build alone and won't let one actor strip-mine
THE SHADOW
Code formalizes rules but not the face-to-face trust-building and graduated sanctioning Ostrom's communities relied on; on-chain 'boundaries' can harden into exclusion.
EXAMPLES
Commons Stack, 1Hive/Gardens, Token Engineering Commons
ANALOGS
Guild-maintained shared loot/resource pools governed by member-written rules (play); Ostrom's 8 design principles for commons governance (systems)
DATA
  • Rozas et al. map six blockchain affordances onto Ostrom's commons-governance principles (tokenisation, formalisation/decentralisation of rules, autonomous automatisation, decentralisation of infrastructure power, transparency, codification of trust): 6 affordances mapped to 8 principles doi.org/10.1177/21582440211002526 · 2021-03
CHAIN · Cautionary · RUNG 4

ConstitutionDAO

17,000+ strangers raised $47M to buy the Constitution in under a week, and still lost.

Go deeper
STRUCTURE
A Discord/Twitter-coordinated ad hoc DAO used Juicebox to crowdfund ETH toward a single-purpose goal (winning a Sotheby's auction), with contributor shares tracked by the PEOPLE token — a self-organizing structure assembled and dissolved faster than any legal entity could incorporate.
WHY IT IS LEVERAGE
Proved, empirically and publicly, that leaderless strangers can mobilize nine figures of capital toward a shared goal faster than any institution — the clearest demonstration of blockchain's actual coordination-speed ceiling, independent of whether the bid won.
THE DRIVER
the thrill of being part of something improbable and meaningful with thousands of strangers, right now
THE SHADOW
Lost the auction anyway, outbid at $43.2M by a buyer willing to pay more; refund gas fees ate into small donors' returns, and the DAO had no real governance plan for what 'winning' would have even meant operationally.
EXAMPLES
ConstitutionDAO, Juicebox Protocol, PEOPLE token
ANALOGS
A server-wide pickup raid forming in hours for a one-time world-boss event, then disbanding (play); Self-organization emerging ad hoc around a single shared goal (systems, Meadows rung 4)
DATA
CHAIN · Identity · RUNG 5

Account Abstraction, Passkeys & Intents

Let a contract be your wallet, and what you want replace which buttons you click.

Go deeper
STRUCTURE
Rule change: EIP-7702 (Pectra) lets any externally owned account attach contract code per transaction for gas sponsorship, batched calls, and passkey auth. Intents flip the flow: users sign a desired outcome; competing solvers find the execution path.
WHY IT IS LEVERAGE
Moves the complexity of how a transaction actually executes off the user and onto a competitive solver market. This is the biggest lever on mainstream usability, letting crypto UX finally look like tapping buy instead of managing gas and chain selection.
THE DRIVER
New users bounce off seed phrases and gas; apps want one-tap flows; solvers compete for the spread between a signed intent and the cheapest execution path.
THE SHADOW
Paymasters and solvers reintroduce centralized chokepoints and new fee-extraction layers; session keys and delegated code widen the attack surface of what used to be a simple keypair.
EXAMPLES
ERC-4337, EIP-7702, ERC-7683, CoW Swap
ANALOGS
Rules leverage point governing who may act as an agent, Meadows rung 5 (systems); Quest-goal abstraction; state the objective and let pathfinding handle the route (game design)
DATA
CHAIN · Funding · RUNG 5

Quadratic Voting & Quadratic Funding

Cost rises with the square of votes; small donors get matched like whales.

Go deeper
STRUCTURE
QV: cost of n votes = n^2. QF: matching-pool share is proportional to (sum of sqrt(contributions))^2 — a rewrite of the aggregation rule itself so breadth of support outweighs depth of one wallet.
WHY IT IS LEVERAGE
Directly counters plutocracy and the free-rider problem in public-goods provision by changing the funding rule to reflect breadth of support, not wealth — leverage at the rules layer that reorders who has power.
THE DRIVER
wanting one's small voice to count for something next to a whale's
THE SHADOW
Vulnerable to Sybil attacks and collusion — splitting one large donor's capital into many fake 'small' identities — unless paired with strong proof-of-personhood.
EXAMPLES
Gitcoin Grants, Clr.fund, RadicalxChange
ANALOGS
Crowdfunded guild-hall stretch goals where many small backers outweigh one patron (play); Rules-level redesign of an aggregation function (systems, Meadows rung 5)
DATA
  • Gitcoin Grants cumulative public-goods funding distributed to date: over $60M distributed to 3,700+ projects across GG1-GG23+ gitcoin.co/blog/gitcoin-grants-23-retro · 2025-11
  • Founding quadratic-funding mechanism paper: Buterin, Hitzig & Weyl, 'A Flexible Design for Funding Public Goods,' Management Science arxiv.org/abs/1809.06421 · 2018-09
  • Gitcoin Grants cumulative distribution to open-source/public-goods projects: over $60M distributed to 3,700+ projects across GG1-GG23+ gitcoin.co/blog/gitcoin-grants-23-retro · 2025-11
CHAIN · Roles · RUNG 5

Hats Protocol — On-Chain Roles

Authority as a wearable, revocable token instead of a permanent title.

Go deeper
STRUCTURE
Each role ('Hat') is a non-transferable ERC-1155 token in a hierarchical tree; a wearer gets exactly the permissions coded into that hat, and an admin can revoke or reassign it instantly — role-based governance replaces full-token-vote governance for routine operational decisions.
WHY IT IS LEVERAGE
Separates authority to act from ownership of the token, letting a DAO delegate real operational power without handing over governance votes or requiring a full proposal for routine work — a rules-layer fix for 'DAOs can't do anything fast.'
THE DRIVER
wanting clear, revocable job titles inside a DAO instead of either total centralization or voting on every single task
THE SHADOW
Still depends on the admin hat being well-governed; a captured or compromised top-level hat can reassign every role beneath it instantly.
EXAMPLES
Hats Protocol, ENS DAO (Hats pilot), Hats protoDAO
ANALOGS
Guild officer ranks with scoped permissions (loot master, raid leader) instead of one shared admin password (play); Rules that define roles and permissions directly rather than relying on norms (systems, Meadows rung 5)
DATA
  • Hats Protocol's core design: roles represented as non-transferable ERC-1155 'Hats' with revocable delegation of authority and responsibility, organized hierarchically github.com/Hats-Protocol/hats-protocol · 2025-06
CHAIN · Identity · RUNG 5

Proof of Personhood & Sybil Resistance

Prove you're one real human, without proving which one.

Go deeper
STRUCTURE
Biometric (iris-scanning Orb) or social-graph (web-of-trust) attestations produce a unique-human credential that downstream apps verify via zero-knowledge proof — a rule change on who/what counts as one voting unit, closing the Sybil loophole underneath every one-person-one-vote mechanism.
WHY IT IS LEVERAGE
Without this layer, every 'one person, one signal' mechanism above (QF, QV, airdrops) degrades into 'one wallet, one vote' and gets Sybil-farmed into plutocracy by default — the precondition that makes the empowering half of this cluster real rather than theoretical.
THE DRIVER
the need to be counted as a person online without surrendering who you actually are
THE SHADOW
Biometric approaches trade Sybil-resistance for a coercible, surveillance-adjacent chokepoint (one Orb network deciding who is 'real'); social-graph approaches are gameable by sufficiently patient collusion rings.
EXAMPLES
World ID / Worldcoin, BrightID, Human Passport (formerly Gitcoin Passport)
ANALOGS
Account verification stopping one player from running 50 bot alts in a guild vote (play); A rule defining the unit of agency the whole system counts (systems, Meadows rung 5)
DATA
CHAIN · Deliberation · RUNG 5

Citizens' Assemblies & Sortition

Pick the deciders by lottery, brief them like jurors, let them out-deliberate parliament.

Go deeper
STRUCTURE
A demographically-stratified random sample of citizens is convened, given balanced expert testimony over weeks, deliberates in structured small and plenary sessions, and issues recommendations — replacing election (selects for ambition and money) with sortition (selects for a representative cross-section).
WHY IT IS LEVERAGE
Removes campaign-donor and re-election incentives from the room entirely; a juror who will never run for office again has no structural reason to posture, which is why assemblies have repeatedly broken deadlocks elected politicians couldn't touch.
THE DRIVER
distrust that people who need to get re-elected can ever give a costly issue a straight answer
THE SHADOW
No enforcement power beyond the government's choice to hold the referendum it recommends; selection and facilitation design can still be captured to pre-shape the 'random' sample's framing.
EXAMPLES
Irish Citizens' Assembly, France's Convention Citoyenne pour le Climat, Ireland's four consecutive assemblies
ANALOGS
EVE Online's CSM seats filled by election rather than lottery, as the road not taken (play); A rule for who gets decision-input, redesigned around representativeness rather than ambition (systems, Meadows rung 5)
DATA
CHAIN · Infrastructure · RUNG 6

MEV & Proposer-Builder Separation

Split ordering from proposing so MEV becomes an open auction, not a backroom deal.

Go deeper
STRUCTURE
Information-flow redesign: builders compete to pack the most valuable block; relays show the proposer only a header, value not contents, until the proposer has already signed, which prevents the proposer from stealing or discriminating.
WHY IT IS LEVERAGE
MEV is unavoidable wherever transaction ordering has value. PBS does not eliminate it, it legalizes and opens it into a visible marketplace, converting invisible backroom extraction into an auctioned, partly redistributed revenue stream.
THE DRIVER
Searchers and builders chase arbitrage and liquidation profit; validators want MEV revenue without building blocks themselves.
THE SHADOW
Relays are centralized trust points in a system sold as trustless; sandwich attacks still tax ordinary users; enshrining PBS on-chain remains unresolved.
EXAMPLES
Flashbots, MEV-Boost, SUAVE
ANALOGS
Information-flow leverage point, Meadows rung 6 (systems); Sealed-bid auction mechanics replacing hidden pay-to-win (game design)
DATA
CHAIN · Money · RUNG 6

Spot BTC/ETH ETFs

Bitcoin and Ether get a brokerage on-ramp; institutional capital gets a regulated door.

Go deeper
STRUCTURE
A regulated wrapper converts an exchange-listed share into a claim on custodied BTC or ETH. Creation and redemption arbitrage between share price and NAV is the balancing loop keeping the ETF tracking spot, while flows become a visible proxy for institutional demand.
WHY IT IS LEVERAGE
Collapses the operational, custody, and compliance barrier that kept pensions, RIAs, and retirement accounts out. It is the biggest access-flow change in crypto's history, turning self-custody into clicking buy in an existing brokerage account.
THE DRIVER
Institutions want regulated, custodied exposure without operational risk; issuers like BlackRock and Grayscale want fee income on a new asset class.
THE SHADOW
Re-concentrates custody into a handful of TradFi issuers, the opposite of the self-sovereign holding crypto was built for; ETF flows now swing spot price with no on-chain news at all.
EXAMPLES
BlackRock IBIT, Grayscale ETHE, Fidelity FBTC
ANALOGS
Information and access-flow leverage point, Meadows rung 6 (systems); A casual-mode on-ramp that imports hardcore-mode assets (game design)
DATA
CHAIN · Markets · RUNG 6

Prediction Markets as Information Engines

Betting with skin in the game out-forecasts polls, pundits, and committees, in real time.

Go deeper
STRUCTURE
Information-flow loop: traders stake capital on outcomes; the resulting price is a continuously updated, incentive-weighted probability estimate that self-corrects as new information arrives, because wrong beliefs are directly taxed through losses.
WHY IT IS LEVERAGE
This takes the information-flows leverage point, who has access to what and when, and turns it into a live, tradable market price that often outpaces slower institutional forecasting, because every participant is paid to find the error first.
THE DRIVER
Traders seek profit from being right before the crowd; platforms earn fees on volume; the public gets a continuously updating forecast as a byproduct.
THE SHADOW
Thin markets are easy to manipulate for a cheap narrative win; offshore vs CFTC-regulated arbitrage creates uneven protection, and it can shade into sports betting.
EXAMPLES
Polymarket, Kalshi
ANALOGS
Information-flow leverage point, Meadows rung 6 (systems); Meta-game forecasting ladders and community wagering on match outcomes (game design)
DATA
CHAIN · Voting · RUNG 6

Liquid Democracy & Delegation

Don't vote yourself; lend your vote to someone who will, until you take it back.

