Creating: Map, Move, Make

Working notes, not prose. This page is a research packet for drafting the chapter: the brief, source material, questions, examples, reader perspectives, and the earlier draft.

Brief

Must: Show intention meeting material constraint through bearings, cheap attempts and revisable results, with if-then plans and stopping rules included.
Serves: It is the mode where information returns to matter and changes the world.

Quote options

“Everyone designs who devises courses of action aimed at changing existing situations into preferred ones.” — Herbert A. Simon, The Sciences of the Artificial (1969), ch. 3 “The Science of Design,” p. 55 (1st ed.). [verified: https://academics.design.ncsu.edu/student-publication/wp-content/uploads/2016/11/Simon_H_ScienceofArtificial.pdf (Design Issues reprint)] Why: Defines creating as intention turned back onto the world, where information returns to matter.
“The situation talks back, the practitioner listens, and as he appreciates what he hears, he reframes the situation once again.” — Donald Schön, The Reflective Practitioner (1983), pp. 131–132. [verified: https://sopper.dk/speciale/arkiv/book49.pdf (excerpt compilation with page numbers)] Why: Shows material pushing back on intention: cheap moves, back-talk, and results you can revise.
“Every good craftsman conducts a dialogue between concrete practices and thinking; this dialogue evolves into sustaining habits, and these habits establish a rhythm between problem solving and problem finding.” — Richard Sennett, The Craftsman (2008), p. 9 per some sources (Wikiquote gives p. 18; check against the Yale edition). [verified: https://de.wikiquote.org/wiki/Richard_Sennett ; https://www.anecdote.com/2008/04/craftsmanship/] Why: Presents making as a loop between hand and head, not a plan carried out once.
“Whenever situation x arises, I will initiate the goal-directed response y!” — Peter M. Gollwitzer, “Implementation Intentions: Strong Effects of Simple Plans,” American Psychologist 54(7) (1999), p. 493 (abstract). [verified: https://www.prospectivepsych.org/sites/default/files/pictures/Gollwitzer_Implementation-intentions-1999.pdf] Why: Gives the literal form of the if-then plan the outline asks for, and makes a good inline example.
“Ever tried. Ever failed. No matter. Try again. Fail again. Fail better.” — Samuel Beckett, Worstward Ho (1983). [verified: https://en.wikiquote.org/wiki/Samuel_Beckett] Why: The line for cheap attempts. Use it with its bleak original context in mind, perhaps set against a stopping rule: knowing when not to try again.

Research

Core argument

Creating is the return leg. Chapters 11–14 follow the world into information: distinctions, models, tests. Creating runs the other way. A pattern held in a head (a recipe, a seating plan, an apology) gets pushed back into matter, and matter answers. That answer is the whole point: the world is where our representations get graded.
Intention alone is cheap and underspecified. Wanting something (Act I’s volition) doesn’t tell you where you stand, what’s available or what can’t give. Research on the intention-behaviour gap backs this up: people with “good” health intentions acted on them only about half the time (Sheeran 2002, as summarised in Gollwitzer and Sheeran). The chapter can start from that gap.
Map: take your bearings relative to a purpose, not the territory in full. A bearing is partial on purpose. It covers who the thing is for, what’s on hand, what the deadline is and what’s non-negotiable. A good map lists the constraints that enthusiasm would rather leave out. Simon’s satisficing fits here: we can’t optimise, so we define “good enough” up front. James C. Scott’s warning belongs here too. A map drawn from above that erases local metis can wreck the very thing it’s trying to improve.
Move: make contact before you’re certain, and make the first contact cheap. Planning can’t reveal what only contact reveals (Moltke, Suchman’s canoe, Schön’s “back-talk” of the situation). Cheap moves are pencil sketches, one spoonful of sauce, a glider instead of an aeroplane, one question instead of a speech. They buy information about the material at a low price. Nature does the same thing: bacteria run and tumble, slime moulds grow out everywhere and then prune.
Make: leave a result that can be inspected and revised. A made thing is sturdier than an intention because it can answer back. The rice is soft, the wing gives a third of the lift you calculated, the draft reads badly. Making isn’t the victory lap. It’s how you get consequences solid enough to learn from, which hands the result back to Learning (Ch. 14). Gall’s law and Brooks’s “plan to throw one away” both say the first working thing should be small and meant to be revised.
The three moves overlap, and each failure mode is missing one of them. A plan that never gets tried is Map without Move. Frantic effort with no bearings is Move without Map. Endless tinkering that never produces anything inspectable is Move without Make. Naming the absences is more useful than naming the stages.
If-then plans are a small handle where intention tends to slip. Implementation intentions (“If situation Y, then I will do Z”) move the decision from the moment of action to a cue chosen in advance. The meta-analytic effect is real (d = .65 across 94 studies) but it’s an average, and it depends on the cue being noticed and the action being right. It’s a handle, not a spell.
Stopping rules belong inside the plan. Without one, creating becomes a renovation that never ends, or the sunk-cost trap (Concorde). You set the stop in advance, just as you set the cue in advance: a timer, a budget, a “kill criterion”, a date to review. Optimal-stopping maths (the 37% rule) shows the question of when to stop can be reasoned about, not just felt.
Limits: some good work isn’t creating. Maintenance, care and replenishment preserve things rather than produce them. Some changes need resources or cooperation no individual can supply, and the framework should show that plainly rather than blame weak will. This leads into Becoming (Ch. 16) and Maintenance (Ch. 17).
Close on why this is “unnatural”. Every living thing reshapes its surroundings. Humans can hold a map of a change that doesn’t exist yet, test it cheaply in symbols, and then commit matter to it. That goes back to Act III’s theme: how sophisticated our loops are.

