Personal Knowledge Mgmt Agents -- A 52-Week arXiv Curriculum In Building A Better Brain

My intention now is more intently upon planning the year ahead IN THINKING.

It is more of an evolution, than a revolution ... it will ALWAYS be true that I am looking for better knowledge engineering tools; that always was true, but even moreso no ... I will ALWAYS be working today on making tomorrow better, working tomorrow on making the week ahead better, working in the week ahead on making the next month better ... with a serious intention of making the next 52 weeks into the best year in the history of 52 week periods.

Since I don't have that many years or maybe days or hours left in life, I need to be more blunt about my time management and prioritization.

There is NO TIME WHATSOEVER for nostalgia, reunions, funerals or any form of looking back.

I will REPEAT for emphasis, if anyone was not paying attention.

There is NO TIME WHATSOEVER for nostalgia, reunions, funerals or any form of looking back.

Other people can do things in their own way, but in my case ... I am going to allow the DEAD to bury their dead ... and I am absolutely dead serious about that -- from now on, there is NO TIME WHATSOEVER for nostalgia, reunions, funerals or any form of looking back. NONE!

Planning the system, then living and improving the system.

Plans are always imperfect, good intentions pave the road to hell ... but the whole point of planning is not the document, but SUCK that comes from when plans don't pan out -- embracing the SUCK means finding and embracing the root cause of the failure and caving in the melon of that demon.

Planning is not navel gazing, but instead 99% of it is the process of implementing, then failing ... then refactoring, revising and changing the plan ... the POINT of a plan is to have a starting roadmap to work from, while realizing that as soon as the work starts, the plan will need to be completely overhauled and adjusted, because new OPPORTUNITIES will arise.

Finding the root cause of future failures is why we LOOK FORWARD ... we plan to be able to accomplish early detection, we plan to EMBRACE the neck of root cause demon and cave its melon in ... in terms of planning for the FIGHT AHEAD, we have to PLAN for the likelihood of being punched in the face; we have to plan for desparately wanting to quit.

The MOST IMPORTANT part of planning is gratitude discipline which is ALWAYS, ALWAYS, ALWAYS realizing that the situation is far better than one hoped for, although almost certainly not exactly what one hoped for, ie the plan will NEED TO BE adjusted -- that's the REASON for why planning and preparation are key.

A Second-Brain Curriculum for One Ordinary Working Person

Free. Self-led. No tool required. No branding. No system to buy.

Soli Deo gloria.

0. Why bother

You are going to be programmed either way.

That is not a rhetorical flourish. It is a description of how a person is made. What you read this week, what you scroll, whose voice you let into the room and how often — that becomes the material you think with, and it shapes your attention long before you have any say about the shape. There is no neutral setting. The only real question is whether you are the one doing it on purpose.

This document is about doing it on purpose. It is long, it is dull in places, and it asks for an hour a week for a year. Here is the case for the hour.

The short answer

A second brain is not a filing cabinet. It is a harness — the structure your thinking runs inside. And a badly built harness does not merely slow you down. It quietly determines what you are able to think, the way a badly laid out workshop determines what can be built in it, and a badly chosen vocabulary determines what can be said.

Everything else in these pages is detail. This is the claim.

This is not new, and the last time it happened it changed everything

There is a phrase, the People of the Book, and it was coined by outsiders as a category — but it stuck because it was accurate. Jews and Christians were the people whose God had written something down, and who were expected to read it.

A community organized around a text behaves differently from a community organized around memory and custom. It can argue with itself across centuries. It can compare readings at a distance. It can hold a standard that is not the person currently in charge. You cannot have "it is written" without a place to write it, and you cannot have a reformation without people who own copies.

Two further things came out of that soil, and both are load-bearing here.

The first is that every human bears the image of God. Not the king, not the priest, not the literate — every human. That claim installs moral agency and individual conscience at the centre of a person, and it implies that an ordinary person's judgment about what is true is worth something. Historians still argue about how much of Western political thought descends from that claim versus arriving independently, and this document does not need to settle it. What it needs is the smaller, plainly observable fact: people who can read, and who know they are responsible for what they read, become a different kind of people. As do their children.

The second is that print changed the speed at which all of this could happen. Scattered communities could finally compare the same page instead of comparing memories. Argument became cumulative. Access stopped being a privilege of the few.

The important part is not that literacy made people cleverer. It is that literacy made a particular kind of thinking possible for ordinary people: holding many sources at once, in one head, over time, and then putting them in a new order. That is the entire ambition of a second brain. The printing press is its ancestor, and the family resemblance is not a metaphor.

What is actually NEW now ... with agentic harness engineering

Two things, and they are why this is worth doing now rather than way back in 2026 or decades before.

Your notes become much MORE a part of your future mind, not a record of its past. That is not mysticism; it is a mundane observation about how anyone does arithmetic, keeps an appointment, or thinks through a decision on paper. The notebook is in the loop. Good notetaking skills always mattered for actually dwelling on different facets and actually LEARNING material as one did more problems or exposed weaknesses in understanding, but now iterative improvements that are part of good notetaking or improving agent skills [after failures] are even more important. If the notebook is disordered, the loop is disordered, and you will experience the disorder as a personal failing rather than as a property of the harness. As we move into the realm of crafting better and better agentic skills and the trajectories of building knowledge graphs, the process of notetaking and ruminating on those notes becomes even more intensely part of our consciousness, ie we must really THINK about how we want our agents to think as they gather information and perform tasks that shape our human connections and drive our future interactions.

And the reader is no longer only you. Whatever you write down, and however you structure it, is now also the material that your tools reason from ... and that material can be cloned and used by others, in the way that Paul's letters not only were for Paul to clarify his ideas, but they also shaped the ideas of the revolutionary, evangelical body of believers in Jesus Christ. Assistants, search, agents, automations — all of them inherit your structure and your omissions. A vault with no stated purpose produces confident answers to the wrong question, at scale, quickly. A vault with an explicit purpose and a clean set of containers is something closer to an executable specification: not just a place you remember things, but a briefing you can hand to something that will work while you sleep.