Go deeper
STRUCTURE
Token-weighted voting rights flow through a delegate graph instead of terminating at the holder; holders may re-delegate or reclaim at any time — an information-flow redesign meant to convert low-turnout apathy into informed proxy participation.
WHY IT IS LEVERAGE
Solves the 'nobody has time to read every proposal' problem by letting expertise concentrate in willing specialists, while nominally keeping exit/reclaim power distributed.
THE DRIVER
wanting a voice in the room without doing the homework for every single decision
THE SHADOW
Concentrates real power in a small, often unaccountable delegate class; 'ghost delegates' hold power and never vote, and visibility bias lets early, well-known delegates entrench.
EXAMPLES
ENS DAO, Uniswap, Arbitrum DAO, Compound
ANALOGS
Guild officers voting on behalf of a raid roster that trusts their judgment (play); Information-flow redesign of who gets to act on what signal (systems, Meadows rung 6)
DATA
  • Arbitrum DAO voting-power concentration (Gini coefficient): mean Gini coefficient of approximately 0.995 among active wallets across 15 of the DAO's highest-turnout Snapshot proposals (~850,000 vote records); on some proposals the top 10 wallets/delegates controlled up to 86.87% of voting weight papers.ssrn.com/sol3/papers.cfm?abstract_id=7103878 · 2025-01
  • Cross-protocol delegation-concentration study across Uniswap, Aave, Arbitrum, ENS, Compound (history through June 19, 2025): delegation mechanisms reinforce visibility bias; a small set of top-ranked delegates hold disproportionate influence regardless of alignment with token holders arxiv.org/abs/2510.05830 · 2025-10
  • Arbitrum DAO voting-power concentration: mean Gini coefficient of approximately 0.995 among active wallets; on some proposals the top 10 wallets/delegates controlled up to 86.87% of total voting weight researchgate.net/publication/410908562_An_Empirical_Evaluatio… · 2025-01
  • ENS DAO delegate structure: 142 active delegates hold 64% of voting power; the top 10 delegates control 28%; Nakamoto coefficient of 18 (vs. 11 for Uniswap, 8 for Compound) discuss.ens.domains/t/ens-delegate-accountability-what-the-da… · 2026-04
  • DAO-wide voter turnout: DeepDAO-tracked data: roughly 3.7% of governance-token holders actually vote overall; major DAOs range 15-22% on contested proposals (Uniswap 22%, Optimism 19%, Arbitrum 18%, ENS 15%) daotimes.com/deepdao-dao-tool-report-for-2025 · 2025-01
CHAIN · Voting · RUNG 6

Holographic Consensus (DAOstack)

A prediction market pre-filters which proposals deserve the whole DAO's attention.

Go deeper
STRUCTURE
GEN-token 'predictors' stake on whether a proposal will pass; proposals attracting enough confident staking get 'boosted' into fast-track absolute-majority voting, letting most decisions clear at low quorum while reserving full attention for contested ones — an information-flow throttle on collective attention itself.
WHY IT IS LEVERAGE
Solves the plutocracy-vs-apathy trap by decoupling 'can this be trusted to a small majority' from 'does this need everyone' — attention becomes the resource being allocated, not just votes.
THE DRIVER
wanting a DAO to move fast on everyday decisions without demanding everyone's attention on everything
THE SHADOW
Depends on a healthy, liquid prediction market for GEN; adoption stayed low and the token itself has gone largely inactive, so the filter nobody uses filters nothing.
EXAMPLES
DAOstack, Genesis DAO
ANALOGS
A guild's officer-only fast-track vote for routine calls, with full-council votes reserved for charter changes (play); Information flow that gates attention itself as the scarce resource (systems, Meadows rung 6)
DATA
CHAIN · Identity · RUNG 6

Attestations (EAS) & Soulbound Tokens

Make trust legible: credentials and claims as a public, queryable graph.

Go deeper
STRUCTURE
EAS lets any address issue a signed, schema-typed on- or off-chain attestation about another address; SBTs (non-transferable tokens) anchor those claims to a persistent 'Soul' — together turning reputation from private memory into a shared information flow other contracts and DAOs can read and act on.
WHY IT IS LEVERAGE
Reputation, credentials, and affiliation are the raw material every other mechanism here (sybil-resistance, delegation, roles, funding eligibility) needs to read; opening that layer is a foundational information-flow fix beneath all the others.
THE DRIVER
wanting contributions and trust earned in one community to travel with you instead of evaporating when you leave
THE SHADOW
Non-transferability can calcify a reputation that should be revisable; composability means one bad or coerced attestation propagates everywhere it's read.
EXAMPLES
Ethereum Attestation Service (EAS), Optimism's badgeholder attestation system, Gitcoin/Human Passport stamps
ANALOGS
A character's achievement/reputation panel that follows them across guilds (play); A new information channel feeding every balancing and reinforcing loop downstream (systems, Meadows rung 6)
DATA
  • The founding soulbound-token paper: Weyl, Ohlhaver & Buterin's 'Decentralized Society: Finding Web3's Soul' proposes non-transferable SBTs encoding commitments, credentials, and affiliations for sybil-resistant governance papers.ssrn.com/sol3/papers.cfm?abstract_id=4105763 · 2022-05
  • EAS network scale: over 9 million attestations by more than 450,000 attesters across networks, including 1.3M+ attestations on Optimism alone x.com/eas_eth · 2026-10
CHAIN · Deliberation · RUNG 6

Community Notes / Bridging-Based Ranking

A note only shows up if people who usually disagree both rate it helpful.

Go deeper
STRUCTURE
Notes on a post are scored by a matrix-factorization model that looks for raters across the ideological spectrum converging on 'helpful' — the algorithm explicitly rewards bridging rather than majority agreement, flipping the usual engagement-maximizing feedback loop.
WHY IT IS LEVERAGE
Most platform ranking reinforces whatever a user's own cluster already believes, a reinforcing loop toward polarization; bridging-based ranking is a rare case of an information-flow rule that actively selects for cross-tribal agreement instead.
THE DRIVER
the exhaustion of watching every fact-check look like it came from 'the other side'
THE SHADOW
Still gameable by sufficiently broad coordinated brigades; can be hollowed out by platforms that drop enforcement teeth (label-only, no removal) or by skewed contributor pools.
EXAMPLES
X/Twitter Community Notes (ex-Birdwatch), Meta Community Notes, TikTok Community Notes
ANALOGS
A raid wipe post-mortem only becoming 'the lesson' once both the tank and healer factions agree on it (play); Selecting the information flow for cross-cutting signal, not volume (systems, Meadows rung 6)
DATA
  • Meta's 2025 move to a Community Notes model: Meta discontinued its US third-party fact-checking program and began testing a Community Notes model across Facebook, Instagram, and Threads platformer.news/meta-community-notes-launch/ · 2025-01
  • TikTok's Community Notes contributor base: nearly 80,000 users approved as contributors since the April 2025 announcement; TikTok's notes supplement rather than replace professional fact-checking, unlike Meta's and X's approach tech.yahoo.com/social-media/articles/tiktok-adding-community-… · 2025-04
CHAIN · Deliberation · RUNG 6

Pol.is & vTaiwan

Cluster the crowd's opinions into a visible map before voting on anything.

Go deeper
STRUCTURE
Pol.is collects short statements and agree/disagree votes, then uses unsupervised clustering to visualize opinion groups in real time; vTaiwan runs that map through a structured 4-stage process (crowdsource, reflect, legislate, feedback) before any legislation is drafted.
WHY IT IS LEVERAGE
Most online deliberation collapses into whatever take gets the most upvotes; surfacing the actual shape of disagreement finds non-obvious consensus a show-of-hands vote would hide — a change in what decision-makers even get to see.
THE DRIVER
wanting to be heard as a nuanced position, not flattened into 'agree' or 'disagree' with the loudest take
THE SHADOW
Participation skews toward the already-online and already-engaged; the process produces legitimacy and clarity but has no binding force unless a government chooses to act on it.
EXAMPLES
vTaiwan (Uber regulation, 2015), Pol.is (used globally, including by AI labs for public-input processes), Taiwan's Join platform
ANALOGS
A guild forum's sentiment map showing where raid-night-time factions actually overlap before a vote (play); Information-flow redesign: show the structure of disagreement, not just its tally (systems, Meadows rung 6)
DATA
CHAIN · Voting · RUNG 6

AI Delegates & AI-Assisted Governance

Train an AI on your values, then let it vote the thousand proposals you'll never read.

Go deeper
STRUCTURE
An AI agent is trained on a token-holder's past votes and stated preferences (a 'digital twin') or on a DAO's documented values, then votes, drafts rationale, or flags proposals on the holder's behalf — a new information-processing layer inserted between proposal and vote, typically bounded by hard-coded 'circuit breakers.'
WHY IT IS LEVERAGE
Targets the real bottleneck in every delegation mechanism above — human attention — by automating the reading and reasoning, not the values; done honestly it could deliver the 'informed participation' liquid democracy always promised but rarely achieved.
THE DRIVER
wanting to actually participate in the governance you're entitled to, without a second full-time job reading every proposal
THE SHADOW
Accountability collapses the moment an AI delegate votes wrong, with no settled answer for who is responsible; also concentrates real influence in whoever designs the most-adopted agent, a subtler version of the same delegate-concentration problem.
EXAMPLES
Near Foundation 'digital twins', Optimism Foundation Mission Request AI governance-agent pilot, StableLab AI-assisted delegate research
ANALOGS
An add-on that auto-votes 'ready check' based on your historical attendance pattern (play); A new information-processing layer inserted into an existing feedback loop (systems, Meadows rung 6)
DATA
CHAIN · Markets · RUNG 7

Points Programs & Airdrops

Promise a future token for present behavior, and the fake behavior scales right along with it.

Go deeper
STRUCTURE
Reinforcing growth loop: protocols emit non-binding points for usage; expectation of a future token pulls in capital and activity now, which the protocol showcases to pull in more. Sybil detection becomes a balancing loop fighting the exploit this same reinforcement invites.
WHY IT IS LEVERAGE
Decouples token distribution from a sale, letting protocols bootstrap real usage and a distributed holder base before a market exists. When it works, users market the protocol for a lottery ticket instead of a wage, which is remarkably capital-efficient growth.
THE DRIVER
Protocols want fast liquidity and usage bootstrapping without an SEC-risky token sale; users want a free, asymmetric bet on future airdrop value.
THE SHADOW
Sybil farms, thousands of wallets run by one actor, capture a disproportionate share meant for real users; mercenary capital exits the instant the token lists, leaving the usage number hollow.
EXAMPLES
LayerZero, Blast, EigenLayer airdrop, Hyperliquid airdrop
ANALOGS
Reinforcing feedback loop, Meadows rung 7 (systems); Battle-pass grinding that rewards anticipated loot, later patched against bot farms (game design)
DATA
CHAIN · Cautionary · RUNG 7

Play-to-Earn Collapse: Axie Infinity & Ronin

When the reward loop outruns the fun loop, both the economy and the bridge under it get drained.