Key questions

If information “returns to matter,” what exactly crosses over when a recipe becomes a meal? Where does the pattern live at each step?
Is a beaver building a dam creating in this chapter’s sense? A bacterium swimming up a sugar gradient? Where does “trying” start?
Why is it so much easier to plan than to begin? What is the threshold made of?
How much map is enough? When does more planning become a way to put off contact?
What makes an attempt “cheap”? Cheap in money, time, pride or reversibility? And cheap for whom?
Can a first move be too cheap to teach you anything? (A model kitchen isn’t a kitchen.)
Why do finished-looking things resist revision? Is polish the enemy of learning?
Why do kindergarteners beat business-school graduates at the marshmallow challenge, and what does that say about adult planning?
When the material “talks back,” how do you tell a real constraint from your own lack of skill?
Does an if-then plan make you more free (acting on your own earlier choice) or less free (running on autopilot)?
Why do people keep going after it’s clear they should stop? Is stopping ever a failure of character, or always a failure to design the stop?
Who gets to draw the map when many people will live inside the result (Brasília, collectivised farms, a family’s shared kitchen)?
What’s lost when a plan from above overrides local knowledge (metis)? What’s gained?
Is “good enough” a compromise, or the only honest target a finite creature has?
A child builds a sandcastle knowing the tide will take it. Is that making or Becoming?
If most prototypes fail (Dyson’s 5,126), is failure part of the method, or is that just a story winners tell?
Can you “create” in relationships (an apology, a friendship) the way you create objects, or does the other person refuse to be material?
What does a stopping rule look like for something with no natural end: parenting, art, a career?
How does Creating hand off to Learning, and when should it?
The cynic’s question: isn’t all this just productivity advice in philosophical dress?

Examples

The dinner (running example from Ch. 13). You planned a mushroom pie and found half a bag of rice, two carrots, lentils and 40 minutes. Map is who’s eating and what’s allowed. Move is tasting the stock and texting about coconut milk. Make is dinner exists, and the rice is too soft.
Bacterial run-and-tumble (cell scale). E. coli swims straight, tumbles to a random new heading, and swims on. It runs longer when things are improving and tumbles sooner when they’re getting worse. That’s the cheapest possible move with a built-in stopping rule, and no map at all. Wikipedia, Chemotaxis
Slime mould and the Tokyo rail network. Tero et al. (Science, 2010) laid oat flakes in the pattern of Tokyo and 36 surrounding towns. Physarum polycephalum spread everywhere, then pruned itself into a network comparable to the real railway in efficiency, fault tolerance and cost. It found the shortest path in a maze the same way (Nakagaki et al., Nature, 2000). This is cheap attempts plus pruning, with no nervous system involved. Wikipedia
Betty the crow (animal scale). In 2002, Kacelnik’s Oxford group reported in Science that a New Caledonian crow bent a straight wire into a hook to lift a food basket. She did it in nine of ten successful retrievals, the first time she’d been given wire. The material answered and she reshaped it. Wikipedia
The Wright brothers, 1901–1903. The 1901 glider gave about a third of the calculated lift, and Wilbur said man wouldn’t fly “in a thousand years.” So they built a wind tunnel, tested about 200 model wings (38 analysed in detail), and found that the Smeaton coefficient everyone used was wrong (about 0.0033, not 0.0054). The 1902 glider flew over a thousand times. On 17 December 1903 they flew 120 ft in 12 s, and later 852 ft in 59 s. Gliders were the cheap attempts and the wind tunnel was the cheaper map. Wikipedia
Dyson’s 5,127 prototypes. Five years of prototypes before the G-Force launched in Japan in 1983. It works as a story about cheap iteration, and also as a warning about survivorship: we don’t hear about the people who built 5,000 and quit. Wikipedia
The marshmallow challenge. Twenty spaghetti sticks, tape, string and one marshmallow in 18 minutes. Tom Wujec (TED 2010) reported that kindergarteners outperform recent business-school graduates, because they build and test early while the graduates plan one tower and put the marshmallow on at the end. Verify the specifics against Wujec’s TED talk or marshmallowchallenge.com before citing.
Apollo 13’s “mailbox” (April 1970). The command module’s CO2 canisters were square and the lunar module’s sockets were round. Ground engineers designed an adapter using only what was on board (plastic covers torn from procedure manuals, duct tape and so on). Joseph Kerwin read the steps up over about an hour and Swigert and Haise built it. It’s the clearest case of intention meeting a fixed inventory. Wikipedia
Sydney Opera House. Estimated at $7 million for completion in 1963. It opened in 1973, scaled down, at $102 million. A cautionary tale about a grand map with no cheap first moves. Wikipedia, Planning fallacy
The planning fallacy in students. Buehler, Griffin and Ross (1994): 37 students predicted their theses would take 33.9 days on average (27.4 in the best case, 48.6 in the worst). The actual average was 55.5 days, longer even than their worst case, and only about 30% finished on time. Wikipedia
Edinburgh Trams (city scale). In 2004 the forecast was about £320m. A reference-class estimate put the 80th percentile at £400m. The final cost was £776m. The “outside view” as a mapping tool, and its limits. Wikipedia, Reference class forecasting
Scott’s high-modernist schemes (civilisation scale). Brasília, Soviet collectivisation and Tanzania’s ujamaa villagisation were grand maps imposed without cheap local moves, and they overrode metis (Scott, Seeing Like a State, 1998). This is the political version of “Map without Move.”
Concorde and the sunk-cost trap. The British and French governments kept funding Concorde after it was clearly uneconomic. It became the textbook case for continuing because of past investment, and the argument for writing the stopping rule down before you start. Wikipedia, Sunk cost
The secretary problem. When candidates arrive in sequence and you can’t go back, the optimal policy is to look at about 37% (n/e), then take the first one better than everything so far. This picks the best candidate about 37% of the time. A precise, surprising stopping rule; don’t import it naively into dating or hiring. Wikipedia
Implementation intentions in health. If-then plans helped with remembering vitamin pills, booking mammograms and doing testicular self-exams. Across 94 studies the effect was medium to large (d = .65) (Gollwitzer and Sheeran 2006, summarised in the NCI chapter linked under Source material). A small everyday version: “If I put the kettle on after dinner, I clear one section of the counter for five minutes.”
Thought experiment: the perfect apology. You spend three days drafting it. Compare that with one sincere sentence tomorrow and then listening. Which one creates something? Relationships are material that talks back more than anything else does.