That is what people mean when they talk about harness engineering for agentic systems. The interesting part is not the agents. It is the harness — and the harness is your second brain, which is why it is worth a year rather than an afternoon.

The Big Why has to be written down, or it is a mood

Ask why five times and you find out what you actually want. Most chains stop early. Why organize my notes? So I can find things. Why? So I can finish work. Why? So I can do my job. Why? — and here almost everyone runs out of words, because the fourth answer was the last one they had ever said out loud.

A purpose that survives five why's is a purpose. A purpose that stops at so I can be more productive is a feeling with a calendar attached.

And your notes are where the answer becomes checkable. Not because the notes are a spiritual instrument, but because they are a truthful record of what you actually attended to. You can say what you love. Your last three months of captures will tell me what you loved, and the two accounts will not always agree. That gap is the most useful thing in this entire document, and the whole year is a way of closing it.

So the Big Why here is not an abstraction and it is not a productivity goal. It is stated plainly, once: to adore and imitate Jesus Christ — not a figure in a stained-glass window, but a person who lived in a world with bodies in it, interruptions in it, Roman roads in it, and neighbors in it who were frequently inconvenient. Everything below is an attempt to be more useful to that. Ordering what you know is not the point; it is the preparation for the point.

This also rules out a counterfeit that is very popular. Compassion that only distributes is not compassion. Handing somebody calories, or a check, or an opinion, or a link, and calling it love, is usually a transaction that makes the giver feel finished. Love that actually helps requires knowledge — of the person, of the problem, of what has already been tried and already failed. That knowledge has to be collected, sorted, remembered, and kept current. There is no shortcut, and the people who skip the collection phase are the ones who keep helping in ways that do not help.

Why a year

Because this is a gigantic task, and gigantic tasks are done in small pieces by people who keep showing up.

Not because a year is sacred. Because the thing being changed is not your software, it is your reflexes — what you reach for when something arrives, what you do on a Friday afternoon, what happens in the third week of a bad month. Reflexes do not change in a weekend and they are not changed by reading. They change by repetition, under load, with somebody watching, and then they change again next year because you are not the same person.

That is also why this schedule is meant to be run again. Each pass the structure gets simpler and the Big Why gets clearer, and neither of those things can be rushed by wanting them.

What is being asked of you

An hour a week. A pencil. Your own real material, however embarrassing. And one uncomfortable habit: saying the true number out loud.

That is all. There is nothing to buy, no product to install, no community to join, and no version of this that requires you to become the sort of person who has a system. You will be rebuilding something you already have — a memory, a set of commitments, a body of knowledge — in a form you can actually work with.

And if the system ever costs more than it returns, delete it. It is a servant. It was always a servant. The point of the year is not the harness. It is that a person who has ordered what they know is free to get on with what they are for.

Contents

• 0. Why bother
• 1. What this document is, and what it is not
• 2. How to read a week's entry
• 3. How to use this document: read all of it, then jump
• 4. The whole person: four things we are actually training
• 5. Why the spiritual dimension is not decoration
• 6. The fixed shape of every session
• 7. The four-week rhythm inside every block
• 8. Staircase, rotation, and the annual refactor
• 9. The eleven things that never change
• 10. Seasonal operating modes and the calls made in advance
• 11. What we say, and what we refuse to say
• 12. The year, at a glance
• 13. The cross-reference map
• Part Two: The Blocks
• Appendix A — Words we use, and what they mean
• Appendix B — The forty things, in one page
• Appendix C — Input discipline: the full case
• Appendix D — Scope, budget, and the arithmetic of attention
• Appendix E — The reading list, and the rule about reading lists
• Appendix F — Where the design comes from
• Appendix G — The fifty-two drills, and what each one needs
• Appendix H — Running this as a group
• Appendix I — Troubleshooting: the hard cases

1. What this document is, and what it is not

This is a full year of weekly sessions for learning to keep a personal knowledge system: a place where what you read, decide, learn, promise, and intend to do is captured, sorted, kept findable, and turned into work that actually gets finished.

It draws on four traditions that were invented separately and mostly do not know about each other.

Building a Second Brain, from Tiago Forte, supplies the vocabulary of capture, organize, distill, express and the four containers of PARA: Projects, Areas, Resources, Archives. It supplies the crucial claim that a note is worth keeping only if it will be used, and the crucial warning that collecting is not the same as thinking.

Getting Things Done, from David Allen, supplies the mechanics of capture, clarify, organize, reflect, engage, the discipline of the next physical action, the two-minute rule, the four criteria for choosing, and above all the weekly review. Allen's actual contribution is not a list method; it is the observation that the mind is for having ideas, not for holding them, and that an open loop you have not decided about will cost you attention every day until you decide.

Toyota, and the shop floor — 5S, kanban, andon, poka-yoke, standard work, kaizen, and the seven wastes — supplies the physical discipline that knowledge work usually lacks. A workbench either has a place for every tool or it does not. A note system either has a place for every note or it does not. The shop floor has spent seventy years learning how to make that sentence true, and it did it with visual management, written standards, small numbers, and relentless small corrections rather than with enthusiasm.

Deming and the quality tradition supplies the rest: that most defects are process defects rather than character defects, that you cannot improve what you do not measure, that measurement without a stable process is superstition, and that the correct response to an error is to change the process rather than to try harder next time.

It is not a note-taking course. It does not teach a method for taking notes on books, and it takes no position on whether you should write in the margins, in a separate document, or not at all.

It is not an app guide. No product is recommended, endorsed, or required anywhere in these pages. Where an example names a category of tool it names a category, and the specific product is deliberately omitted, because this document is meant to survive the death of every product currently installed on your machine. If a week's build task cannot be completed without buying something, the task is skipped, not bought. That sentence is a rule of the program and not a suggestion.