Go deeper
STRUCTURE
A reinforcing loop, new players fund payouts to earlier players, attracting more players, works only while inflow exceeds outflow. SLP emissions ran about 4x redemption and burn, so the token had nowhere to go but down once new-player inflow slowed.
WHY IT IS LEVERAGE
The cleanest public case study of why fun and yield are different reward systems that cannot be collapsed into one token without one cannibalizing the other. The $625M Ronin bridge hack then showed the custodial layer underneath was just as fragile as the token design.
THE DRIVER
Players, many in the Philippines and Southeast Asia, needed real income during COVID; the studio needed explosive user growth to justify its valuation.
THE SHADOW
SLP lost over 99% of its value as emissions outpaced sinks; Ronin's 5-of-9 validator multisig was compromised by the Lazarus Group for roughly $625M, the largest DeFi hack to that date.
EXAMPLES
Axie Infinity, Smooth Love Potion (SLP), Ronin Network
ANALOGS
Unchecked reinforcing loop, Meadows rung 7 (systems); Inflationary gold-farming economy collapse in early MMOs (game design)
DATA
CHAIN · Markets · RUNG 7

Hyperliquid: Fully Onchain Order-Book Perps

A CEX-grade order book where the matching engine is the blockchain's own consensus.

Go deeper
STRUCTURE
HyperBFT consensus runs the central-limit order book as the chain's state machine itself, not a settlement layer under a separate matching engine. Fee revenue flows into an Assistance Fund that buys and burns HYPE, a reinforcing loop tying usage directly to token scarcity.
WHY IT IS LEVERAGE
Proves an onchain venue can match a centralized exchange on latency and depth while keeping custody, matching, and settlement transparent and verifiable, collapsing the assumption that DeFi must always be slower or thinner than a CEX for perps specifically.
THE DRIVER
Traders want CEX-speed execution without CEX custody risk; the protocol wants fee revenue to accrue visibly to token holders instead of a company's cap table.
THE SHADOW
Hyperliquid is a single chain and team, a soft-centralization point despite onchain mechanics; large liquidations have stressed the vault before.
EXAMPLES
Hyperliquid, HyperCore, HYPE token
ANALOGS
Reinforcing loop tying usage to scarcity, Meadows rung 7 (systems); A fully server-authoritative competitive match where the server is the public replay (game design)
DATA
CHAIN · Funding · RUNG 7

Retroactive Public Goods Funding (RetroPGF)

Fund what already proved valuable, not what promises to be.

Go deeper
STRUCTURE
Flips the funding arrow: badgeholders score completed impact after the fact, then token emissions pay backward — converting a speculative grant process into a reinforcing loop that feeds proven usage with more resources.
WHY IT IS LEVERAGE
Removes the grant committee's forecasting problem entirely; impact is measured, not pitched, which starves grift and compounds resources toward what is already demonstrably working.
THE DRIVER
the builder's frustration at being asked to pitch a future instead of being paid for a past that already shipped
THE SHADOW
Badgeholder subjectivity and lobbying still shape scores; early rounds skewed toward well-connected, legible projects over quiet infrastructure work.
EXAMPLES
Optimism Collective RetroPGF 1-7, Retro Funding (2025 OSO evaluation model)
ANALOGS
Raid loot distributed after the kill based on logged contribution, not pre-raid promises (play); Reinforcing feedback loop re-aimed at verified output (systems, Meadows rung 7)
DATA
  • Cumulative RetroPGF / Retro Funding OP distributed since program launch: over 60M OP tokens distributed to hundreds of projects and contributors since 2021 optimism.io/blog/retro-funding-2025 · 2025-01
  • Cumulative RetroPGF distribution since launch: over 60M OP tokens distributed to hundreds of projects and contributors optimism.io/blog/retro-funding-2025 · 2025-01
  • RetroPGF Round 2 badgeholder allocation: 69 of 71 badgeholders allocated 10M OP across 195 nominated projects/people, all of whom received funding optimism.io/blog/announcing-the-results-of-retropgf-2 · 2023-12
  • 2025 shift from annual rounds to continuous impact evaluation: RetroPGF 7 began Jan 16, 2025, transitioning into ongoing 'Retro Funding' across Dev Tooling, Onchain Builders, and OP Stack Contributions optimism.io/blog/retro-funding-2025 · 2025-01
CHAIN · Consensus · RUNG 8

Nakamoto Consensus & Proof-of-Work

Longest chain wins; self-adjusting difficulty turns electricity into unforgeable history.

Go deeper
STRUCTURE
Stock = accumulated chain-work. Flow = one block ~10min. Balancing loop: difficulty retargets every 2,016 blocks to hold the interval steady as hash-power enters or leaves. Halving is a hard-coded rule cutting issuance every 210,000 blocks.
WHY IT IS LEVERAGE
Removes any human arbiter from what happened and in what order, the deepest primitive other models borrow trust from. No permission is needed to add hash-power or to verify; truth is majority-compute, not majority-authority.
THE DRIVER
Miners chase block reward plus fees; users want censorship-resistant settlement with no counterparty.
THE SHADOW
51% attacks hit smaller chains; energy cost centralizes mining into few pools/regions; MEV reintroduces the extraction PoW tried to remove.
EXAMPLES
Bitcoin, Litecoin, Bitcoin Cash
ANALOGS
Balancing loop via self-adjusting parameter (systems); Dynamic difficulty adjustment keyed to player population (game design)
DATA
CHAIN · Markets · RUNG 8

Overcollateralized Lending & Liquidations

Trustless credit: demand more collateral than borrowed, force-sell if the margin slips.

Go deeper
STRUCTURE
Stock = collateral deposited versus debt owed. Balancing loop: a health factor of collateral value over debt triggers automatic keeper-bot liquidation once it crosses a threshold, no court or credit check, just code watching a ratio live.
WHY IT IS LEVERAGE
Replaces underwriting, which needs identity, history, and legal recourse, with a math inequality enforced by economic incentive. This is how strangers lend to strangers at internet speed with no KYC and mostly no bad debt.
THE DRIVER
Borrowers want leverage or liquidity without selling their position; liquidators want a margin for enforcing solvency.
THE SHADOW
Oracle manipulation or thin-market gaps trigger cascading liquidations that overshoot; bad debt can still outrun keepers in extreme volatility.
EXAMPLES
Aave, MakerDAO/Sky, Compound
ANALOGS
Balancing feedback loop, Meadows rung 8 (systems); Margin-call or permadeath-on-threshold mechanics (game design)
DATA
CHAIN · Money · RUNG 8

EIP-1559 Base-Fee Burn

A self-adjusting fee burns instead of paying validators, making usage itself deflationary.

Go deeper
STRUCTURE
Balancing loop: the base fee rises or falls to target half-full blocks; the base-fee portion is burned, removed from the supply flow, while only the tip goes to the proposer. Net ETH supply becomes issuance minus burn.
WHY IT IS LEVERAGE
Decouples paying for blockspace from paying a rent-seeking validator. The burn makes usage itself a value-accrual mechanism for every ETH holder, aligning network activity with the asset's scarcity instead of only the fee-recipient's income.
THE DRIVER
Users want predictable fees instead of first-price-auction volatility; holders want usage growth to benefit supply, not just validators.
THE SHADOW
Low-activity periods flip ETH back to net-inflationary; L2 rollups now capture most execution fees, muting L1 burn versus the 2021 design assumptions.
EXAMPLES
Ethereum (London upgrade, Aug 2021)
ANALOGS
Balancing loop with a reinforcing side-effect (systems); A currency sink that burns spent gold to fight inflation (game design)
DATA
  • EIP-1559 specification: a per-block network fee that is burned, with block size flexing to absorb congestion: base fee burned; created 2019-04-13, live August 2021 (London) eips.ethereum.org/EIPS/eip-1559 · 2019-04
CHAIN · Voting · RUNG 8

Dual Governance (Optimistic Veto)

Token holders propose; the people with skin in the staked asset can brake it.

Go deeper
STRUCTURE
A second, independent stock of locked stETH can trigger a dynamic timelock: 1% opposed locks in an escalating 5-45 day delay, 10% opposed triggers a full 'rage quit' halt until opposers exit — a balancing loop layered on top of simple majority token voting.
WHY IT IS LEVERAGE
Converts governance from one-token-one-vote into a two-chamber check where the people most exposed to protocol risk (stakers) can brake decisions made by people who merely hold the governance token.
THE DRIVER
the staker's fear of being overruled by a governance token they don't hold, about an asset they do
THE SHADOW
Still requires stETH holders to organize and notice a bad proposal in time; apathy defeats the brake exactly as it defeats ordinary voting.
EXAMPLES
Lido DAO
ANALOGS
A raid leader's call overridden by a tank/healer veto when a wipe is imminent (play); A balancing loop added to check a reinforcing one (systems, Meadows rung 8)
DATA
CHAIN · Infrastructure · RUNG 9

ZK Proofs & Real-Time zkVMs

Prove a computation happened correctly without re-running it; verification turns cheap.

Go deeper
STRUCTURE
A zkVM compresses verifying an execution from re-running everything to checking one short proof, collapsing the delay between computation and trusted verification from minutes or blocks to seconds, a direct attack on the system's slowest feedback loop.
WHY IT IS LEVERAGE
Proving instead of re-executing lets a laptop verify what a datacenter computed. It underlies validity-rollup scaling, privacy-preserving compliance, and trust-minimized bridges, because it fully separates trusting a result from revealing its inputs.
THE DRIVER
Rollups need fast, cheap finality; enterprises and users want privacy-preserving compliance; chains want light clients that skip full nodes.
THE SHADOW
Proving systems are young and complex; bugs in circuits are near-invisible until exploited; heavy GPU proving reintroduces centralization at the prover layer.
EXAMPLES
Succinct SP1, RISC Zero, zkSync, Scroll
ANALOGS
Delay-reduction leverage point, Meadows rung 9 (systems); Deterministic lockstep replay verification instead of trusting the host, as in anti-cheat (game design)
DATA
CHAIN · Cautionary · RUNG 9

Flash-Loan Governance Attacks

Borrow the vote, pass the theft, repay the loan — all in one transaction.

Go deeper
STRUCTURE
An attacker takes an uncollateralized flash loan, uses the borrowed capital to acquire a supermajority of a protocol's governance token within a single block, passes a self-serving proposal with zero time delay between acquiring power and executing it, then repays the loan — exploiting the absence of any delay between voting-power acquisition and execution.
WHY IT IS LEVERAGE
The cautionary anchor for this whole cluster: these attacks prove pure token-weighted voting with no time-lock isn't 'decentralized' at all — it's temporarily rentable, and rentable governance is attackable governance.
THE DRIVER
the attacker's recognition that a system built to trust 'whoever holds the tokens right now' has no way to ask whether that holder will still be around tomorrow
THE SHADOW
is itself the cautionary case; the fix (time-locks, delegation lag, snapshot-before-proposal) is now close to universal, but newer DAOs still occasionally ship without it.
EXAMPLES
Beanstalk Farms (2022), Mango Markets (2022)
ANALOGS
A raid-boss exploit patched the same week it's discovered, versus one that lingers because 'that's just how the system works' (play); A missing delay in a stock-flow structure, exploited the instant it's noticed (systems, Meadows rung 9)
DATA
  • Beanstalk flash-loan governance exploit, total drained: attacker used a roughly $1B flash loan from Aave to pass a malicious supermajority governance proposal in one transaction, draining approximately $182M (netting roughly $80M after costs), April 17, 2022 bloomberg.com/news/articles/2022-04-18/defi-project-beanstalk… · 2022-04
  • Beanstalk flash-loan governance exploit: attacker used roughly a $1B flash loan from Aave to pass a malicious supermajority proposal in one transaction, draining approximately $182M (netting roughly $80M after costs), April 17, 2022 bloomberg.com/news/articles/2022-04-18/defi-project-beanstalk… · 2022-04
  • Mango Markets exploit and its governance-vote aftermath: attacker drained roughly $114-117M by manipulating the MNGO price about 1,000% using $5M in capital, then used Mango DAO's own governance vote to let him keep $47M as a 'bounty' in exchange for returning the rest, with about 96% of voted tokens in favor coindesk.com/business/2022/10/20/mango-markets-will-start-ret… · 2022-10
CHAIN · Markets · RUNG 10

Automated Market Makers (x times y = k) & Bonding Curves

A formula replaces the order book; price is just the ratio of two pooled stocks.