Source material

“Plan to throw one away; you will, anyhow.” — Frederick P. Brooks Jr., The Mythical Man-Month, Anniversary ed. (1995), ch. 11 “Plan to Throw One Away,” p. 116 (first ed. 1975). The full line reads: “Hence plan to throw one away; you will, anyhow.” [verified: https://en.wikiquote.org/wiki/Fred_Brooks] Use: Makes the case for revisable results. The first made thing is a probe, not the product. (Brooks later softened this for incremental development; the author could mention that.)
“No plan of operations extends with any certainty beyond the first contact with the main hostile force.” — Helmuth von Moltke the Elder, “Über Strategie” (1871), in the standard English translation; German: “Kein Operationsplan reicht mit einiger Sicherheit über das erste Zusammentreffen mit der feindlichen Hauptmacht hinaus.” [verified: https://en.wikiquote.org/wiki/Helmuth_von_Moltke_the_Elder] Use: Motivates Move. Use the exact wording rather than the popular “no plan survives contact.” The real line says with any certainty, which is humbler and closer to the chapter’s point.
“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.” — John Gall, General Systemantics (1975), p. 71 (later Systemantics). [verified: https://en.wikipedia.org/wiki/John_Gall_(author)] Use: Supports starting with a small working Make. “Invariably” and “never” overreach, and the chapter can say so.
“Whereas economic man maximizes - selects the best alternative from among all those available to him, his cousin, administrative man, satisfices - looks for a course of action that is satisfactory or ‘good enough.’” — Herbert A. Simon, Administrative Behavior, 2nd ed. introduction (1957), p. xxix (Wikiquote attributes it to the book without separating editions; the word “satisfice” dates from 1956, so this sentence is from the 1957 edition, not the 1947 first edition). [verified: https://en.wikiquote.org/wiki/Herbert_A._Simon] Use: Grounds Map in “good enough” bearings and ties in with stopping rules. Satisficing is itself a stopping rule for search.
“Metis, far from being rigid and monolithic, is plastic, local, and divergent.” — James C. Scott, Seeing Like a State (1998), ch. 9 “Thin Simplifications and Practical Knowledge: Metis” (page not given in source; check the Yale ed., around pp. 309–341). [verified: https://en.wikiquote.org/wiki/James_C._Scott] Use: Shows why a map from above fails without local contact. Know-how in the hands answers the material in ways a plan can’t.
“Overall, forming implementation intentions had a medium-to-large effect on rates of goal attainment across the 94 studies included in the review (d = .65).” — Peter M. Gollwitzer and Paschal Sheeran, “Implementation Intentions,” NCI Health Behavior Constructs summary (c. 2008), summarising their meta-analysis in Advances in Experimental Social Psychology 38 (2006): 69–119. [verified: https://cancercontrol.cancer.gov/sites/default/files/2020-06/goal_intent_attain.pdf] Use: Gives the if-then section an honest number. Pair it with the same source’s finding that people act on “good” intentions only 53% of the time (Sheeran 2002).
“workmanship using any kind of technique or apparatus, in which the quality of the result is not predetermined, but depends on the judgment, dexterity and care which the maker exercises as he works” — David Pye, The Nature and Art of Workmanship (1968), p. 4, defining “the workmanship of risk.” [verified: https://en.wikipedia.org/wiki/David_Pye_(furniture_designer)] Use: Names what Make is when the result isn’t guaranteed. Pye sets it against the “workmanship of certainty” (jigs, machines), which is a nice way into when you want to take the risk out.
“Change the attitude toward errors. Think of an object’s user as attempting to do a task, getting there by imperfect approximations.” — Donald A. Norman, The Design of Everyday Things (2002 ed.), p. 131. [verified: https://en.wikiquote.org/wiki/Donald_Norman] Use: Recasts error as the normal texture of Move. Approaching a goal by imperfect approximations is exactly the cheap-attempt loop.
“In planning to run a series of rapids in a canoe, one is very likely to sit for a while above the falls and plan one’s descent.” — Lucy Suchman, Plans and Situated Actions (1987), p. 52 (commonly cited; the passage goes on to say the plan is useful for positioning, not for getting you through the rapids). [UNVERIFIED — wording from memory; the only check found was Wikipedia’s paraphrase at https://en.wikipedia.org/wiki/Situated_cognition. Check against the book before use.] Use: The best single image for “plans as resources, not scripts.” Map orients you, Move gets you through.