It is not a productivity aesthetic. There is nothing here that is designed to be photographed. The most useful artifacts in this document are a wall chart in pencil and a one-page checklist with coffee on it.

It is also not a program that makes anybody smarter. What it trains is narrower and much more testable than intelligence. By the end of the year a participant should be able to say, with evidence: I can capture reliably; I can find almost anything in thirty seconds; I can name the next physical action on any project I own; I know how much work I can honestly have in flight; I can tell you what I finished last month; and I can tell you, without flinching, what I dropped.

That is the whole claim. Everything below is machinery for making that claim true, and for making it true again next year, because a system that is not rebuilt annually is not a system — it is a sediment.

2. How to read a week's entry

Every week in Part Two is laid out the same way, so that a person can open to any page in the document and know exactly where they are and what to do next.

• Build task. A concrete, physical change made to your own system, during the session, before you leave the room. Not homework. Not a plan to build something. The thing itself: a folder created, a list rewritten, a board assembled, a file renamed, a page deleted. If you cannot point at the result, the build task was not done.

• Drill. A named, repeatable skill, practiced in ten minutes with a timer running. It has a name so that you can say it out loud, teach it to somebody else, and recognize it when you see another person doing it badly. Names matter more than they sound like they matter. A person who can say "that is a mind sweep" has a handle for a thing they can then improve. A person who cannot name it can only have opinions about it.

• Concept. One idea, in a single sentence, that you are meant to carry out of the room. It is written to be slightly too compressed on first reading, so that it opens up over the following week.

• Verse. One passage of Scripture, quoted where it is short enough and cited where it is not. This is not a decorative addition and it is not a reward for finishing the work. It is the reason the work is worth doing, and section 5 explains why that ordering is deliberate.

• Teaching. Six to nine sentences, written out rather than summarized, for whoever is leading. It says what the idea means, why it matters to a person who is tired or behind or new, and how it connects to what is on the screen in front of them. The aim is that nobody ever has to improvise a lesson they do not feel ready to give.

There is a sixth thing that is not a slot, and it is the most important one: the link lines. Every block opens with a short See also paragraph naming the other blocks that overlap it. Those are honest, not decorative. Very little in this document is genuinely sequential, and section 3 explains how to use them.

3. How to use this document: read all of it, then jump

This section exists because of a specific failure that happens to almost everyone who tries to learn a knowledge system from a book, and it is worth naming before we start.

The failure is this. A person reads the chapter on capture, and it does not quite make sense to them, because capture only makes sense if you already know what you are capturing for — and that is in the chapter on expression, which is six months away. So they assume they misunderstood, or that the chapter was badly written, or that they need a better tool. Then they stop. Or worse, they push through obediently, build the wrong thing well, and discover in week thirty that they have been diligently maintaining a structure that does not serve their actual work.

So there are three rules for using this document, and they are rules rather than preferences.

Rule one: read the entire schedule before you begin. Not carefully, not with a notebook. Skim all fifty-two weeks and all nine appendices in one sitting, the way you would read the map of a building before walking into it. It should take an hour. The purpose is not comprehension; the purpose is to know what is coming, so that when a week feels arbitrary you can recognize what it is building toward.

Rule two: when something is unclear, ask what it is for, not whether it is wrong. Almost every week in this document is a part of the whole rather than a step in a line. Distillation looks pointless until you have tried to retrieve something you wrote eight months ago. Retrieval looks like an afterthought until you have a vault with four thousand notes in it. The correct response to an unclear week is the question how does this relate to the rest of the system? — and the answer is usually in a different block, findable through the See-also lines at the top of the block you are reading. If the answer is not there, the answer is in Appendix B, which compresses the entire year onto one page.

Rule three: you may jump ahead, and you should, but build with the end in mind. Two things are true at once and the tension between them is real. A person who jumps to block eight because they are excited about expression will produce a beautiful empty vault. A person who refuses to look past block one because the document says "staircase" will produce a tidy pile of nothing. The resolution is not a compromise; it is a different order of operations. Look ahead freely. Build backward. Read where the year ends, decide what you want to be able to say about your own work in week fifty-two, and then come back and build toward it in whatever order your actual life permits.

Rule four: this schedule is meant to be run again. It is not a course you complete. It is a year you repeat. Next January you will start at block one again, and roughly a third of it will be wrong for you in a way you could not have known this year, and roughly a third of it will be more useful than it was the first time because you now have real material to run it on. The appendices are written to be edited in place. The weeks are written to be reordered. Nothing here is sacred except sections 1, 5, and 9.

A word about the size of this. The document is long because it is meant to be usable by someone who is having a bad year and cannot remember the plan. A short document assumes a reader with momentum. This one assumes a reader on a Tuesday.

4. The whole person: four things we are actually training

A curriculum that only trains mechanics produces a person with immaculate folders and an unfinished life. This program is organized so that every session touches four dimensions of a person, and every block names which one it is leaning on hardest.

The technical dimension

The mechanics: where a thing goes, what a thing is called, how a thing is found, how fast it can be retrieved. This is the shallow one, it is the one that can be finished, and it is the one that gets all the attention in books about note-taking.

It matters anyway, and it matters for a reason that is easy to miss. A system you cannot operate while tired is not a system. The technical dimension exists to make the right action the easy action — one inbox instead of six, one naming convention instead of judgment calls at eleven at night, one board with a limit instead of a list of ninety things. Every mechanic in this document is there to remove a decision, because a decision you have to make repeatedly is a decision you will eventually stop making.

The cognitive dimension

Attention and judgment: naming, deciding, deferring, deleting, and the difference between a topic and a project, between an area and a resource, between something you are interested in and something you are responsible for.

A person who can explain why their taxes are an Area and not a Project has gained something no folder structure gives them, because the same distinction will show up in a budget meeting, a family conversation, and a job description. This is why the teaching sentences exist, why the drills have names, and why the leader is supposed to ask questions rather than issue instructions. Naming a distinction is the whole skill. The folders are downstream.