Go deeper
STRUCTURE
Two token stocks sit in a shared pool; the invariant x*y=k is the stock-flow structure itself, every swap is a flow that must hold the product constant, so price is purely a function of relative depletion. Bonding curves generalize this to one-sided issuance.
WHY IT IS LEVERAGE
Removes the market-maker as a privileged intermediary. Anyone can supply liquidity and anyone can price against the curve 24/7 with no order book, no matching engine, and no listing committee.
THE DRIVER
LPs seek fee yield on idle assets; traders want always-on liquidity without needing a counterparty to show up.
THE SHADOW
Impermanent loss transfers value from LPs to arbitrageurs; thin pools are trivially manipulated or sandwiched; bonding-curve tokens are a classic rug-pull rail.
EXAMPLES
Uniswap v2/v3/v4, Curve, Bancor, pump.fun
ANALOGS
Stock-flow structure, Meadows rung 10 (systems); Formulaic crafting sinks and faucets setting an in-game exchange rate (game design)
DATA
CHAIN · Infrastructure · RUNG 10

Restaking & Shared Security

The same staked capital re-rented to secure a dozen other networks at once.

Go deeper
STRUCTURE
Stock = staked ETH and liquid staking tokens. Restakers delegate that same capital's slashing risk to operators, who in turn secure multiple Actively Validated Services, one stock feeding many parallel security flows, each with its own slashing condition.
WHY IT IS LEVERAGE
Turns bootstrapping your own validator set, previously the biggest cost new chains faced, into a rental market: new infrastructure buys credible security off Ethereum's existing stake instead of begging for its own from scratch.
THE DRIVER
New AVSs need cheap, credible security fast; stakers want extra yield on capital already locked and otherwise idle beyond base issuance.
THE SHADOW
Correlated slashing risk: one bad AVS or operator bug can slash capital meant to secure something else entirely; re-hypothecation echoes 2008-style leverage stacking.
EXAMPLES
EigenLayer, Symbiotic, Babylon
ANALOGS
Stock-flow structure reused across multiple flows, Meadows rung 10 (systems); A shared resource pool rented out to multiple guilds at once (game design)
DATA
CHAIN · Infrastructure · RUNG 10

Rollups & Modular Data Availability

Execute off-chain, settle on-chain: scale by unbundling compute from consensus.

Go deeper
STRUCTURE
Layered stock-flow: L2s batch transactions and post compressed data or proofs to L1 blobs, a cheaper pruned stock than full calldata. EIP-4844 added blob space; Fusaka's PeerDAS lets validators sample slices instead of downloading all data, raising the ceiling again.
WHY IT IS LEVERAGE
Separates the three jobs a chain does, execute, order and settle, store data, into markets that scale independently, so blockspace stops being one monolithic bottleneck and becomes a stack where every layer can be optimized or swapped.
THE DRIVER
Users want cheap, fast transactions; L1 wants to stay the neutral settlement and data-availability root, not compete on throughput.
THE SHADOW
Fragmented liquidity and security assumptions across dozens of L2s recreate the bridge-risk and UX-fragmentation problems rollups were meant to solve; sequencer centralization is common in practice.
EXAMPLES
Arbitrum, Base, Optimism, zkSync
ANALOGS
Unbundling a monolithic stock-flow structure, Meadows rung 10 (systems); Client-side prediction with server reconciliation; do work locally, settle authoritative state centrally (game design)
DATA
CHAIN · Money · RUNG 10

Tokenized Real-World Assets

Treasuries and credit move onto programmable rails, carrying TradFi yield into DeFi.

Go deeper
STRUCTURE
An off-chain stock, US T-bills or private credit, is wrapped into an onchain token whose NAV flow updates from the custodian or issuer. Once tokenized, the asset becomes composable collateral inside the same lending and AMM primitives as any native crypto asset.
WHY IT IS LEVERAGE
Bridges the two largest pools of capital on earth, short-duration sovereign debt and DeFi's idle stablecoin liquidity, letting onchain protocols earn real yield and giving TradFi instruments 24/7 settlement and composability for the first time.
THE DRIVER
Issuers like BlackRock want distribution into crypto-native capital; DeFi protocols want yield-bearing collateral instead of idle cash.
THE SHADOW
Legal recourse still runs through the off-chain issuer or custodian; a token is a claim, not the asset, and redemption gates or issuer trouble break the as-good-as-cash assumption fast.
EXAMPLES
BlackRock BUIDL, Franklin Templeton BENJI, Ondo Finance
ANALOGS
Stock-flow structure linking two separate systems, Meadows rung 10 (systems); Cross-game currency exchange bridges between virtual and real economies (game design)
DATA
CHAIN · Voting · RUNG 10

Conviction Voting

Votes accrue like compound interest the longer you hold them on a proposal.

Go deeper
STRUCTURE
Each voter's signal is a stock that accumulates and decays continuously rather than a one-shot tally; a proposal passes once accumulated conviction crosses a threshold scaled to the size of its ask — continuous funding, no snapshot deadline.
WHY IT IS LEVERAGE
Replaces single-moment plutocratic snapshots with a time-weighted signal rewarding sustained conviction over flash-mobbed whale votes, letting funding flow continuously instead of in discrete, game-able rounds.
THE DRIVER
wanting patience and sustained belief to count for more than one large, momentary stake
THE SHADOW
Low-liquidity treasuries can still be drained slowly by patient capital; decay-rate and threshold parameters are easy to mis-tune and opaque to newcomers.
EXAMPLES
1Hive / Gardens (HNY), Commons Stack, Token Engineering Commons
ANALOGS
A guild's DKP priority building the longer a member waits their turn for loot (play); Stock-and-flow structure: conviction as an accumulating stock gated by inflow and decay (systems, Meadows rung 10)
DATA
  • Conviction voting's origin and first production deployment: derived from Michael Zargham/BlockScience research; first deployed by Commons Stack and 1Hive as an Aragon app on Gnosis Chain, with a default 48-hour conviction half-life github.com/1Hive/conviction-voting-app · 2024-01
CHAIN · Commons · RUNG 10

Protocol Guild

Pay the maintainers of the commons everyone depends on and no one owns.

Go deeper
STRUCTURE
A continuously-funded pool, fed by ecosystem donations and routed via 0xSplits, vests to a weighted list of Ethereum core-protocol contributors over 4 years, with membership and weights periodically re-evaluated — a stock-flow fix for public infrastructure's classic underfunding.
WHY IT IS LEVERAGE
Core-protocol maintenance is the textbook invisible public good: critical, uncompensated by any single market transaction, and chronically underfunded; routing ecosystem-wide value capture back to maintainers closes that loop structurally rather than through one-off grants.
THE DRIVER
guilt and self-interest both, from everyone who profits off software whose maintainers they've never had to think about paying
THE SHADOW
Funding still depends on voluntary ecosystem generosity, not a mandatory protocol-level tax; membership and weighting decisions sit with a small steering group, reintroducing a political center.
EXAMPLES
Protocol Guild, 0xSplits
ANALOGS
A game studio's core engine team funded by a slice of every third-party mod's revenue (play); A stock (the funding pool) with a steady inflow and vesting outflow replacing one-off grants (systems, Meadows rung 10)
DATA
CHAIN · Consensus · RUNG 11

Proof-of-Stake Staking & Slashing

Security bought with capital at risk instead of energy at cost.

Go deeper
STRUCTURE
Stock = ETH locked as validator bonds, a buffer. Flow = issuance rewards in, slashing and exit queue out. Balancing loop: misbehavior burns a slice of the bond, pricing dishonesty directly instead of out-competing it with more hashrate.
WHY IT IS LEVERAGE
Converts security from a physical resource, power plants and ASICs, into a financial one, bonded capital, which is programmable: stake can be restaked, delegated, liquid-wrapped, and sliced into new security markets.
THE DRIVER
Validators want yield on idle capital; the network wants costly-to-attack consensus without PoW's energy bill.
THE SHADOW
Liquid staking centralizes into a few operators; correlated client bugs can cascade slashing; compounding issuance favors the already-large.
EXAMPLES
Ethereum (The Merge), Lido, Rocket Pool
ANALOGS
Buffer / safety-stock leverage point (systems); Gear-at-risk raid mechanics, lose your bond on death (game design)
DATA
CHAIN · Exit · RUNG 11

Moloch DAO Rage-Quit

Disagree hard enough and you can walk out the door with your share of the treasury.

Go deeper
STRUCTURE
Any member can 'ragequit' at any time before a proposal executes, burning their shares to withdraw their proportional stake from the treasury before the contested decision takes effect — exit implemented as a hard buffer against capture.
WHY IT IS LEVERAGE
Operationalizes Hirschman's 'exit' as an individually-exercisable right rather than a mere sentiment, so a minority cannot be trapped funding a direction it rejects — the feature that made pooling DAO treasuries safe in the first place.
THE DRIVER
the need for an emergency door when voice has failed and the group is about to fund something you can't live with
THE SHADOW
Mass ragequits during a contested vote can drain and kill a DAO's treasury faster than the bad proposal would have; only works cleanly for fungible, liquid treasuries.
EXAMPLES
MolochDAO, MetaCartel, Raid Guild, Nouns DAO (forked exit)
ANALOGS
A guild member banking their DKP and leaving rather than raid under a new loot rule (play); Exit as a safety-valve buffer against a runaway loop (systems, Meadows rung 11)
DATA
  • Moloch v2/Minion design popularized ragequit as a standard DAO primitive: open-sourced by Ameen Soleimani and collaborators github.com/MolochVentures/moloch · 2019-02
CHAIN · Exit · RUNG 11

Nouns DAO & the Fork-as-Exit Mechanism

Any 20% of the DAO can take their share of the treasury and start over.

Go deeper
STRUCTURE
Any member can trigger a fork window; if 20% quorum of Nouns joins, the DAO splits — forkers receive a proportional treasury withdrawal into a new contract while non-forkers keep the original — ragequit scaled from individual exit up to a whole faction's exit.
WHY IT IS LEVERAGE
Makes credible exit available to an organized minority, not just a lone dissenter, which forces the remaining majority to actually negotiate rather than simply outvote — exit as a structural check on majority capture, built into the DAO's own rules.
THE DRIVER
a faction's refusal to keep funding a direction it lost the vote on but still has to pay for
THE SHADOW
Each fork round has bled a large share of treasury to arbitrageurs who buy in right before a fork purely to ragequit for a cut, not out of real disagreement, turning the exit valve into an extraction mechanism.
EXAMPLES
Nouns DAO, Nouns forks (2023-2025)
ANALOGS
A guild splitting in two after a loot-rule dispute, each half taking its banked gold (play); Exit-as-buffer scaled from individual to coalition size (systems, Meadows rung 11)
DATA
SYSTEMS · Change · RUNG 6Added 7 October 2026

Granovetter's Threshold Models of Collective Behavior

Each person joins once enough others have joined, so the lowest thresholds decide whether a crowd forms.