Counterarguments and limits

“This is productivity advice in a philosophy costume.” Map/Move/Make sounds a lot like lean startup, design thinking and OODA. The chapter needs to say what it adds: a place for these moves in a natural history of information going back into matter, plus its explicit limits (stopping rules, maintenance, Becoming). Otherwise the cynic wins.
Iterate-cheaply bias. Some things can’t be prototyped cheaply or undone: surgery, bridges, nuclear plants, a first impression, having a child. There, heavy up-front mapping is right (Flyvbjerg: “think slow, act fast”). The chapter shouldn’t treat planning as the villain. The Wright brothers’ wind tunnel was more map, not less.
Survivorship bias. Dyson, Edison and the Wright brothers are the iterators who won. Plenty of people iterated diligently and failed for lack of capital, luck or timing. Cheap attempts are only cheap if you can afford a lot of them, so the method assumes slack.
Effect sizes are averages. The d = .65 hides wide variation. Later work, including some replication debate and moderator findings, suggests effects shrink with weak goals, in the real world and over long periods. Don’t promise.
Gall’s law and Brooks overstate their case. “Never” and “invariably” are rhetorical. Some designed-from-scratch systems do work (the Apollo program had huge up-front design). Say “usually.”
People aren’t material. Using craft language for relationships (“make an apology,” “prototype a conversation”) risks turning the other person into an object. The earlier draft already flags this; keep it.
Constraints aren’t all physical. Poverty, law and other people’s refusal are constraints that no amount of bearings or cheap attempts can get round. The framework has to name structural limits, not recast them as bad technique (the earlier draft does this; strengthen it).
“Stopping rule” can rationalise quitting too early. Grit research and ordinary experience both say some things only pay off after a long stretch of apparent failure. The secretary-problem maths assumes conditions (no going back, known n) that rarely hold.
The nature analogies stretch. Bacteria and slime moulds don’t intend. Calling chemotaxis “cheap attempts” is a metaphor, and the chapter should mark where the analogy stops (volition, Act I).
Scott’s own critics say he underrates what legibility gave people (public health, sanitation). High modernism wasn’t only failure.

Connections

Ch. 13 (Three Ways to Meet a Problem): Keep the same dinner as the running example, and bring back the claim that the modes are overlapping tools, not a ladder.
Ch. 14 (Learning: Measure, Model, Manipulate): Make hands its results back to Measure. Manipulate in Learning is a test, while Move in Creating is an attempt with a purpose, and the author should draw that line clearly.
Ch. 16 (Becoming): The earlier draft’s final line (“usefulness loosens its grip”) sets up the handoff. Stopping rules make room for Becoming.
Ch. 17 (Maintenance, Meaning, and Magic): Washing the pot, replacing the soil, and the fact that preserving isn’t producing.
Ch. 18 (Loops That Learn and Loops That Don’t): Iteration can spiral or run away. The sunk-cost escalation is a loop that doesn’t learn.
Ch. 20 (Raising the Floor): Scott’s high modernism, and program theory as a Map of how collective Make is supposed to work.
Act I (volition chapters, 1–3): Creating is where the want from “Before Thought There Is Want” / “Will Creates Difference” finally touches matter.
Ch. 9 (The Cost of the Blade) and Ch. 10 (Thingification): A map is a cut, and bearings leave things out, so the cost of the blade applies to plans too.
Ch. 11 (Knowledge): Know-how (metis, craft) as a kind of knowledge that the ontology/epistemology map needs to include.

Exercise ideas

The two-towers test (Map vs Move). Instruction: Get 20 sticks of dry spaghetti, a metre of tape, a metre of string and one marshmallow, and set a timer for 18 minutes. First round: plan the whole tower on paper for 10 minutes, then build it, and put the marshmallow on at the end. Second round, another day: put the marshmallow on a tiny tower in the first two minutes, then grow it, re-testing every few minutes. Measure both. Notice: When did you first find out the marshmallow was heavy? What did round two tell you that the plan in round one couldn’t? Why it serves: You feel directly that the material holds information you only get through contact, and that a cheap early Make can beat an expensive map.
Write one if-then plan with a stopping rule, then audit it. Instruction: Pick one small change you’ve been putting off. Write it on a card in this form: “If [a specific cue that already happens daily], then I will [one action under 10 minutes]. I stop when [timer, quantity or condition]. On [a date 7 days out] I will check [one observable result].” Keep the card where the cue happens. On the review date, write down how many times the cue happened, how many times you acted, and whether the result held up. Notice: Did the cue fire without you noticing? Did the stop rule make starting easier? Was the action the right one? Why it serves: It turns both of the outline’s required tools into one testable artefact, and the audit brings Creating back round to Learning.
The cupboard inventory (bearings before ambition). Instruction: Before cooking your next meal (or starting a repair, a letter, any small project), write down in two minutes: who it’s for, what’s actually on hand, how much time you have, and one thing that can’t be compromised. Only then choose what to make. Afterwards, note one thing the result “said back” to you (too salty, the hinge still squeaks). Notice: How far did the plan move once the constraints were written down? What did the finished result reveal that the inventory didn’t? Why it serves: It practises Map as partial bearings tied to a purpose, and treats Make as something that answers back.

