Haute Lumière

Commerce · I.08 · MMXXVI · daylight

La Bourse  /  Volume I  /  Nº I.08  /  Workbook — the student

WORKBOOK — THE STUDENT

Chapter I.08 · When It Stalls

For the person studying this alone, or in a seminar, with no organisation to change yet. You have stalled on things. Everybody has. What you will build in these ninety days is the instrument that makes your next stall readable — and the habit of reading a stall is worth more over a lifetime than the habit of avoiding one, because you will not avoid one.

---

WHY THIS WORKBOOK IS DIFFERENT

The chapter is written for someone whose programme has stopped inside a firm with a facility, a sponsor and an approval path. You may have none of those. You have something better for learning this particular lesson: you have a much shorter cycle time and many more stalls.

A dissertation chapter abandoned in March. A training plan that survived eleven weeks. A language you have restarted four times. A society you founded that ran for two terms and then quietly did not. A practice, a habit, a project, a band.

Every one of those is a completed natural experiment, and you have never read one of them. That is the entire opportunity of this workbook. You are sitting on a dataset about what you can actually build that runs without you, and it has been free the whole time.

The method is identical to the chapter's. Declare the pause. Take the survivor ratio. Compute the warm fraction. Name the room. Then restart at the survivors or close it in writing — and the closing is not the consolation prize, it is half the skill.

---

PART ONE — DISCOVERY

Days 1–30: read the stalls you already have

Exercise 1.1 — The stall inventory (90 minutes)

List every project, practice, habit, commitment or piece of work you have started in the last five years and are not currently doing. Aim for fifteen. Include the small ones and the embarrassing ones; especially those.

For each, in one line: what it was, roughly when it stopped, and — this is the one that matters — what the last thing that happened was.

Almost none of them will have an ending. Most will have a last Tuesday.

What you are looking for. Two patterns, and you will see both.

First, the shape of your stalls. Do they stop at the same point? A great many people stall at exactly the transition from building the thing to maintaining it, which is a design problem and not a character problem. Some stall when a single other person leaves. Some stall at the first external deadline that was not self-imposed. Your pattern is information about what you need to build into the next one, and it is not available from any source other than a list like this.

Second, the outliers. One or two things on your list did not stop. Do not put them on the list — put them on a second list, and treat that second list as the most valuable page in this workbook.

Exercise 1.2 — The survivor register, applied to yourself (2 hours)

Pick the largest thing on the stall inventory. Reconstruct, as honestly as you can, the full list of activities that were running at the point it stopped. Be granular: not "the dissertation" but the reading log, the weekly writing block, the supervision meeting, the reference database, the Sunday review, the shared document with the two people in your cohort.

Then mark each one: is it still happening, with nobody pushing it?

Count. That fraction is your σ.

Almost certainly it is low, and almost certainly two or three things are still alive that you had not noticed. The reference database that you still add to. The friend you still send papers to. Those are the only parts of that project you actually built. Everything else was being carried by your attention, and when your attention went elsewhere it went with it.

Exercise 1.3 — The appreciative interview about a stall (45 minutes, with another person)

Find someone whose work you admire and ask exactly this:

"Tell me about something you started that stopped — and then came back. What was still there when you came back to it? What had you built that survived without you?"

Then stay quiet. Take notes on what survived, not on why it stopped.

This is the version of Chapter I.01's appreciative interview aimed at the hardest material, and the answers are startlingly consistent. People restart on a relationship, a file, a physical object, a standing arrangement, or a skill in their hands. Almost nobody restarts on motivation. Do this ten times and you will have a working theory of what durability is made of that no course will give you.

---

PART TWO — THE ARITHMETIC

Days 31–45: compute before you decide

Exercise 2.1 — Reproduce the chapter's figures (2 hours)

Do not take the numbers on trust. Open lib/verify/I_08.py, read it, and then compute these independently — by hand, in a spreadsheet, or in whatever language you use.

1. W(t) at t = 1, 2, 3 and 5, from the five components and their rates. Confirm 0.823, 0.681, 0.567, 0.400.
2. The instantaneous decay rate at t = 0, which is −Σ wᵢ ln(kᵢ). Confirm 19.9 percent a year. Notice that it is larger than the 17.8 percent actually lost in year one, and be able to say why.
3. The carrying-cost schedule on a £180,000 build with £62,000 of annual saving. Confirm that cumulative cost crosses the build cost at about month 24.
4. The three restart windows at f = 0.15, 0.35 and 0.60. Confirm 11.3, 5.8 and 2.7 years.
5. The survivor-ratio table. Confirm that σ = 0.25 takes the f = 0.60 window from 2.7 years to 7.5.

Exercise 2.2 — Build your own W (2 hours)

The chapter's weights and four of its five decay rates are assumptions, and the module says so in its own comments. This exercise is where you stop being a reader of a model and start being an author of one.