Go deeper
STRUCTURE
Each person carries a threshold: the number or share of others who must act before they will. Start from the distribution of thresholds and the outcome can be worked out step by step. Whoever has a threshold of zero acts, that act tips everyone whose threshold is one, and the chain runs until it reaches the first gap. Two groups with the same average willingness can end in opposite places because one gap among the low thresholds stops the cascade.
WHY IT IS LEVERAGE
It moves attention from persuading the average person to the order and visibility of the first few. A first room, a first wave or a first night succeeds or fails on who is in it and on whether each next person can see the count, and a builder controls both.
THE DRIVER
the habit of reading a crowd's behavior as the sum of its members' opinions, and so working on opinions when the gap is in the sequence
THE SHADOW
Counts can be staged with fake early movers or inflated numbers, which buys one cascade and spends the trust that every later cascade needs. The model also says nothing about whether the crowd should move.
EXAMPLES
Granovetter's riot example, where a hundred people with thresholds from zero to ninety-nine all riot in turn, and raising one person's threshold from one to two leaves a single rioter; the paper's suggested applications to strikes, voting, migration and the spread of rumors and innovations
ANALOGS
Game design: matchmaking that seats a new player beside a friend who already plays; DAO: a proposal page that shows its supporter count so later voters can see it; Corpus: network effects describe the growth that follows once a first cascade completes
THE MOTUS EVOLUTION
Proposal: before any first room, wave or night, the host orders the people who said yes by the count each one needs, invites the lowest thresholds first, and shows a live count to everyone still deciding.
DATA
  • Threshold defined as the number or proportion of others who must decide before a given actor does, with the finding that groups with similar average preferences can produce very different outcomes: Threshold Models of Collective Behavior, Mark Granovetter, American Journal of Sociology sociology.stanford.edu/publications/threshold-models-collec… · 1978
SYSTEMS · Commons · RUNG 4Added 7 October 2026

Legitimate Peripheral Participation and Communities of Practice

Newcomers learn a practice by doing real, low-risk parts of it among people who already can.

Go deeper
STRUCTURE
Lave and Wenger studied apprenticeship and found that most learning ran through journeymen and more advanced apprentices, inside a community that acted as a living curriculum. A newcomer starts at the edge with tasks that are real and forgiving, and moves toward full participation as skill and identity grow. A community of practice has a shared domain, members who interact and learn together regularly, and a practice of shared methods and tools.
WHY IT IS LEVERAGE
It answers a limit set by one teacher's hours. A community that gives newcomers real work at its edge grows its own practitioners, and the people one step ahead do most of the teaching. The handle is the design of the first tasks.
THE DRIVER
the belief that learning happens in a course delivered by an expert, after which the learner goes out to practice alone
THE SHADOW
Edge work can harden into cheap labor that never leads inward, and a core group can guard the center so newcomers stay at the edge for good. The Wenger-Trayner introduction lists among common myths that communities of practice are always harmonious or a solution to everything.
EXAMPLES
The communities in the 1991 study: Yucatec midwives, Vai and Gola tailors, navy quartermasters, supermarket meat cutters and members of Alcoholics Anonymous; the Impressionists, who met in cafes and studios to work out a style they painted alone
ANALOGS
Game design: World 1-1 onboarding, where a new player does the real thing at low stakes; DAO: contributor ladders where a first bounty leads to a working group seat; Corpus: the founder-hours limit drawn on the MotusMoves page names guides who graduate from the practice as its way out
THE MOTUS EVOLUTION
Proposal: every body publishes the smallest real tasks a newcomer can finish in an hour, each one shipped work, and the person one step ahead signs each move inward.
DATA
  • The book that introduced legitimate peripheral participation as the way newcomers move toward full participation in a practice: Situated Learning: Legitimate Peripheral Participation, Jean Lave and Etienne Wenger, Cambridge University Press doi.org/10.1017/CBO9780511815355 · 1991
  • The term community of practice was coined while studying apprenticeship, for the community that acts as a living curriculum for the apprentice: Introduction to communities of practice, wenger-trayner wenger-trayner.com/introduction-to-communities-of-practice/ · 2022
SYSTEMS · Cybernetics · RUNG 3Added 7 October 2026

Double-Loop Learning (Argyris and Schön)

Correcting the action is one loop; questioning the goals and assumptions behind it is the second.

Go deeper
STRUCTURE
In single-loop learning an error is found and fixed within the existing goals and assumptions, the way a thermostat holds the temperature it was set to. In double-loop learning the error leads back to those governing variables, which are examined and changed. Argyris found that what people say guides them differs from what guides their actions, and that defensive routines protect those working assumptions from the errors that would revise them.
WHY IT IS LEVERAGE
Every instrument an organization builds feeds the first loop by default. Without a deliberate second loop, better measurement only makes the organization faster at pursuing a goal nobody may question.
THE DRIVER
defensive reasoning that treats a surfaced error as a threat, so bad news about a goal travels upward slowly or never
THE SHADOW
Second-loop talk can turn into endless reopening of every goal, so nothing holds still long enough to be tested. Argyris also found that the most successful professionals often resist it hardest, because they have had the least practice at failing.
EXAMPLES
Argyris's 1977 case of Product X, where at least five people knew the product was in serious trouble six years before management dropped it, at a loss above 100 million dollars; the thermostat that could ask why it is set where it is
ANALOGS
Game design: a balance team that rewrites the win condition after a third nerf to the same dominant build fails; DAO: a constitutional review called after repeated proposals fail for the same reason; Corpus: Goodhart's Law is what single-loop learning on a metric produces
THE MOTUS EVOLUTION
Proposal: every field reading and every bench run answers one more question, whether it changed a move or a rule, and each reading that would change a rule becomes a Votus for ratification.
DATA
  • Problem solving described as single-loop learning, and double-loop learning as the further step of reflecting on assumptions and testing them: Teaching Smart People How to Learn, Chris Argyris, Harvard Business Review hbr.org/1991/05/teaching-smart-people-how-to-learn · 1991
  • The failure to surface errors and unpleasant truths traced to faulty organizational learning, with double-loop learning as the method that reverses it: Double Loop Learning in Organizations, Chris Argyris, Harvard Business Review hbr.org/1977/09/double-loop-learning-in-organizations · 1977
SYSTEMS · Structure · RUNG 5Added 7 October 2026

Brand's Pace Layering

Fast-changing parts try things while slow-changing parts keep what works, and a healthy system needs both speeds.

Go deeper
STRUCTURE
Brand proposes six layers of a working civilization, ordered from fast to slow: fashion and art, commerce, infrastructure, governance, culture and nature. Fast layers propose and slow layers dispose. Small, fast layers instruct the large, slow ones through accumulated innovation and occasional revolution, and the slow layers constrain the fast ones through constancy. The same idea appears in his shearing layers of a building, from the furniture inside up to the site.
WHY IT IS LEVERAGE
It turns the question of what may change into a rule about how fast each kind of thing may change and who may change it. A system that lets its slowest layer move at the speed of its fastest loses its memory, and one that holds every layer still loses its ability to learn.
THE DRIVER
the wish to give everything one speed, either freezing all of it for consistency or shipping all of it at release pace
THE SHADOW
Layer labels can shield a favored layer from challenge, so a mistake gets called culture and is never fixed. Brand also notes that the fast layers get the attention while the slow layers hold the power.
EXAMPLES
Brand's six layers, first set out in The Clock of the Long Now in 1999; the historic preservation movement, through which culture told commerce it could change tenants and leave the buildings standing; the shearing layers of a building in How Buildings Learn
ANALOGS
Game design: a live game that keeps its core rules for years while seasons and cosmetics turn over monthly; DAO: a constitution amendable only by supermajority over months while working groups reset budgets every cycle; Corpus: Brand credits ecologists including C. S. Holling with the same question, which the adaptive cycle answers
THE MOTUS EVOLUTION
Proposal: Motus pace layers for the whole map, with August's lines and the Mantra in the slowest layer, definitions and rungs in a ratified middle layer, and passes and moves in the fast layer, and each fact kept in one home at its own layer.
DATA
  • Six levels of pace and size in an adaptable civilization, where the fast layers innovate, the slow layers stabilize, and the whole combines learning with continuity: Pace Layering: How Complex Systems Learn and Keep Learning, Stewart Brand, Long Now longnow.org/ideas/pace-layers/ · 2018
SYSTEMS · Loops · RUNG 7Added 7 October 2026

Hyde's The Gift: Gifts That Keep Moving

A gift keeps its force while it moves on to someone new, and the circle stops when it is sold.

Go deeper
STRUCTURE
Building on Mauss's 1925 essay on gift exchange, Hyde describes two economies that run side by side. In the gift economy what is received is passed on, often to a third person, so the increase belongs to the circle and binds its members to each other. In the market economy goods trade for equivalent value and the parties part even and unbound. A gift converted into a commodity stops circulating, and the bond it carried stops with it.
WHY IT IS LEVERAGE
It names a reinforcing loop that product design often breaks by accident: every gift passed on raises the chance of the next one. Where a paywall, a resale market or a reward for giving touches the gift path, the loop turns into exchange and the sense of community it produced goes with it.
THE DRIVER
the instinct to charge at the moment of greatest goodwill, because that is when people are most willing to pay
THE SHADOW
Gift language can disguise unpaid labor or a sales funnel. A gift that cannot be refused or returned becomes a debt, which the reciprocity model in this corpus already warns about.
EXAMPLES
Mauss's studies of exchange in Polynesia, Melanesia and the Pacific Northwest; the kula ring of the Trobriand Islands, where armshells and necklaces travel in opposite directions around a circle of islands; Hyde's argument that a work of art can survive without a market and cannot survive without the gift
ANALOGS
Game design: gifting systems that let a player pass a received item on to a third player; DAO: retroactive funding where past recipients fund the next round of public goods; Corpus: the gift loop drawn among the Motus tokens, where help earns a gift and gratitude makes the next ask easier
THE MOTUS EVOLUTION
Proposal: in every Motus body the act of giving stays outside every paywall and moves on to someone other than the giver, and money is charged only for the tools around the gift.
DATA
  • A defense of creativity and its value in a culture governed by money and commodities, first published in 1983: The Gift, Lewis Hyde lewishyde.com/the-gift/ · 1983
SYSTEMS · Strategy · RUNG 5Added 7 October 2026

Axelrod and Hamilton's Evolution of Cooperation

Cooperation built on reciprocity can start and hold when the same parties expect to meet again.

Go deeper
STRUCTURE
In a repeated Prisoner's Dilemma, a strategy that cooperates first and then copies the other side's last move won Axelrod's computer tournaments. Cooperation based on reciprocity can get started among a small cluster in a hostile population and resist invasion once established. It depends on the shadow of the future: the chance of meeting again must be high enough that defecting today costs more than it gains.
WHY IT IS LEVERAGE
It turns a trust slogan into conditions anyone can check. Trusting first wins when interactions repeat, when each side can see the other's past moves, and when defection meets a response. Take away repetition or visibility and the same trusting policy gets exploited.
THE DRIVER
the one-shot frame, where each deal is treated as the last and taking the most today looks rational
THE SHADOW
Strict tit for tat can lock two parties into echoing retaliation after one misread move, so working systems add forgiveness. A long shadow of the future can also be manufactured through lock-in that traps people without earning their return.
EXAMPLES
The 1980 computer tournaments, both won by Anatol Rapoport's TIT FOR TAT; the live-and-let-live system between opposing trenches in the First World War, analyzed in Axelrod's 1984 book; the paper's own suggested applications to territoriality, mating and disease
ANALOGS
Game design: persistent worlds where reputation follows a player from one session to the next; DAO: vesting and slashing that tie tomorrow's stake to today's behavior; Corpus: the reciprocity model shows the obligation of one gift, and this model shows what repetition adds
THE MOTUS EVOLUTION
Proposal: every Motus trust rail makes the next meeting likely and the last one visible, by vesting across seasons and by recording each delivery as a claim anyone can read.
DATA
  • A model of cooperation in the Prisoner's Dilemma whose deductions and computer tournament show cooperation based on reciprocity getting started in an asocial world and resisting invasion once established: The Evolution of Cooperation, Robert Axelrod and William D. Hamilton, Science pubmed.ncbi.nlm.nih.gov/7466396/ · 1981
SYSTEMS · Commons · RUNG 10Added 7 October 2026

Benkler's Commons-Based Peer Production

Strangers build large works together when the work splits into small pieces that cost little to integrate.