Open questions for the author

Keep “Map, Move, Make,” or rename? “Map” clashes with the book’s wider map/territory language (a map as a model in Ch. 14). “Bearings” might be clearer for the first M, if the alliteration can give.
How much biology? Should the chapter open at the cell (run-and-tumble, slime mould) to keep Act III’s natural-history thread, or stay human and domestic (the dinner)?
Where does Creating stop and Learning’s Manipulate begin? Both involve acting on the world. Pick a line, e.g. testing what’s true versus changing what is.
How hard to lean on implementation-intention research? Given replication concerns in social psychology, keep it as one modest paragraph, or build an exercise around it?
Stopping rules: a practical rule of thumb (timer, budget, kill criterion), the formal maths (secretary problem), or both? And should stopping be framed as part of Creating or as the doorway to Becoming?
Collective scale: include Scott’s high-modernism case here, or save it for Ch. 20 and only point to it?
Relationships as material: keep the apology and listening examples (they humanise the chapter) or drop them to avoid the “people as projects” problem?
Moltke: use the exact, more hedged translation (recommended) or the popular short version with a note?
Suchman’s canoe: verify the wording and page against the 1987 Cambridge edition before quoting. It may be the best image in the chapter.

Reader perspectives

Curious young child

First reactions:

“Creating is where an intention meets a cupboard” makes them laugh, because the cupboard never has what you want.
They love making things, so this feels like the chapter about them. They’d be disappointed it’s about lentils and not forts, slime or cardboard robots.
The if-then plan looks like a rule in a game, which they like: “If you touch the tree, you’re safe.”
The kettle-and-counter example is the most boring sentence in the chapter to a child.
“Stopping rule” is new and surprising. Nobody ever told them it’s OK to stop making something.

Questions they’d ask:

“What’s a map for if you’re not going anywhere?”
“Why do you have to plan? Can’t you just start building?”
“What if your first try is bad? Do you have to throw it away?”
“How do you know when something is finished?”
“If I stop before it’s done, isn’t that quitting? My coach says never quit.”
“What does ‘cheap attempt’ mean? Does it cost money?”
“Why did the guest not like coconut? Couldn’t you have asked first?”
“Is making a mess part of making?”
“Can you make something by accident?”
“What’s a plan that never happens? Is that still a plan?”
“Why does ‘if-then’ work? Isn’t it just saying what you’re gonna do?”

Where they’d get lost, bored, offended or unconvinced:

“Some situations require resources or cooperation that personal effort cannot supply” is true and important, but abstract. Say “you can’t build a treehouse if you don’t have a tree, or if a grown-up won’t hold the ladder.”
“Another person is not a project to optimize” goes right over their heads; the listening example is about adult arguments.
They’d be unconvinced by “stop when the timer sounds” if the thing they’re making is fun. The chapter should admit that stopping rules matter most when you want to keep going (one more level, one more coat of glitter).
The Beckett quote “fail better” would sound like a poster in their classroom, but if the author uses its bleak original context, a kid would be confused why a sad line is being used cheerfully.
James Scott and Seeing Like a State would never land at this age without a picture. The picture could be: a grown-up designs a playground from an office, and kids still wear their own shortcut path across the grass.

Examples they’d bring:

A cardboard box fort: Map (how many boxes, where’s the tape, where can it go without Mum tripping), Move (tape two boxes and see if they stand), Make (it exists, and then the roof falls in and you fix it).
Sandcastles: The sea is the ultimate material constraint. You learn wet sand holds and dry sand doesn’t only by trying. And you have to decide when to stop before the tide decides for you.
Drawing with a pencil first, then pen: a kid’s own version of a cheap attempt.
Birthday card for Grandma: If-then plan in kid form: “If I finish my homework, then I’ll make the card before dinner.” And the stopping rule: it’s done when it’s done, not when it’s perfect.
Desire paths in the park: A well-worn shortcut across the grass is a real, visible example of people’s moves overruling a planner’s map.
Slime recipes: Too sticky, add more activator; too hard, add lotion. Making that talks back.

What would win them over:

Swap at least one grown-up example (the kettle and counter) for a kid-sized one, like a fort, a card or a sandcastle, so they can see themselves in it.
Tell them the first try is supposed to be a bit bad; that’s what makes it cheap. It takes the fear out of starting.
Frame stopping rules as “knowing when your sandcastle is done so you can enjoy it before the wave comes.”
Show the made thing talking back: the fort roof falls in, and that’s how you learn about tape.