The behavioral dimension

What happens under load. A late Tuesday. A flooded inbox. A deadline week. A month where the whole thing was abandoned and now re-opening it produces dread.

This is where knowledge systems actually fail, and they fail quietly, and the failure is almost always moralized. People decide they are disorganized, or undisciplined, or not the kind of person who can keep a system. They are almost always wrong about this. What happened is that the system had no capacity mode: no written answer to the question what do I do when I cannot do the full version? So the choice became all-or-nothing, and nothing was the cheaper branch.

The capacity weeks exist for this and for nothing else. Every one of them is an attempt to answer that question in writing, in advance, while somebody is calm, so that the answer can be read later by somebody who is not.

The spiritual and stewardship dimension

The honest gap between the system you have and the system you claim to have.

A knowledge system's characteristic sin is not laziness; it is quiet self-deception that is dressed as diligence. The note saved instead of read. The "someday" list nobody opens. The book summarized to avoid thinking about it. The project that has been in progress for eleven months and has been talked about in three meetings. The vault that grows while the output does not. All of it feels like work, all of it is measurable, and none of it is the thing.

Everything in this document that sounds harsh is aimed at that sin and at nothing else. Section 5 makes the argument for why it is a spiritual matter and not merely an efficiency matter.

5. Why the spiritual dimension is not decoration

This section is the one that some readers will skip, and it is the one that determines whether the other fifty-one weeks survive contact with a real year. So it goes near the front and it is written plainly.

First: the material is enormous and a person can drown in it. PARA has four categories and about eleven edge cases that will consume a weekend. GTD has six horizons of focus, and each horizon has its own review, and people have built careers selling the difference between them. 5S has five steps, and lean has seven wastes, and quality management has control charts, and kanban has Little's Law, and a thoughtful person can spend a decade learning how to organize their reading and never once finish anything they read. This is not hypothetical. The literature on personal knowledge management is full of people who are extremely good at managing inputs and have not shipped an argument in three years.

The traditions in this document did not begin as self-improvement. 5S came off a factory floor where the standard was set by whether a person could find the right tool in five seconds while a machine was stopped and thirty people were waiting. Kanban came out of a shop where the constraint was real and the cost of work-in-progress was visible. GTD came out of a man who had been, by his own account, a mess, and who was looking for something that survived an ordinary bad week. Deming's whole argument is a moral one: that most of what we call employee failure is management's failure to design a process, and that blaming the person is both unkind and ineffective. These are not aesthetic traditions. They are traditions about stewardship of limited things — tools, time, attention, materials, and other people's patience.

Second: attention is the thing being spent, and it is not refundable. Every week of this curriculum costs an hour, and every hour of attention that goes into maintaining a structure is an hour that does not go into the work the structure was supposed to serve. That is a real cost and it deserves to be weighed every single week rather than assumed. The recurring question in this document — does this cost more than it returns? — is not a productivity question. It is a question about what you are doing with a life.

Third: the honest answer to "why bother" is not "to be more productive." A person who organizes their knowledge in order to optimize their output will eventually discover that output is not a coherent goal, because you can always want more of it. A person who organizes their knowledge in order to be faithful with what they have been given has a stopping condition. That is the difference between a system that serves you and a system that eats you, and it is the reason this document has a verse in every week rather than a quotation.

Fourth, and this is the argument that actually matters: the work of ordering is itself a devotional act, or it is nothing. Scripture is unusually interested in this. The tabernacle instructions are obsessive about order, and about the fact that the order is not Moses's idea. The Proverbs are full of diligence, which in Hebrew is closer to sharp than to busy — the diligent person is the one whose work cuts. The parable of the talents is about stewardship of what was given rather than about talent in the modern sense, and the servant who is condemned is not the one who produced least; he is the one who did nothing with what he had, out of fear. The man who built bigger barns is not condemned for poor planning. He is condemned for planning as if the barns were the point.

So the devotional frame is not a garnish sprinkled over a productivity curriculum. It is the load-bearing wall. Three consequences follow, and they are the operating theology of this document:

1. The system is a servant and never a master. When the system and the person conflict, the person wins, every time, without discussion. A review that costs a marriage is not a good review. A vault that displaces the work is a barn.
2. Faithfulness is measured in what was actually finished, not in what was collected. This is why blocks seven through nine exist, and why the year does not end with a beautiful archive but with a handover document and a teach-back.
3. Rest is a command, not a reward. A system with no sabbath in it will be abandoned, and the abandonment will feel like failure, and it will not be failure. It will be a sabbath that arrived without permission. Block twelve makes that permission explicit and writes it down.

If a reader takes nothing else from this document, take the ordering in point two: collect nothing you will not use, and finish something every single week. Everything else here is plumbing.

6. The fixed shape of every session

The session is the same every week of the year. That sameness is not laziness; it is the single most important design decision in the document, and it is copied deliberately from the way a shop floor runs and the way a good small group runs.

A person who drops in twice a year should recognize the shape immediately. A person who is anxious about new situations should not have to learn a new format on top of learning new behavior. And a person who is having a genuinely terrible week should be able to participate by following the room instead of by reading the instructions.

Seventy-five minutes. In December it is cut to sixty by shortening the build segment to fifteen minutes. In a session where only two people have come, the drill stays, the build stays, the measure stays, and the circle shortens to three minutes. On a week where a person is drowning, the build is replaced by the re-entry protocol from week forty-seven and nobody says a word about it.

Three things are never cut: the measure, the verse, and the closing commitment. The measure because a system you cannot see is a system you will start lying about. The verse because it is the reason. The closing commitment because an intention that has not been said out loud and written down is not an intention, it is a mood.

What the wall chart is actually for

The wall chart is a shop-floor device and it is doing shop-floor work: it makes the invisible visible, so that variation becomes a fact instead of a feeling.

One line per week, one number per line, twelve to fifteen weeks visible at a time. Nothing digital. Nothing pretty. You are not tracking productivity; you are learning what your own normal variation looks like, which is the difference between "my inbox is out of control" (a feeling that produces shame and no action) and "my inbox is at 240, it is usually under 60, and it was also at 240 in September after the conference" (a fact that produces a plan).