Take the project from Exercise 1.2 and rebuild the five components for it:

Assign your own weights (they must sum to 1) and your own annual survival rates. Justify each rate in one sentence. Then compute W at the actual elapsed time.

The one that will surprise you is your own fluency. Most people rate it far too high. A dissertation chapter you have not touched in two years requires substantially more than two weeks to get back into, and that re-entry cost is the decayed component. Measure it next time by timing the re-entry.

Exercise 2.3 — Estimate your own f (45 minutes)

f is the premium you pay for the fact that this stopped. In an organisation it is counted in signatures. In a life it is counted in the people who would have to be asked again and in your own record with yourself.

Write down, for the same project:

• Who would have to agree for this to resume? List them by name.
• Which of them watched it stop?
• What would you have to say to each one that you did not have to say the first time?
• And honestly: what is your own internal f? How much harder is it to begin something for the second time than the first?

That last question is not a soft one. It has an answer and you know it. Write a number between 0 and 1.

Exercise 2.4 — Run the inequality (30 minutes)

You now have W, f and σ for a real project of yours. Compute:

        restart while    W  >  1 + f  −  1/(1 − σ)

Write the verdict down and date it. Then do what it says. Both outcomes are wins: a restart at σ is a cheap restart of something you wanted, and a close is the discharge of a debt you have been carrying without noticing.

---

PART THREE — DESIGN

Days 46–75: build the instrument that survives you

Exercise 3.1 — The personal dormancy note (one page, 40 minutes)

For every project you currently have running, write a one-page note now, while it is running, that says what happens if you stop.

• What are the four things that must keep running even if I stop pushing?
• Who is the second person, and what do they hold?
• What is the sunset date, after which this is ended rather than dormant?
• What is the survivor register I would take at month three?

This is thirty minutes per project and it is the single most useful piece of paper in this workbook, because the version of you who needs it will not be in a state to write it.

Exercise 3.2 — Embed one thing (a fortnight)

Take your current main project and find one activity inside it that could be made to run without your attention. Not all of it — one activity.

Candidates: a recurring calendar item somebody else owns; a shared document another person also writes in; an automated export; a standing arrangement with one other human ("we send each other a paragraph every Sunday"); a physical object placed where it is used.

The test is Chapter I.08's test: would this still happen if you stopped? Build it so the answer is yes. Then, if you ever stall, your σ is not zero, and the arithmetic in Part Two showed you exactly what a non-zero σ is worth.

Exercise 3.3 — Write one close (90 minutes)

Choose something from the stall inventory that is genuinely over. Write its close. Four parts, one page:

1. What it cost — hours, money, the things you did not do instead.
2. What it returned while it ran. Be generous and be accurate; almost everything returned something.
3. Its final survivor ratio, and what survived.
4. One paragraph: which layer proved load-bearing.

Then put it somewhere you will find it. This is not journalling. It is the document your next project will be designed from, and it is the form of writing this book values most.

---

PART FOUR — THE TERM PROJECT

Days 76–90 and beyond: the stall ledger

The project. Keep a stall ledger for a full academic year.

One page per stalled thing, and a new page whenever something stops. Each page: the thing, the date of the last Tuesday, the survivor register with σ, your W at the time of writing, your f, the verdict of the inequality, and the decision you took. Then, six months later, a single added line: was the verdict right?

What this produces. By the end of a year you will have between six and fifteen pages, and three things that are genuinely rare:

1. Your own decay rates, measured. After a year you will know how fast your fluency, your records and your collaborations actually decay, because you will have re-entered several of them and timed it. You can then replace the chapter's assumptions with numbers that are true about you.

1. Your own f, calibrated. The gap between the f you estimate at the moment of stalling and the f you observe at the moment of restarting is a measurement of your own optimism, and it is stable enough to correct for.

1. A design principle you did not read anywhere. The pattern across your own survivors — what it is, in your life, that keeps running — is different from the pattern in mine, and it is the most actionable thing you will own at the end of this year.

Self-assessment, at the end of the term. Score each honestly, one to five.

The one to be honest about is the last row, because it is the only one that changes σ next time, and σ is what buys the restart.

---

A NOTE ON WHAT THIS IS ACTUALLY FOR

You are going to stall. Repeatedly, for the whole of a working life, on things that mattered. This is not a flaw in your temperament and this workbook is not a productivity system for preventing it.

What separates people whose work compounds from people whose work does not is almost never the frequency of their stalls. It is whether anything survived them — and that is a question of what got embedded, which is a design decision you make while things are still moving.

So the exercise that matters most here is 3.2, and it takes a fortnight. One activity, in your current project, built so that it would keep happening if you stopped. Do that once per project for ten years and you will arrive at mid-career with a substrate underneath you: records that stayed comparable, collaborators who are still collaborators, methods written down, instruments still running.

That substrate is the whole asset. Everything else in this chapter is arithmetic about how fast it decays when you neglect it — and the arithmetic is only interesting because the asset is real.