Cynical adult

First reactions:

“If-then plans and stopping rules. So this is the productivity chapter. I have read Atomic Habits. I have a drawer of unused habit trackers.”
The counter-clearing example (“If I put the kettle on after dinner, I will clear one section of the counter for five minutes”) is so small it’s almost insulting. Also, I don’t own a kettle.
To its credit, the draft keeps undercutting itself (“neither a spell nor a substitute for care”, “another person is not a project to optimize”). That’s more honest than most self-help. But too much hedging and I wonder why it’s giving advice at all.
The line “Creating without enough can turn every life into an unfinished renovation” is the best sentence in the chapter.
“Information returns to matter” (from the brief) is a big idea that the draft never actually shows. The mushroom pie becomes lentil rice; that’s not a philosophy of making, that’s dinner.

Questions they’d ask:

“How is Map, Move, Make different from ‘plan, try, finish,’ and why rename it?”
“Do if-then plans actually work, or is this another shiny finding that won’t replicate? What’s the effect size?”
“You cite Gollwitzer, who invented implementation intentions. Isn’t that like quoting the inventor on how good the invention is?”
“What’s a stopping rule for things that don’t have natural endings, like a career, a relationship, or a startup?”
“Isn’t ‘cheap attempts’ just ‘fail fast’, the Silicon Valley slogan that ran companies into the ground and people into burnout?”
“Who gets to make cheap attempts? Some people can’t afford to fail even once.”
“What’s the point of Beckett’s ‘fail better’ if the original is bleak? Are we being sold an inspirational poster by someone who knows it’s a misreading?”
“Where does sunk cost fit? The chapter should tell me when to walk away, not just when to stop the timer.”
“What does Seeing Like a State have to do with my counter? The chapter mentions constraints lost by abstract plans but never shows me one.”
“Is maintenance creation or not? You mention washing the pot, then wave it off to Chapter 17.”

Where they’d get lost, bored, offended or unconvinced:

The exercise is too trivial to be persuasive. A five-minute counter-clearing habit does nothing to show “intention meeting material constraint.”
The if-then example about listening (“If I notice myself preparing a rebuttal, then I will first summarize what I heard”) is good, but it reads like a couples therapist’s handout, and the chapter never checks whether it worked.
The brief says the mode is “where information returns to matter and changes the world,” but the chapter is personal-scale only. No buildings, no products, no policies. The biggest claim is left unsupported.
The class issue: cheap attempts and stopping rules assume slack. The draft nods to this in the last paragraph (“resources or cooperation that personal effort cannot supply”) but it arrives late and feels like a disclaimer.
Sennett’s page number is marked unverified; a chapter that tells me to measure carefully shouldn’t cite carelessly.

Examples they’d bring:

Gollwitzer & Sheeran’s 2006 meta-analysis reported a medium-to-large average effect (d around 0.65) across 94 studies. Say the number, and mention that later work on publication bias has questioned how large effects in this literature really are. Honesty about the size sells it better than hype.
The Sydney Opera House: Jørn Utzon’s design (1957) collided with engineering reality; it opened in 1973, ten years late and famously over budget. Intention meeting material at civic scale, and a case where no stopping rule was applied.
Brasília or the collectivized agriculture cases in James C. Scott’s Seeing Like a State: master plans that ignored local knowledge. The book already cites Scott; use him.
A home renovation that never ends: the reader has lived this or watched a friend live it. That’s the “unfinished renovation” line made literal.
The Concorde: kept going for decades partly because of what had already been spent. It gave economists the “Concorde fallacy” as a name for sunk-cost reasoning. A perfect stopping-rule cautionary tale.

What would win them over:

One example at world scale (a building, a product, a law) next to the kitchen, so “information returns to matter” means something.
Say how well if-then plans work and where they don’t, with a real number, and keep the stopping rule central instead of an afterthought.
Acknowledge up front who can afford cheap attempts, and offer a version for people with little slack.
Replace the counter exercise with something where the material actually pushes back (fixing a thing, writing a letter that gets a reply).

Believer / spiritual reader

First reactions:

Every Abrahamic reader will think of Genesis at the word “Creating.” The first model of making they know ends with a stopping rule: on the seventh day God rested (Gen. 2:2–3). The draft’s “Include the stopping rule” will land as unintentionally scriptural.
They’ll warm to the line “another person is not a project to optimize.” It’s the most spiritually serious sentence in the draft.
They’d appreciate “Some worthwhile activity preserves rather than produces.” Tending, keeping and stewardship are central to many traditions (Gen. 2:15 gives the gardener’s task as “to work it and to keep it”).
If-then plans will strike them as a secular version of something old. Religious life is full of cue-based practices: a blessing on seeing lightning, prayer at the call to prayer, grace before bread.

Questions they’d ask:

Are we creators, or sub-creators working with material we didn’t make? Does the difference change how we should make?
Where is the Sabbath in this chapter? A weekly stopping rule has been tested on a whole people for three thousand years. Isn’t that the strongest example you could use?
If-then plans: “If I say I’ll do it tomorrow, then I’ll add ‘if God wills.’” Isn’t that humility about outcomes also a kind of implementation intention?
You say making isn’t a victory but a way of “acquiring consequences.” What about the fruits of action? Should we be attached to them at all?
Who is the cupboard’s owner? Your “Map” lists what’s available. Traditions would add what was given, and ask whether you’re entitled to use it all.
Is a stopping rule just efficiency, or is it also a boundary on our own ambition, a guard against the tower of Babel?
How do I know whether a failed attempt means “try again” or “this isn’t mine to make”? Discernment traditions have criteria. Does the book?
Is the apology example really Creating? Repentance is a change of heart, not a product. Doesn’t calling it “making” miss something?
Should maintenance (washing the pot, tending a friendship) be called making, or does it need its own sacred standing?
Does the book know Beckett’s “fail better” comes from a bleak late text? A believer might prefer a hope-bearing version, and would want the difference noted.