Go deeper
STRUCTURE
Beside markets and firms, Benkler identifies a third mode of production in which many people contribute to a shared resource without prices or managers. It thrives when a project is modular, so pieces can be made independently; fine grained, so the smallest useful piece takes little effort, ideally alongside pieces of other sizes; and cheap to integrate, so checking and assembling contributions costs little, often because software does it. Benkler argues those three properties set its limit, whatever the size or complexity of the project.
WHY IT IS LEVERAGE
The handle is structural and sits with the builder. Cutting the work into smaller modules and automating the check that admits each one widens the pool of people who can contribute, and a small motivation per person is enough when each piece is small.
THE DRIVER
the assumption that serious work needs either payment per task or a manager assigning it
THE SHADOW
When integration is expensive, the integrator ends up taking the residual value of the shared project and contributors stop showing up, a failure Benkler names directly. Firms can also take from an open commons and give nothing back.
EXAMPLES
Linux and free software; NASA Clickworkers, where volunteers marked craters on Mars images in sessions of a few minutes; Distributed Proofreaders, which split Project Gutenberg proofing into single pages
ANALOGS
Game design: user-made levels that pass an automatic validation step before they publish; DAO: bounty boards with automated tests as the merge check; Corpus: Ostrom's principles describe how a commons defends itself once contributions arrive
THE MOTUS EVOLUTION
Proposal: every open body publishes its smallest useful contribution, such as one bench item or one model card, together with the automated check that admits it, so a stranger can add one piece in under an hour.
DATA
  • Peer production is limited by the modularity and granularity of a project and by the cost of integrating contributions, which covers both quality control and assembly: Coase's Penguin, or, Linux and The Nature of the Firm, Yochai Benkler, Yale Law Journal yalelawjournal.org/article/coases-penguin-or-linux-and-the-… · 2002
SYSTEMS · Strategy · RUNG 6Added 7 October 2026

Wardley Mapping

Map the chain of needs, place each part on its path from novel to commodity, and only then decide what to build.

Go deeper
STRUCTURE
A map starts from a user need and draws the chain of components that meets it, ordered by how visible each one is to the user. The other axis is evolution. Supply and demand competition moves every component from the uncharted, where things are uncertain, rare and changing, toward the industrialized, where they are known, common and stable, through the stages Wardley calls genesis, custom built, product and commodity. Position decides method: explore the novel parts, and buy or rent the commodity ones.
WHY IT IS LEVERAGE
A small team's scarcest stock is builder hours. The map shows where those hours change what users get, in the genesis and custom parts they can see, and where the hours go into rebuilding something already sold as a commodity.
THE DRIVER
the pull to build every component in house because building is the enjoyable part
THE SHADOW
A map drawn alone becomes a confident picture of guesses, since evolution positions are judgments that need challenge from other people. Treating a component as commodity too early can also hand a distinguishing piece to a vendor.
EXAMPLES
Wardley's map of a television company, drawn from viewers' need to be entertained; power supply, a commodity everyone plugs into without thinking about it; Nokia, whose value chains changed as it went from paper mill to plastics to telecommunications
ANALOGS
Game design: studios that license engines and middleware and build the parts players see; DAO: rollups and data availability bought as infrastructure while governance stays custom; Corpus: the Theory of Constraints finds the bottleneck, and the map shows which side of it to build on
THE MOTUS EVOLUTION
Proposal: each body keeps one map that marks every component genesis, custom, product or commodity, and builder hours go only to the first two.
DATA
SYSTEMS · Cybernetics · RUNG 8Added 7 October 2026

Shewhart's Control: Chance Causes and Assignable Causes

A process is in control when past data predicts its variation within limits, and only points outside those limits call for a hunt.

Go deeper
STRUCTURE
Shewhart separates the variation in any process into a constant system of chance causes, which the process produces by itself, and assignable causes, which can be found and removed. A phenomenon is controlled when past experience lets you predict, within limits, how it will vary in future. Limits computed from the process's own data become the yardstick. Points inside them call for work on the system as a whole, and points outside them call for finding the specific cause.
WHY IT IS LEVERAGE
It sets the gain of the balancing loop around every metric. Reacting to each point as if it were a signal can add variation, while ignoring real signals lets a shift run on. The limits separate the two before anyone moves.
THE DRIVER
the urge to explain and act on every rise and fall of a number, especially one shown live on a public page
THE SHADOW
Limits computed from too few points give false comfort, and a process can be stable at a bad level. Control says the system is predictable and leaves open whether it is good.
EXAMPLES
Shewhart's example of writing the letter a many times, where no two are alike and yet each writer's a's vary within a range of their own; the eclipse of the sun, a phenomenon predictable with great precision; the control chart that industry took up from his work
ANALOGS
Game design: live operations teams that wait for a retention change to clear normal week to week noise before patching; DAO: treasury dashboards that flag only spending outside its usual range; Corpus: Goodhart's Law warns about the target, and this model warns about the reaction
THE MOTUS EVOLUTION
Proposal: every Motus counter shows its run with limits computed from its own history, and says whether the latest point falls inside them, before anyone writes a move about it.
DATA
  • A phenomenon is controlled when past experience lets one predict, within limits, how it may vary, and assignable causes of variation may be found and eliminated: Economic Control of Quality of Manufactured Product, W. A. Shewhart dspace.gipe.ac.in/xmlui/handle/10973/18338 · 1931
SYSTEMS · Complexity · RUNG 4Added 7 October 2026

Holling's Adaptive Cycle and Panarchy

Growth stores capital until the system locks up, and the release that follows is where new forms start.

Go deeper
STRUCTURE
The adaptive cycle runs through four phases. Rapid growth gives way to conservation, where capital accumulates and connections tighten until the system grows rigid. A disturbance then releases what was locked up, and in reorganization the freed resources recombine and new forms get their chance. Cycles nest across scales into a panarchy: slower, larger levels protect faster ones and hold their memory, while faster, smaller cycles invent and can set off change in the levels above.
WHY IT IS LEVERAGE
It says when intervention is cheap. Late in conservation a system resists change and every fix costs more, while in release and reorganization small moves set the structure of the next cycle. It also makes endings a phase whose freed resources can be steered.
THE DRIVER
the wish to keep any successful thing growing forever and to treat every release as failure
THE SHADOW
The cycle can be used to romanticize collapse or to excuse letting things break. Holling's work also describes traps, such as rigid systems that cannot release and poor ones that never store enough to grow.
EXAMPLES
Boreal forests where spruce budworm outbreaks release decades of stored growth; managed forests where fire suppression prolongs conservation and makes the eventual fire larger; seasonal programs that close each season and reopen with what they kept
ANALOGS
Game design: seasonal resets that release accumulated advantage so new players can reshape the ladder; DAO: sunset clauses that dissolve working groups at the end of each term; Corpus: the two loops model draws the release beside the new loop, and pace layering draws the scales
THE MOTUS EVOLUTION
Proposal: every season and every wave states at its start what it will release at its end and where the freed people and money go next, so each ending seeds the next cycle.
DATA
  • Hierarchies and adaptive cycles across scales form a panarchy in which each level runs at its own pace, protected by slower levels and invigorated by faster ones, combining learning with continuity: Understanding the Complexity of Economic, Ecological, and Social Systems, C. S. Holling, Ecosystems link.springer.com/article/10.1007/s10021-001-0101-5 · 2001
SYSTEMS · Structure · RUNG 4Added 7 October 2026

Alexander's Pattern Language and the Fifteen Properties

A shared language of solved design problems lets ordinary people make places that feel alive.

Go deeper
STRUCTURE
A Pattern Language gives more than 250 patterns, each made of a problem statement, a discussion with an illustration, and a solution, at scales from towns down to window seats. Patterns connect: each one completes larger patterns and is completed by smaller ones, so a sequence of patterns generates a coherent design. In The Nature of Order Alexander later named fifteen structural properties that recur in things that have life, such as levels of scale and strong centers, and used them to judge whether a made thing works as a whole.
WHY IT IS LEVERAGE
It moves design power to the people who live with the result. Once the language is shared, any group can design for itself and still get a coherent whole. A catalog written as patterns becomes generative, because each entry points to what it needs above and below it.
THE DRIVER
the professional habit of handing people finished designs, which leaves them no way to design their own
THE SHADOW
Patterns get copied as a style without their problem statements, which keeps the look and loses the reason. A language presented as timeless can also shut out forms that a new community needs.
EXAMPLES
Patterns such as Light on Two Sides of Every Room and Street Cafe; the Oregon Experiment on the University of Oregon campus; the software design patterns movement and the first wiki, both built on the form Alexander gave patterns
ANALOGS
Game design: level kits whose rooms and encounters combine by rules players come to learn; DAO: governance templates that compose into a constitution; Corpus: the Sanctum's starting structures already compose like patterns, and each Frontier pattern states its own smallest version
THE MOTUS EVOLUTION
Proposal: each model in this corpus names the larger patterns it completes and the smaller ones that complete it, so the Sanctum can compose models the way Alexander's patterns compose a town.
DATA
  • More than 250 patterns, each a problem statement, a discussion and a solution, offered as a language any person or group can use to design their own surroundings: A Pattern Language, Christopher Alexander, Sara Ishikawa and Murray Silverstein, Oxford University Press patternlanguage.com/bookstore/pattern-language.html · 1977
  • Fifteen structural features that appear again and again in things which have life: The Nature of Order, Christopher Alexander, The Center for Environmental Structure patternlanguage.com/bookstore/nature-of-order.html · 2004
SYSTEMS · Change · RUNG 2Added 7 October 2026

Emergent Strategy (adrienne maree brown)

Large change is built from small, intentional interactions, at the pace trust allows.

Go deeper
STRUCTURE
Drawing on Octavia Butler's fiction and on patterns found in nature, brown treats change as constant and shapeable. Her principles hold that the large reflects the small, so the quality of the smallest interaction sets the quality of the whole. People who are trusted become trustworthy. Work moves at the speed of trust, and critical connections among a few people build more resilience than a critical mass of shallow ones.
WHY IT IS LEVERAGE
It works on what a group believes about how change happens, and that belief governs every smaller choice. A group that believes in critical mass spends on reach. A group that believes in critical connections invests in the few relationships that will carry the next step.
THE DRIVER
the scale reflex, where a movement measures itself by the size of its crowd and hurries to grow it
THE SHADOW
Small and slow can become a reason never to test anything at scale, and trust-first principles can be exploited by people who never return the trust. The evolution of cooperation model names the conditions that guard against that.
EXAMPLES
The book's roots in Octavia Butler's Parable novels; organizers who plan a meeting around the conversation only the people in that room can have; the teachers from nature that the book draws on, such as starling murmurations and mycelium
ANALOGS
Game design: small guild raids that build the trust a large alliance later depends on; DAO: pods and working groups that prove a practice before the whole DAO adopts it; Corpus: Meadows' Dancing with Systems is its analytic partner, and the threshold model is its counterweight on how crowds tip
THE MOTUS EVOLUTION
Proposal: each body sizes its next step to the trust it has already earned, counted as relationships that have delivered at least once, and grows that count before it courts a crowd.
DATA
  • A book on shaping change through small interactions and decentralized leadership, drawing on Octavia Butler: Emergent Strategy: Shaping Change, Changing Worlds, adrienne maree brown, AK Press akpress.org/emergentstrategy.html · 2017
  • Its core principles, including small is good and small is all, trust the people, and critical connections more than critical mass, as summarized by an organizing library: Emergent Strategy, Commons Library commonslibrary.org/emergent-strategy/ · 2024

12 · SOURCES

Receipts.

179 unique sources behind 192 data points. Every number on this page links to one of them.