Appendix D explains how to read the chart, including the one mistake everyone makes with it.

7. The four-week rhythm inside every block

The year is thirteen blocks of four weeks. Each block has one theme, and inside each block the four weeks follow the same pattern, so that learning the pattern once means you can predict any week of the year, including next year's.

Week one of a block: Teach. Slow, no pressure, a great deal of talking, nothing timed that does not need to be. The goal is that everyone leaves able to name the thing and describe when they would use it. If people are tired at the end, the teaching was too long. If people cannot say the name of the thing, the teaching was too short. There is no third failure mode, and the fix for both is the same: cut material, keep the name.

Week two of a block: Repetition. The same skill, a hundred times, on real material rather than examples. Volume without speed. This is where the skill stops being something you know and becomes something you do — and it is the least interesting week of every block, which is why it is the one most often skipped, which is why the people who skip it are still reorganizing their folders in year three.

Correction in this week is offered quietly and individually, in a single sentence, and never as a public demonstration of somebody's error. The shop floor learned this the expensive way: an andon cord that gets you yelled at does not get pulled.

Week three of a block: Capacity. The same skill, under a realistic and slightly unpleasant load: a full inbox, a genuine deadline, a busy week, a phone instead of a desk, a month of neglect. In a controlled dose, with a written fallback. This is the week that trains the behavioral dimension, and it is the week that tells you whether the thing you built in weeks one and two is a system or a hobby.

Week four of a block: Flow and circle. Light, integrated work that uses the block's theme without demanding anything new. Followed by a longer verse segment and a longer conversation. This is the week a person can bring a friend, arrive late, leave early, miss half of it, and still take something home.

Why this order and not some other. Teach before repetition because a skill you cannot name cannot be corrected. Repetition before capacity because load applied to a sloppy movement produces a fast sloppy movement. Capacity before flow because flow is what a skill looks like after it has survived pressure, and doing it earlier produces the pleasant illusion of competence that collapses the first busy month.

8. Staircase, rotation, and the annual refactor

The martial-arts version of this document is deliberately a rotation: you can join in week forty-three and lose nothing, because the blocks are a cycle and the person who arrives late is not behind.

This document is not that, and it should say so plainly instead of pretending.

This one is a staircase. Block one is genuinely before block two, because you cannot sort a pile you have not collected, and block seven is genuinely after block six, because you cannot express a thought you have not distilled. Pretending otherwise would be exactly the kind of flattering lie this whole document is against. There is an order here, it is defensible, and section 3's rule three already tells you what to do when your life disagrees with it: look ahead, build backward, jump if you must, and accept that you will redo some of it.

What is preserved from the rotation model is the more important half: every week is complete on its own. If a person attends only week thirty-four, they leave the room with a working board, one number, one idea, one verse, and one commitment. They will lack the earlier context, and the teaching says the context out loud in its first two sentences, which is precisely why the teaching is written down rather than left to the leader's memory.

The annual refactor, which is not optional

This schedule is meant to be run again, and it is meant to be wrong the second time.

A first pass through this year will produce three kinds of result. About a third of the blocks will be exactly what you needed and will become permanent. About a third will be roughly right but wrongly weighted — the right idea at the wrong size for your actual work. And about a third will be wrong for you in a way you could not possibly have predicted in January, because you did not yet know what your own material looked like.

The third category is the valuable one, and it is only available if you write down, in week fifty-two, which blocks those were. That is what the redesign week is for, and it is why the year does not end with a celebration.

Three rules for the annual refactor:

1. Keep four things. Not thirteen, not "everything that worked." Four. A person who keeps everything has not refactored, they have decorated.
2. Change the standard, not the intention. If the weekly review did not happen, the fix is a five-minute version with a written trigger, not a resolution to try harder. Deming's rule applies to your own Saturday: change the process.
3. Date the new document. A plan without a date becomes a policy. A plan with a date is allowed to expire, which is the only thing that keeps it honest.

9. The eleven things that never change

These are not seasonal and they are not optional. They are what keeps a good system alive past its second year, and every one of them exists because somebody's system died without it.

1. It is free. No course is sold from this document, no template is monetized, no tool is recommended in exchange for anything. If you are being taught this material and somebody wants money for the next part, leave.
2. No tool evangelism. The system must survive a change of tool. A week that only works in one application is a failed week, and a person who cannot describe their system without naming a product does not yet have a system.
3. One home for everything. An item lives in exactly one place. Duplicates are the only true defect in this entire document, and they are the only thing worth being pedantic about. Everything else is a preference.
4. Nothing is deleted before it is archived, and nothing is archived in order to avoid deciding. The first half is a mercy to your future self. The second half is a guard against the most respectable form of procrastination available to an organized person.
5. If it costs more than it returns, delete it. In front of other people if possible, and say why. Section 5 explains why this is a spiritual rule and not an optimization.
6. No one's notes, inbox, or screen is shown without their specific spoken permission. Stated before the first session, not after a problem. This is the knowledge-work equivalent of the rule about photographs, and it exists for the same reason: the things people keep are not decorative.
7. Honest numbers only. If a number on the wall chart is embarrassing, it goes up anyway. A chart you curate is worse than no chart, because it produces false confidence at exactly the moment you needed a warning.
8. One person speaks at a time, and no one is required to speak. Silence is a legitimate contribution and is treated as one.
9. No reading about systems in place of using one. The reading list is one page, it is in Appendix E, and it does not grow during the year.
10. The weekly review is not cut for time. It is shortened, never skipped. A missed review is data, not a moral failure, and it goes on the chart.
11. The last ten minutes belong to the number and the commitment. They are not traded for more build time, no matter how interesting the build has become. This is the rule that gets broken first and the one that matters most, because the build is always more fun than the accounting, and the accounting is what makes the build count.