Where they’d get lost, bored, offended or unconvinced:

Unconvinced: by the if-then research presented as the main source of wisdom about turning intention into action. The science is fine and the caution is good (“neither a spell nor a substitute for care”). But without religious precedents it looks as if the book thinks the idea began with Gollwitzer.
Mildly offended: by treating the apology as a craft object with a draft and a result. Many traditions separate the work of repair (restitution, teshuvah) from the inner turning. Name both, or the reader will feel the heart has been left out.
Bored: by a kettle-and-counter example for the stopping rule. It’s too small to show why stopping matters. Stakes need to come from somewhere, such as a relationship, a vocation or a building project.
Lost: by the word “Map” arriving in this chapter after “Model” in the last. A reader will ask how bearings differ from a model. Luke 14:28, counting the cost before building a tower, gives a concrete answer: a map is the inventory for this purpose.

Examples they’d bring:

The Sabbath (Genesis 2:2–3; Exodus 20:8–11) and Heschel’s The Sabbath (1951). Heschel calls the Sabbath “a palace in time.” A weekly stopping rule for all creating, built into law, is the best-tested stopping rule in human history.
Shmita, the sabbatical year (Leviticus 25:1–7). The land lies fallow every seventh year. The draft’s “replacing soil after harvest” is already a religious law.
Qur’an 18:23–24. “Do not say of anything, ‘I will do that tomorrow,’ except [with] ‘If God wills.’” It’s an if-then plan with humility about outcomes built in, and it matches the draft’s caveat that the cue may never arrive.
Bhagavad Gita 2:47. “You have a right to your action, but never to its fruits.” It’s the classic teaching on making without being ruled by results. It reframes stopping rules as detachment rather than efficiency.
Luke 14:28–30. Counting the cost before building a tower so you aren’t mocked for leaving it half-finished. It’s a scriptural “Map” step, and a warning about “unfinished renovation.”
Shikinen Sengu at the Ise Grand Shrine. The main shrine buildings are rebuilt every twenty years, and have been for well over a millennium. Maintenance, making and the passing on of craft become one ritual, a strong case for “preserves rather than produces.”

What would win them over:

Using the Sabbath, or Gita 2:47, as the book’s example of a stopping rule, alongside the modern research rather than beneath it.
A sentence acknowledging that people with faith often experience making as participation in something larger (sub-creation, stewardship), and that the framework doesn’t need to deny this.
Separating repair from repentance in the apology example.
Keeping “the framework should reveal those conditions, not translate them into insufficient will.” Believers who have watched prosperity preaching blame the poor for weak faith will cheer.

Skeptical scientist

First reactions:

The cupboard scene is concrete and relatable, and “Creating is where an intention meets a cupboard” is the chapter in one line. The planning fallacy is sitting right there in the gap between the mushroom pie and the forty minutes, unnamed.
The if-then paragraph is carefully hedged, and the hedges are right. But it cites “research on implementation intentions” without giving an effect size. A scientist wants to know how big the effect is and how robust.
“Stopping rule” is doing double duty. In statistics, stopping rules are a known source of false positives; in decision theory, optimal stopping is a real mathematical field. The chapter uses the phrase in an everyday sense without acknowledging either, and a technical reader will notice.
The brief’s “information returns to matter” is the most interesting idea and gets the least development. Evolutionary biology and motor neuroscience both have precise versions of it.

Questions they’d ask:

How big is the implementation-intentions effect, in what populations, for what kinds of goals, and has it survived correction for publication bias?
What’s the evidence that cheap early attempts beat careful planning? It depends on the cost of failure. Surgery and bridges don’t prototype the way cooking does. Where’s the line?
How do you set a stopping rule in advance without it being arbitrary? What’s the principled version?
Isn’t Map just Chapter 14’s Model under a new name? What distinguishes a map for acting from a model for explaining?
Does creating need intention? Evolution builds eyes with no plan. If intention is essential, say so; if not, say what natural “creating” looks like.
Other animals build things: beavers, bowerbirds, New Caledonian crows bending hooks. What exactly is unnatural about human making?
The kettle example has four components (cue, action, limit, review). Which of them does the evidence actually support?
Why does the draft end on maintenance and structural constraints? That’s important, but is it this chapter’s job or Chapter 17’s?
How do sunk costs fit? The hardest stopping decisions are the ones where you’ve already invested.