Open every source

PLAY · 51

  1. users.cs.northwestern.edu/~hunicke/MDA.pdf · 2004-01 · MDA Framework
  2. goodreads.com/quotes/990231-that-s-what-games-are-in-the-end-… · 2004-01 · Theory of Fun: Fun Is Learning
  3. en.wikipedia.org/wiki/Magic_circle_(game_studies) · 2003-01 · The Magic Circle
  4. who.int/teams/mental-health-and-substance-use/alcohol-drugs-a… · 2019-05 · WHO Gaming Disorder, Prevalence & the Telemetry Correction
  5. pubmed.ncbi.nlm.nih.gov/36433839/ · 2021-02 · WHO Gaming Disorder, Prevalence & the Telemetry Correction
  6. ora.ox.ac.uk/objects/uuid:4aedc046-eaab-4b59-8239-546584f36ed8 · 2021-02 · WHO Gaming Disorder, Prevalence & the Telemetry Correction
  7. vuorre.com/bibliography/files/3K5WXM3G/vuorre-etal_2022_royal… · 2022-03 · WHO Gaming Disorder, Prevalence & the Telemetry Correction
  8. warcraft.wiki.gg/wiki/Raid · 2008-11 · Guilds, Raids & Asymmetric Co-op Roles
  9. forums.eveonline.com/t/monthly-economic-report-february-2026/… · 2026-03 · EVE Online: Player-Run Economy + Elected Council
  10. nosygamer.blogspot.com/2026/02/a-different-look-at-eve-online… · 2026-02 · EVE Online: Player-Run Economy + Elected Council
  11. investing.com/news/company-news/roblox-q1-2026-slides-strong-… · 2026-05 · UGC Platforms & Creator Payout Economies
  12. thebloxline.com/articles/roblox-now-has-123-million-daily-use… · 2026-08 · UGC Platforms & Creator Payout Economies
  13. tubefilter.com/2026/06/17/epic-games-unreal-editor-for-fortni… · 2026-06 · UGC Platforms & Creator Payout Economies
  14. axis-intelligence.com/minecraft-statistics/ · 2026-09 · UGC Platforms & Creator Payout Economies
  15. yukaichou.com/gamification-study/white-hat-black-hat-gamifica… · 2026-09 · Octalysis — White-Hat vs Black-Hat Drives
  16. researchgate.net/publication/309020043_The_Gamer_Motivation_P… · 2016-10 · Bartle Types & the Quantic Foundry Motivation Model
  17. nirandfar.com/hooked/ · 2014-01 · The Hooked Model & Its Critiques
  18. doi.org/10.1037/a0034857 · 2014-01 · Prosocial & Cooperative Play Benefits
  19. gamedeveloper.com/design/gdc-2012-sid-meier-on-how-to-see-gam… · 2012-03 · A Series of Interesting Decisions
  20. cheatcc.com/articles/breath-of-the-wild-s-chemistry-engine-ex… · 2017-03 · Emergent / Systemic Design
  21. arxiv.org/html/2606.29511 · 2026-06 · Onboarding: Teach by Play (World 1-1)
  22. en.wikipedia.org/wiki/Reciprocity_(social_psychology) · 2026-09 · Reciprocity & Social Obligation Mechanics
  23. dl.digra.org/index.php/dl/article/download/889/889/886 · 2013-05 · Dark Patterns in the Design of Games
  24. lostgarden.com/2012/04/30/loops-and-arcs/ · 2012-04 · Skill Atoms & Loops-and-Arcs
  25. gdcvault.com/play/1016487/juice-it-or-lose · 2012-08 · Game Feel / Juice
  26. blog.udonis.co/mobile-marketing/mobile-games/among-us-player-… · 2026-01 · Social Deduction: Hidden Roles + Public Voting
  27. selfdeterminationtheory.org/SDT/documents/2006_RyanRigbyPrzyb… · 2006-12 · Self-Determination Theory & PENS
  28. its.caltech.edu/~jkenny/nb250c/papers/Schultz-1997.pdf · 1997-03 · Dopamine Reward-Prediction Error
  29. en.wikipedia.org/wiki/Cookie_Clicker · 2013-08 · Idle / Incremental Games
  30. en.wikipedia.org/wiki/Reinforcement · 2026-09 · Operant Conditioning, Variable-Ratio Schedules & the Near-Miss Effect
  31. doi.org/10.1016/j.neuron.2008.12.031 · 2009-02 · Operant Conditioning, Variable-Ratio Schedules & the Near-Miss Effect
  32. en.wikipedia.org/wiki/Social_comparison_theory · 2026-09 · Status & Social Comparison
  33. gamedeveloper.com/design/revisiting-the-mda-framework · 2021-05 · Flow Channel & Dynamic Difficulty Adjustment
  34. en.wikipedia.org/wiki/Skill-based_matchmaking · 1970-01 · Ranked Matchmaking & Elo / MMR
  35. en.wikipedia.org/wiki/Flow_(psychology) · 2026-09 · Flow & the Challenge-Skill Balance
  36. journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0… · 2018-11 · Loot Boxes, Problem Gambling & the Regulatory Response
  37. thesixthaxis.com/2018/04/25/belgium-says-overwatch-fifa-18-cs… · 2018-04 · Loot Boxes, Problem Gambling & the Regulatory Response
  38. washingtonpost.com/video-games/2022/07/19/video-game-look-box… · 2022-07 · Loot Boxes, Problem Gambling & the Regulatory Response
  39. ftc.gov/news-events/news/press-releases/2022/12/fortnite-vide… · 2022-12 · Loot Boxes, Problem Gambling & the Regulatory Response
  40. thewirechina.com/2024/01/14/beijing-plays-a-dangerous-game-wi… · 2023-12 · Loot Boxes, Problem Gambling & the Regulatory Response
  41. en.wikipedia.org/wiki/Battle_pass · 2018-02 · Live Service Seasons & Battle Pass
  42. stepico.com/gaming/how-much-money-has-fortnite-made/ · 2026-01 · Live Service Seasons & Battle Pass
  43. jstor.org/stable/1914185 · 1979-03 · Loss Aversion
  44. siliconcanals.com/t-car-wash-loyalty-cards-endowed-progress/ · 2006-02 · Endowed Progress Effect & the Goal-Gradient Hypothesis
  45. yukaichou.com/behavioral-analysis/goal-gradient-hypothesis-hu… · 2006-02 · Endowed Progress Effect & the Goal-Gradient Hypothesis
  46. en.wikipedia.org/wiki/Zeigarnik_effect · 2026-09 · Zeigarnik Effect, Open Loops & Completionism
  47. blog.duolingo.com/how-streaks-keep-duolingo-learners-committe… · 2023-01 · Streaks & Sunk Cost
  48. academic.oup.com/jrsssa/article/185/1/102/7068450 · 2022-01 · Progression: XP, Levels, Skill Trees & Prestige
  49. gamedeveloper.com/design/the-designer-s-notebook-machinations… · 2011-01 · Internal Economy: Faucets, Sinks & Machinations
  50. thegamer.com/hades-sales-1-million-copies/ · 2020-10 · Roguelike Meta-Progression
  51. shanethegamer.com/research/genshin-impact-revenue-statistics/ · 2025-12 · Gacha & Loot Boxes

SYSTEMS · 30

  1. donellameadows.org/dancing-with-systems/ · 2001 · Meadows' Dancing with Systems
  2. donellameadows.org/wp-content/userfiles/Leverage_Points.pdf · 1999 · Meadows' Twelve Leverage Points
  3. en.wikipedia.org/wiki/Twelve_leverage_points · 2024 · Meadows' Twelve Leverage Points
  4. internationalfuturesforum.com/p/three-horizons-the-patterning… · 2013 · Three Horizons and the Berkana Two Loops
  5. chriscorrigan.com/parkinglot/the-two-loops-model-of-change-pa… · 2004 · Three Horizons and the Berkana Two Loops
  6. en.wikipedia.org/wiki/The_purpose_of_a_system_is_what_it_does · 1970s · Viable System Model and POSIWID
  7. systemsthinkingalliance.org/stafford-beer-the-father-of-manag… · 1972 · Viable System Model and POSIWID
  8. santafe.edu/about/history · 1984 · Emergence and Self-Organization
  9. sciencedirect.com/science/article/abs/pii/S0048733302000628 · 2002 · Geels' Multi-Level Perspective on Transitions
  10. idratherbewriting.com/trends/systems-thinking-fifth-disciplin… · 1990 · Archetype: Tragedy of the Commons
  11. pdfcoffee.com/daniel-kim-systems-archetypes-pdf-pdf-free.html · 1992 · Archetype: Success to the Successful
  12. research.fit.edu/media/site-specific/researchfitedu/coast-cli… · 2008 · Meadows' System Traps and Their Ways Out
  13. nobelprize.org/prizes/economic-sciences/2009/summary/ · 2009 · Ostrom's Eight Design Principles for Governing the Commons
  14. dlc.dlib.indiana.edu/dlc/bitstreams/159de9c2-ac05-4088-8179-9… · 1990 · Ostrom's Eight Design Principles for Governing the Commons
  15. en.wikipedia.org/wiki/Goodhart%27s_law · 1975 · Goodhart's Law and Campbell's Law
  16. en.wikipedia.org/wiki/Goodhart's_law · 1979 · Goodhart's Law and Campbell's Law
  17. nobelprize.org/prizes/economic-sciences/2007/summary/ · 2007 · Mechanism Design
  18. pespmc1.vub.ac.be/REQVAR.html · 1956 · Ashby's Law of Requisite Variety
  19. hbr.org/2007/11/a-leaders-framework-for-decision-making · 2007-11 · Cynefin Framework
  20. en.wikipedia.org/wiki/OODA_loop · 1976 · OODA Loop
  21. melconway.com/research/committees.html · 1968 · Conway's Law
  22. arxiv.org/abs/cond-mat/9910332 · 1999 · Network Effects and Preferential Attachment
  23. web.mit.edu/jsterman/www/Order_Stability_070505.pdf · 1989 · Delays and Oscillation: the Beer Game
  24. pubsonline.informs.org/doi/10.1287/mnsc.35.3.321 · 1989 · Delays and Oscillation: the Beer Game
  25. en.wikipedia.org/wiki/Fixes_that_fail · 1992 · Archetype: Fixes that Fail
  26. en.wikipedia.org/wiki/Industrial_Dynamics · 1961 · Stocks and Flows
  27. tocinstitute.org/the-goal-summary.html · 1984 · Theory of Constraints
  28. en.wikiquote.org/wiki/John_Gall · 1975 · Gall's Law
  29. oreilly.com/library/view/game-mechanics-advanced/978013294672… · 2012 · Dormans' Machinations
  30. en.wikipedia.org/wiki/Antifragile_(book) · 2012 · Antifragility