10. Seasonal operating modes and the calls made in advance

This curriculum assumes a temperate northern-hemisphere year in which a person's work has a shape: a January that is quiet and full of resolve, a spring that fills with obligations, a summer that is disrupted, and an autumn that arrives with more due than there are weeks. If your year does not have that shape, rotate the labels and keep the week numbers, because the week numbers are the unit and the seasons are a convenience.

The calls made in advance, in writing, and not negotiated in the moment

Every one of these exists because the decision is easy when you are calm and impossible when you are not. Write them on the same page as your wall chart.

• The week you fall off for a month. You do not restart the system and you do not rebuild it. You run the re-entry protocol from week forty-seven, which takes one hour and loses nothing. Falling off is expected, planned for, and not a moral event.
• The week a tool dies, changes its pricing, or is discontinued. You do not perform an emergency migration during a deadline, ever. The escape hatch is the export test from week thirty-two and the naming convention from week thirty-three, which exist precisely so that a tool's death is an inconvenience rather than a crisis.
• The week the inbox is over two hundred items. You do not process it. You do a mind sweep and then a triage, in that order, because processing a large backlog item by item is how people lose three weekends and their enthusiasm in the same month.
• The week more than three projects are in progress. You do not add a fourth. You finish one or you kill one, in front of a witness if possible. This is the kanban rule and it applies in every month, not only in block ten.
• The week the review would take more than ninety minutes. You are not reviewing, you are re-planning. Stop, do the five-minute version, and put the re-plan on the calendar for a day when it can be done properly.
• The week somebody asks you to adopt their system. You say no, politely, and you write down what they suggested so you can consider it in week fifty-two. Changing systems mid-year is the second most common cause of abandonment after falling off for a month.

11. What we say, and what we refuse to say

Words create expectations, expectations become promises, and promises become the thing you are measured against. So the claims here are deliberately small, and the refusals are deliberately specific.

We say: a sustained weekly habit; a place for everything; one home for each item; a measurable reduction in the time it takes to find something you know you have; fewer open loops; more finished work; a written answer to what happens in a bad month. We say that people who do the review every week tend to end up trusting their own system, and that we have watched it happen, and that it takes about eight weeks before it stops feeling like homework.

We do not say: that this makes anyone smarter, more creative, or more original. That it is evidence-based — it is a synthesis of practices with strong industrial and clinical provenance, applied to an area where the controlled trials do not exist. That a second brain substitutes for thinking, or that retrieval is the same as understanding. That a larger vault is a better one, or that a well-organized archive is an achievement. That this improves career outcomes on its own. That it reduces anxiety in any clinical sense. We do not put a number on the hours it saves, because the honest number varies by an order of magnitude between people and the dishonest number is the one that sells books.

And the standing answer when somebody asks whether it works:

"Come for eight weeks and tell me what you found."

Not because the question is rude, but because the answer is genuinely local. The only evidence that matters is whether you can find your own material in thirty seconds and whether anything got finished. Both of those are checkable, by you, this quarter, without a study.

12. The year, at a glance

The four threads that run through all thirteen blocks. These are not blocks. They are the things you should expect to see again, in a different costume, roughly every three months, and the reason a week can feel familiar in a way you cannot place.

• The door. What enters and on what terms. Blocks 1, 6, 7, 8, 13.
• The container. Where it lives and what kind of thing it is. Blocks 2, 3, 9, 13.
• The limit. How much can be in flight, how long it may sit, how much it may cost. Blocks 4, 10, 11, 12.
• The accounting. What is actually true, measured, and said out loud. Blocks 5, 10, 11, 12.

13. The cross-reference map

The blocks overlap more than the table above suggests, and the overlaps are the substance rather than an accident. This map is the one to photocopy. It is the answer to "how does this relate to the rest of the system," which section 3 says is the correct question to ask when a week is unclear.

Part Two: The Blocks

Block 1 — Capture: the door and the filter

"Write down the revelation and make it plain on tablets, so that a herald may run with it." — Habakkuk 2:2

"Your mind is for having ideas, not holding them." — David Allen

Weeks 1–4 · Winter · Rhythm: teach, repetition, capacity, flow

Why this block, why now. January is the month people rebuild their systems and the month they abandon them, usually in that order, and the reason is almost always the same: they start with folders.

You cannot organize a pile you have not collected. You cannot sort what is still in your head, your phone, three notebooks, two email accounts, and a browser with forty-one tabs. And you cannot know what is worth keeping until you have first seen what you have, in one place, at one time — which is an experience most adults have not had since they were children, and which is genuinely confronting the first time.

So this month contains almost no organizing. It contains collection, and the admission of how much has never been collected. That admission is the actual work of January, and it is what makes every later block possible. A person who arrives in February with an honest capture habit and no structure at all is in a far better position than a person who arrives in February with beautiful folders containing a tenth of their material.

What a newcomer should expect. Nothing is deleted this month and nothing is thrown out. No existing structure is dismantled. You will be asked to write down things you have been carrying for years, and one of these weeks will probably make you angry, and that is normal and is not a sign that you are doing it wrong. The total cost is a cheap notebook and somewhere to put paper.

See also. The capture habit is meaningless without a destination, which is block 2. What you should refuse to capture is block 6, and it will make more sense after you have felt the weight of a full inbox. What you are capturing for is block 8; read that preamble now if this month feels pointless, because it is the answer to "why bother."

Week 1 · Capture is a promise

Build task. Create exactly four capture points and no more: one paper surface that is always on your person, one place on your phone, one place on your computer, and one voice or quick-capture path for when your hands are busy. Label all four with the same word. Then write a single sentence on paper, in pen, saying what those four feed into, and tape it where you will see it. Four doors, one room.

Drill — "The Four Doors." Ten minutes, timer running. Take the four capture points and put something in each one, deliberately, from something already in your head. The point is not the content. The point is that you physically perform the act of capture into each door once, in one sitting, so the path stops being theoretical. Most people discover in this drill that one of their four doors does not work — the phone app takes four taps, the notebook is in the other room, the voice memo never gets listened to — and that discovery is the finding. Fix the door you will actually use, not the one you wish you would use.