Where they’d get lost, bored, offended or unconvinced:

Unconvinced: “Motion reveals features that planning alone cannot” is true in cheap, reversible domains and dangerous in others. Without a condition like “when attempts are cheap and reversible,” it reads as a startup motto.
Bored: the kettle-and-counter example is so low-stakes that it trivialises the idea. A reader who wants to change something real (a research project, a career) gets nothing to hold.
Lost: “another person is not a project to optimize” in the middle of the if-then section changes topic abruptly. It’s a real point but belongs elsewhere.
Offended (mildly): “Creating without enough can turn every life into an unfinished renovation” is a good line, but it moralises slightly. Frame stopping as information management, not virtue.
Loose terms: “map” (a metaphor on top of Chapter 6’s map/territory language, so is it the same map?), “constraint,” “result,” “stopping rule.”

Examples they’d bring:

The Wright brothers’ wind tunnel (1901). Their gliders underperformed the lift predicted from Lilienthal’s tables. Rather than keep building gliders, they built a small wind tunnel and tested some 200 wing shapes cheaply, correcting the published data. It’s the best real case of Map, Move, Make: a map that turned out wrong, cheap attempts that fixed it, and a revisable result.
Evolution as tinkering. François Jacob, “Evolution and Tinkering,” Science (1977). Nature makes by modifying what’s to hand, not by design. It’s the right foil for asking what intention adds, and it echoes “half a bag of rice and two carrots.”
Niche construction. Odling-Smee, Laland and Feldman, Niche Construction: The Neglected Process in Evolution (2003). Organisms modify their environments (beaver dams, earthworm soil), and those changes feed back on selection. It’s the biological version of “information returns to matter,” and the baseline for the unnatural claim.
Tool-making in birds. Weir, Chappell and Kacelnik, “Shaping of hooks in New Caledonian crows,” Science (2002). A crow named Betty bent straight wire into a hook to retrieve food. It shows how far non-human creating goes, and what it lacks (cumulative, taught, symbolic design).
Stopping rules done properly. Optional stopping inflates false-positive rates (Simmons, Nelson and Simonsohn, Psychological Science, 2011), which is why trials pre-specify when they’ll stop. Wald’s Sequential Analysis (1947) gave the principled version. Simon’s “satisficing” (Simon, “Rational choice and the structure of the environment,” Psychological Review, 1956) is the everyday one: stop at good enough.
Planning and sunk costs. Buehler, Griffin and Ross, Journal of Personality and Social Psychology (1994), on the planning fallacy. People underestimate their own completion times even knowing past projects overran. Arkes and Blumer, Organizational Behavior and Human Decision Processes (1985), on the sunk cost effect. Between them they explain why the mushroom pie was planned and why the renovation never ends.

What would win them over:

Quantify the if-then evidence honestly: Gollwitzer and Sheeran’s 2006 meta-analysis reports a medium-to-large average effect, and the chapter should note that effects vary by goal type and that meta-analytic estimates tend to shrink when publication bias is taken into account.
State when cheap attempts are the right strategy (low cost, reversible, fast feedback) and when they aren’t, with one real case of each.
Put the Wright brothers or a comparable case at the centre, so the chapter shows creating at a scale that matters, not only in a kitchen.
A clear distinction between Map (for acting) and Model (for explaining), or an admission that they are the same skill aimed differently.

Earlier draft

From “Creating: Map, Move, Make”

The dairy-free main course begins as a grand idea: perhaps a complicated mushroom pie. Then you find half a bag of rice, two carrots, lentils, and forty minutes.

Creating is where an intention meets a cupboard.

Map means finding your bearings in relation to a purpose. You need to know who will eat, what is available, how long it takes, and what cannot be compromised. This is not a complete model of cooking. It is a selective account of the terrain relevant now. A useful map includes constraints that enthusiasm would prefer to omit.

Move means making contact before certainty arrives. You rinse the lentils, taste the stock, and text the guest to check whether coconut milk works. The first move can be deliberately cheap: sketch the seating plan in pencil; cook one spoonful of the sauce; ask one sincere question before composing the perfect apology. Motion reveals features that planning alone cannot.

Make means leaving a change in the world. Dinner exists. A draft can be read. A repaired hinge holds the door. The result also pushes back. The rice is too soft, the guest dislikes coconut, the apology explains when it should listen. Making is not the victorious final stage. It is a way of acquiring consequences sturdy enough to answer us.

The three moves overlap. Skilled makers observe and explain constantly. They also receive accidents. The names are useful because they expose characteristic absences: a plan with no attempt, frantic effort with no bearings, or endless motion that never becomes an inspectable result.

Vague intentions often fail at the threshold. I should listen better does not specify when or how. Research on implementation intentions suggests that an if-then plan can help in some settings: If I notice myself preparing a rebuttal, then I will first summarize what I heard. This is neither a spell nor a substitute for care. The cue may never be noticed, the action may be wrong for the moment, and another person is not a project to optimize. It simply places a small handle where intention often slips.

Choose one low-stakes change for the coming week. Give it a cue, one manageable action, a limit, and a date for review. For example: If I put the kettle on after dinner, I will clear one section of the counter for five minutes; I stop when the timer sounds; on Sunday I will check whether the space stayed usable. Include the stopping rule. Creating without enough can turn every life into an unfinished renovation.

Some worthwhile activity preserves rather than produces: washing the pot, maintaining a friendship, replacing soil after harvest. Some situations require resources or cooperation that personal effort cannot supply. The framework should reveal those conditions, not translate them into insufficient will.

And sometimes the best evening contains no improvement project at all. The third mode begins when usefulness loosens its grip.