CHAIN · 82

  1. coinbase.com/developer-platform/discover/launches/x402 · 2026-04 · AI Agents + Crypto-Native Payments
  2. allium.so/blog/onchain-ai-identity-what-erc-8004-unlocks-for-… · 2026-01 · AI Agents + Crypto-Native Payments
  3. slatestarcodex.com/2014/07/30/meditations-on-moloch/ · 2014-07 · Meditations on Moloch
  4. vitalik.eth.limo/general/2020/01/28/credible_neutrality.html · 2020-01 · Credible Neutrality
  5. stablecoins.llama.fi/stablecoins?includePrices=true · 2026-10-07 · Stablecoins
  6. whitehouse.gov/briefings-statements/2025/07/the-president-sig… · 2025-07-18 · Stablecoins
  7. lw.com/en/insights/the-genius-act-of-2025-stablecoin-legislat… · 2025-07 · Stablecoins
  8. walkersglobal.com/en/Insights/2025/01/The-EUs-crypto-regulati… · 2024-06 · Stablecoins
  9. cryptoslate.com/crypto-laws/executive-order-14233-strategic-b… · 2025-03 · Sovereign & Corporate Bitcoin Treasuries
  10. lambdafin.com/articles/microstrategy-bitcoin-holdings · 2026-10 · Sovereign & Corporate Bitcoin Treasuries
  11. solanacompass.com/news/metadao-renames-futardio-as-backable-i… · 2026-10 · Futarchy: Vote on Values, Bet on Beliefs
  12. solanacompass.com/learn/Lightspeed/how-metadao-became-solanas… · 2026-09 · Futarchy: Vote on Values, Bet on Beliefs
  13. plurality.institute/blog-posts/book-launch-plurality-the-futu… · 2024-05 · Plurality (Weyl & Tang)
  14. beincrypto.com/pump-fun-meme-coins-trader-losses/ · 2026-01 · Memecoins & pump.fun as Attention Markets
  15. cip.org/blog/ccai · 2024-10 · Collective Intelligence Project — Alignment Assemblies
  16. cip.org/blog/alignment-assemblies-nine-months-in · 2025-02 · Collective Intelligence Project — Alignment Assemblies
  17. defillama.com/ · 2026-10 · Smart Contracts, Composability & Money Legos
  18. blockeden.xyz/blog/2026/03/21/depin-march-2026-reality-check-… · 2026-03 · DePIN: Decentralized Physical Infrastructure
  19. doi.org/10.1177/21582440211002526 · 2021-03 · Ostrom's Commons Principles, On-Chain
  20. docs.juicebox.money/blog/constitutiondao-highlight/ · 2021-11 · ConstitutionDAO
  21. en.wikipedia.org/wiki/ConstitutionDAO · 2021-11 · ConstitutionDAO
  22. cryptonews.net/news/finance/2808797/ · 2021-11 · ConstitutionDAO
  23. auditone.io/blog-posts/key-eips-in-the-ethereums-pectra-upgrade · 2025-05 · Account Abstraction, Passkeys & Intents
  24. gitcoin.co/blog/gitcoin-grants-23-retro · 2025-11 · Quadratic Voting & Quadratic Funding
  25. arxiv.org/abs/1809.06421 · 2018-09 · Quadratic Voting & Quadratic Funding
  26. github.com/Hats-Protocol/hats-protocol · 2025-06 · Hats Protocol — On-Chain Roles
  27. world.org/blog/announcements/world-id-full-stack-proof-of-human · 2026-04 · Proof of Personhood & Sybil Resistance
  28. coindesk.com/business/2025/02/10/digital-identity-startup-hol… · 2025-02 · Proof of Personhood & Sybil Resistance
  29. cryptotimes.io/2023/07/24/vitalik-shares-his-view-on-worldcoi… · 2023-07 · Proof of Personhood & Sybil Resistance
  30. frontiersin.org/journals/political-science/articles/10.3389/f… · 2020-11 · Citizens' Assemblies & Sortition
  31. sortitionfoundation.org/embarrassment_for_irish_politicians · 2024-03 · Citizens' Assemblies & Sortition
  32. chain.link/article/proposer-builder-separation-pbs · 2026-09 · MEV & Proposer-Builder Separation
  33. ryder.id/blogs/post/bitcoin-etf-inflows-in-2026-the-numbers-b… · 2026-09 · Spot BTC/ETH ETFs
  34. techi.com/ethereum-etf-staking-sec-decision/ · 2026-03 · Spot BTC/ETH ETFs
  35. pewresearch.org/short-reads/2026/09/23/prediction-markets-tra… · 2026-07 · Prediction Markets as Information Engines
  36. pewresearch.org/short-reads/2026/05/27/trading-volume-on-pred… · 2026-07 · Prediction Markets as Information Engines
  37. ccn.com/education/gambling/prediction-markets-53b-kalshi-poly… · 2026-06 · Prediction Markets as Information Engines
  38. papers.ssrn.com/sol3/papers.cfm?abstract_id=7103878 · 2025-01 · Liquid Democracy & Delegation
  39. arxiv.org/abs/2510.05830 · 2025-10 · Liquid Democracy & Delegation
  40. researchgate.net/publication/410908562_An_Empirical_Evaluatio… · 2025-01 · Liquid Democracy & Delegation
  41. discuss.ens.domains/t/ens-delegate-accountability-what-the-da… · 2026-04 · Liquid Democracy & Delegation
  42. daotimes.com/deepdao-dao-tool-report-for-2025 · 2025-01 · Liquid Democracy & Delegation
  43. etherscan.io/token/0x543ff227f64aa17ea132bf9886cab5db55dcaddf · 2026-09 · Holographic Consensus (DAOstack)
  44. papers.ssrn.com/sol3/papers.cfm?abstract_id=4105763 · 2022-05 · Attestations (EAS) & Soulbound Tokens
  45. x.com/eas_eth · 2026-10 · Attestations (EAS) & Soulbound Tokens
  46. platformer.news/meta-community-notes-launch/ · 2025-01 · Community Notes / Bridging-Based Ranking
  47. tech.yahoo.com/social-media/articles/tiktok-adding-community-… · 2025-04 · Community Notes / Bridging-Based Ranking
  48. coindesk.com/consensus-magazine/2024/05/22/audrey-tang-learni… · 2024-05 · Pol.is & vTaiwan
  49. cointelegraph.com/news/near-foundation-ai-delegates-dao-voting · 2026-02 · AI Delegates & AI-Assisted Governance
  50. bex.co/blog/2026/02/05/ai-agent-governance-crisis-who-governs… · 2026-02 · AI Delegates & AI-Assisted Governance
  51. web3futurepro.com/how-daos-are-changing-governance-in-2026/ · 2026-02 · AI Delegates & AI-Assisted Governance
  52. thedefiant.io/news/defi/layerzero-labs-identifies-800-000-pot… · 2024-05 · Points Programs & Airdrops
  53. api.coingecko.com/api/v3/coins/smooth-love-potion · 2026-10-07 · Play-to-Earn Collapse: Axie Infinity & Ronin
  54. en.wikipedia.org/wiki/Axie_Infinity · 2026-01 · Play-to-Earn Collapse: Axie Infinity & Ronin
  55. coindesk.com/tech/2022/03/29/axie-infinitys-ronin-network-suf… · 2022-03 · Play-to-Earn Collapse: Axie Infinity & Ronin
  56. defillama.com/protocol/hyperliquid · 2026-10-07 · Hyperliquid: Fully Onchain Order-Book Perps
  57. aminagroup.com/research/hyperliquid-hype-etf-buyback-staking-… · 2026-05 · Hyperliquid: Fully Onchain Order-Book Perps
  58. optimism.io/blog/retro-funding-2025 · 2025-01 · Retroactive Public Goods Funding (RetroPGF)
  59. optimism.io/blog/announcing-the-results-of-retropgf-2 · 2023-12 · Retroactive Public Goods Funding (RetroPGF)
  60. chainalysis.com/blog/bitcoin-halving-2024/ · 2024-04 · Nakamoto Consensus & Proof-of-Work
  61. news.bitcoin.com/bitcoins-first-post-halving-adjustment-raise… · 2024-04 · Nakamoto Consensus & Proof-of-Work
  62. defillama.com/protocol/aave · 2026-10-07 · Overcollateralized Lending & Liquidations
  63. eips.ethereum.org/EIPS/eip-1559 · 2019-04 · EIP-1559 Base-Fee Burn
  64. theblock.co/post/360212/lido-dao-votes-to-enable-dual-governa… · 2025-06 · Dual Governance (Optimistic Veto)
  65. okx.com/en-us/orbit/insight/88798683489824 · 2026-09 · Dual Governance (Optimistic Veto)
  66. blog.succinct.xyz/real-time-proving-16-gpus/ · 2025-11 · ZK Proofs & Real-Time zkVMs
  67. bloomberg.com/news/articles/2022-04-18/defi-project-beanstalk… · 2022-04 · Flash-Loan Governance Attacks
  68. coindesk.com/business/2022/10/20/mango-markets-will-start-ret… · 2022-10 · Flash-Loan Governance Attacks
  69. defillama.com/protocol/uniswap · 2026-10 · Automated Market Makers (x times y = k) & Bonding Curves
  70. bex.co/blog/2026/03/20/eigenlayer-18b-tvl-vertical-avs-specia… · 2026-03 · Restaking & Shared Security
  71. blog.ethereum.org/2025/11/06/fusaka-mainnet-announcement · 2024-03 · Rollups & Modular Data Availability
  72. coindesk.com/tech/2025/10/30/ethereum-developers-lock-in-fusa… · 2025-12 · Rollups & Modular Data Availability
  73. app.rwa.xyz/ · 2026-07 · Tokenized Real-World Assets
  74. yellow.com/research/tokenized-rwas-31b-market-growth-real-rac… · 2026-07 · Tokenized Real-World Assets
  75. cryptotimes.io/2026/05/02/circle-vs-blackrock-15b-tokenized-t… · 2026-05 · Tokenized Real-World Assets
  76. github.com/1Hive/conviction-voting-app · 2024-01 · Conviction Voting
  77. techflowpost.com/en-US/article/18681 · 2025-09 · Protocol Guild
  78. protocolguild.org/blog/20250909-compensation-insights-for-eth… · 2025-09 · Protocol Guild
  79. kucoin.com/blog/ethereum-reaches-historic-staking-milestone-w… · 2026-05 · Proof-of-Stake Staking & Slashing
  80. github.com/MolochVentures/moloch · 2019-02 · Moloch DAO Rage-Quit
  81. blockworks.co/news/nouns-dao-treasury-fork-governance · 2023-09 · Nouns DAO & the Fork-as-Exit Mechanism
  82. decrypt.co/197400/nouns-fork-disgruntled-nft-holders-exit-27-… · 2023-09 · Nouns DAO & the Fork-as-Exit Mechanism

ADDED 7 OCTOBER 2026 · 16

  1. sociology.stanford.edu/publications/threshold-models-collec… · 1978 · Granovetter's Threshold Models of Collective Behavior
  2. doi.org/10.1017/CBO9780511815355 · 1991 · Legitimate Peripheral Participation and Communities of Practice
  3. wenger-trayner.com/introduction-to-communities-of-practice/ · 2022 · Legitimate Peripheral Participation and Communities of Practice
  4. hbr.org/1991/05/teaching-smart-people-how-to-learn · 1991 · Double-Loop Learning (Argyris and Schön)
  5. hbr.org/1977/09/double-loop-learning-in-organizations · 1977 · Double-Loop Learning (Argyris and Schön)
  6. longnow.org/ideas/pace-layers/ · 2018 · Brand's Pace Layering
  7. lewishyde.com/the-gift/ · 1983 · Hyde's The Gift: Gifts That Keep Moving
  8. pubmed.ncbi.nlm.nih.gov/7466396/ · 1981 · Axelrod and Hamilton's Evolution of Cooperation
  9. yalelawjournal.org/article/coases-penguin-or-linux-and-the-… · 2002 · Benkler's Commons-Based Peer Production
  10. blog.gardeviance.org/2015/02/an-introduction-to-wardley-val… · 2015 · Wardley Mapping
  11. dspace.gipe.ac.in/xmlui/handle/10973/18338 · 1931 · Shewhart's Control: Chance Causes and Assignable Causes
  12. link.springer.com/article/10.1007/s10021-001-0101-5 · 2001 · Holling's Adaptive Cycle and Panarchy
  13. patternlanguage.com/bookstore/pattern-language.html · 1977 · Alexander's Pattern Language and the Fifteen Properties
  14. patternlanguage.com/bookstore/nature-of-order.html · 2004 · Alexander's Pattern Language and the Fifteen Properties
  15. akpress.org/emergentstrategy.html · 2017 · Emergent Strategy (adrienne maree brown)
  16. commonslibrary.org/emergent-strategy/ · 2024 · Emergent Strategy (adrienne maree brown)

Foundations cited in the synthesis: Donella Meadows, Leverage Points (1999) and Thinking in Systems (2008); Elinor Ostrom, Governing the Commons (1990); Satoshi Nakamoto, Bitcoin (2008); Jakob Nielsen, Response Times: The 3 Important Limits (1993).