Concept. A capture system is a promise to your future self, and a promise you break weekly is worse than no promise at all.

Verse. Habakkuk 2:2 — "Write down the revelation and make it plain on tablets, so that a herald may run with it."

Teaching. God tells Habakkuk to write it down, and then gives the reason: so that a herald may run with it. The writing is not for the writer's memory. It is so that somebody else, later, at speed, can act on it. That is a much higher standard than "I will remember," and it is the standard capture actually has to meet. The test of a captured item is not whether you feel organized. It is whether a stranger — or you, in March, having forgotten everything — could pick it up and run. Notice also that the instruction is make it plain. Not clever, not complete, not beautifully worded. Plain. Almost all capture failure is traceable to a private shorthand that only made sense while the thought was still in the room, and the whole failure is visible in a single glance at anyone's oldest notebook. Today, write things down as though someone else will read them, because someone else will, and that someone is you with no memory of today.

Week 2 · The mind sweep

Build task. Conduct a full mind sweep and produce a written list of at least one hundred items. Not tasks. Items: everything that has your attention and is not in a trusted place. Use a timer and a prompt list. Keep it ugly and keep it fast. Put the finished list into your paper capture point, and do not sort it.

Drill — "The Sweep." Ten minutes, timer running, prompts read aloud if you are in a group. Work through the standard triggers in order — home, work, family, money, health, commitments you have made to people, commitments people have made to you, things you are avoiding, things you are afraid you have forgotten, things you said yes to and regret — and write one line for each thing that surfaces. Do not fix anything. Do not open anything. The rule for the ten minutes is: capture only. No action, no judgment, no triage. If your hand stops moving, read the next prompt out loud and start again.

Concept. The reason you cannot think clearly is not that you have too much to do; it is that you are using your memory as a storage medium.

Verse. Psalm 139:23–24 — "Search me, God, and know my heart; test me and know my anxious thoughts. See if there is any offensive way in me, and lead me in the way everlasting."

Teaching. The psalm does not ask God to remove the anxious thoughts. It asks him to know them and to see them, and only then to lead. That order is worth sitting with, because our instinct when overloaded is to suppress or to power through, while the psalm's instinct is to bring the whole thing into the light first. A mind sweep is a small, secular imitation of that prayer: you are asking what is actually there rather than what you would prefer to be there. Expect two things. The list will be longer than you guessed, usually by a factor of two or three. And a large fraction of the items have been occupying attention for months without ever being written down, so writing them will feel like setting something heavy on a table. Nothing has been solved. That is the point. The mind is a terrible filing cabinet and an excellent workbench, and the sweep is how you clear the bench — not because clearing is progress, but because you cannot do any work on a bench that is full.

Week 3 · Capture under load

Build task. Take a normal working day and keep a capture surface within reach for all of it. Every interruption, every idea, every "I must remember," every request that arrives — one line, timestamped, into the capture point. At the end of the day, count them. That count goes on the wall chart, and it is the first number of the year.

Drill — "The Interception Log." Ten minutes at the end of the day, timer running. Read back through the day's capture and mark each line into one of three categories: mine to do, mine to decide, someone else's. Do not process any of them. The purpose is to discover the ratio, because the ratio is diagnostic. A person whose day is ninety percent someone else's has a boundary problem, not an organization problem, and no folder structure will fix it. A person whose day is mostly mine to decide has a decision problem, which is block three. A person whose day is almost entirely mine to do is probably capturing too little, because real days are full of other people.

Concept. You cannot manage an inflow you have never measured, and most people have never once measured theirs.

Verse. 2 Corinthians 10:5 — "...we take captive every thought to make it obedient to Christ."

Teaching. The phrase is take captive, which is an act of force applied to something that is moving and does not want to stop. That is an accurate description of what capturing is like on a busy Tuesday: a thought appears, you are in the middle of something, and the thought either gets written down or it wins. Paul is not talking about note-taking, but he is talking about the same structural problem, which is that an untaken thought runs loose and starts issuing orders. Here is the practical arithmetic. An interruption costs more than itself; it costs the twelve minutes of re-entry afterward, and you pay that cost whether or not you wrote the thing down. Writing it down does not remove the interruption. It removes the re-entry, because the thing is now somewhere other than your head. Today you are not becoming more productive. You are measuring the size of the door, because a door you have never measured is a door you will be surprised by every week for the rest of your life.

Week 4 · The collector's audit, and the circle

Build task. Count two numbers and write both on the wall chart: how many items you captured this month, and how many of them you have since acted on or filed. Then look at the gap. Do not fix the gap today. Write one sentence about what the gap is made of.

Drill — "Catch and Miss." Ten minutes, timer running. Go back through the month's capture points and find five things you clearly should have captured and did not — the appointment you nearly missed, the name you lost, the promise you made twice. Write each one down, then write in one line which door failed. Eight times out of ten it is the same door, and it is not the one you have been improving.

Concept. The measure of a capture system is not how much it holds; it is how little it drops.

Verse. Psalm 139:1–4 — "You have searched me, LORD, and you know me. You know when I sit and when I rise... Before a word is on my tongue you, LORD, know it completely."

Teaching. There is a record of your life that is complete, and it is not yours. God knows the word before it is on your tongue; you lose the word between the thought and the pen. That asymmetry is worth holding onto for two reasons. The first is comfort: the things you have forgotten are not forgotten by everyone, and your failure to keep a complete record is not a failure to be known. The second is proportion. You are not building an archive of your life, and you should not try, because the attempt will eat the life it is trying to record. You are building a working memory slightly better than the one you were issued. Four blocks from now you will be sorting this month's material, and the only question that will matter is whether the good things made it through the door. Today, look at the gap, name what it is made of, and close the first month by saying out loud one thing you intend to stop dropping.

Block 2 — The four containers: PARA

"Let all things be done decently and in order." — 1 Corinthians 14:40

"Only the simplest, most effortless habits endure long term." — Tiago Forte

Weeks 5–8 · Winter · Rhythm: teach, repetition, capacity, flow

Why this block, why now. You now have a pile. A pile needs four drawers and one rule for deciding which drawer, and that is the entirety of what PARA is: Projects, Areas, Resources, Archives.

• A Project is a short-term effort with a goal and an end. It is finished when something specific is true. Move the workshop into the garage. Finish the certification. Get the estate paperwork filed.
• An Area is an ongoing standard you are responsible for maintaining, with no end date. Health. Finances. The house. My direct reports. My marriage.
• A Resource is material on a topic you are interested in, with no obligation attached. Bread baking. Roman aqueducts. Kubernetes.
• An Archive is anything from the other three that is no longer active. Nothing dies; things graduate.

The distinction that does almost all the work is Project versus Area, and it is a single test: does it end? If yes it is a project. If no it is an area. Almost every organization system people build fails at exactly this point, because they file by topic — Work, Church, Family — and then cannot tell what to do next, because a topic has no next action.

This is the block where people discover that most of their "projects" are areas and most of their "areas" are hobbies. That discovery is uncomfortable and it is the most valuable hour of the month.

What a newcomer should expect. Nothing is deleted. Everything unmoved gets red-tagged rather than discarded. If you have never built a folder structure and are starting from a shoebox, you are not behind; you are, in one specific way, ahead, because you have less to migrate. Do not rename everything to match the examples in this document. The four words are the structure. Your nouns are your own.

See also. Project versus Area is really a scope question, and if the test feels arbitrary now it will feel obvious after block 11. Sorting without deciding only relocates the problem, which is block 3. The naming convention built in week eight is the same one tested for survival in week thirty-three.

Week 5 · Projects versus Areas, and the one test

Build task. Take your top ten existing folders, notebooks, stacks, or buckets and write each one's name on a card. Then apply the test to every card: does it end? Sort the cards into two physical piles, Projects and Areas. Do not create the structure yet. Just sort the cards. Then write down, on a separate sheet, every card you could not decide about, and why in one line.

Drill — "The Ends Test." Ten minutes, timer running. Take twenty items from your capture pile and ask only one question of each: does this end, or is it a standard I keep? Ten seconds each, maximum. If you cannot answer in ten seconds, the card goes into the undecided pile and you move on. The undecided pile is the interesting output, because the boundary between a project and an area is genuinely blurry for about a fifth of real items. "Get healthier" is an area. "Complete the physical in March" is a project inside it. Learning to say both — to hold a project as a temporary object inside a permanent area — is the actual cognitive gain of this week.

Concept. Filing by topic gives you a library. Filing by commitment gives you a day.

Verse. Luke 14:28 — "Suppose one of you wants to build a tower. Won't you first sit down and estimate the cost to see if you have enough money to complete it?"

Teaching. Jesus's point in that passage is about counting the cost of discipleship, but the mechanism he uses is a project manager's mechanism, and it is worth noticing how casually he assumes his listeners understand it. A tower is a project. It has an end, a cost, and a moment of completion, and it can fail in a way that is visible to everyone who walks past. That visibility is the reason projects are morally useful: they can be finished or not finished, and the difference cannot be argued with. Areas cannot do that. You cannot complete your health. What you can do is notice that "be healthier" is not a thing that ends, and therefore is not a thing that can be scheduled, and therefore is the kind of item that sits in a list for four years producing low-grade guilt. Today's test is not pedantry. It is the difference between a list that clears and a list that accumulates. Be strict with yourself: if it does not end, it does not go on the project list, and the relief you feel when you move it is not laziness. It is accuracy.

Week 6 · 5S Sort, and the red tag

Build task. Sort two hundred items — files, papers, notes, screenshots, saved links — into three destinations: keep in Projects or Areas, red tag, redirect to a Resource. Red-tagged items do not go in the bin. They go in one physical or digital container labelled with today's date and a note that says what happens to it in ninety days. That container is the entire point of 5S's first step.

Drill — "Ten Seconds or Red Tag." Ten minutes, timer running, strict ten seconds per item. If you cannot decide in ten seconds, it is red-tagged. No exceptions, no "let me just look at this one." The discipline of the timer is not about speed; it is about preventing the decision from becoming a research project. The old shop-floor version of this is seiri, and its actual rule is harsher than most people expect: anything not used in the last year, and not required by law or contract, leaves the work area. The red tag exists because leaving the work area is not the same as destroying, and people will not sort honestly if sorting means throwing away.

Concept. Sorting is not deciding what to keep; it is deciding what to remove from the place where work happens.

Verse. Matthew 13:47–48 — "...the kingdom of heaven is like a net that was let down into the lake and caught all kinds of fish. When it was full, the fishermen pulled it up on the shore. Then they sat down and collected the good fish in baskets, but threw the bad away."

Teaching. Two details in that parable are easy to miss and both are about sorting. The first is that the sorting happens after the catch is on the shore, not in the water — which is to say, you cannot sort while you are still fishing, and trying to do both at once is why so many people capture badly. The second is that they sat down. Sorting is a seated, unhurried, deliberate act with the whole catch in front of you, and it is done in one session rather than a bit at a time over a year. That is precisely what the red tag makes possible. You are not deciding the fate of two hundred things this week; you are deciding which of them belong in the place where you do work. The rest go into a labelled container that will ask you again in ninety days, when your judgment will be better because you will not have touched them once in three months.

Week 7 · The migration

Build task. Choose one domain of your life — work, or household, or a single committee, or one volunteer role — and migrate that entire domain into PARA in one sitting, with a timer. Create the four containers for that domain. Move everything that belongs to it. Do not touch any other domain this week. One domain, done completely, beats four domains done halfway, because a half-migrated domain teaches you nothing about whether the structure works.

Drill — "Four Buckets, One Domain." Ten minutes, timer running. With the four containers open side by side, take the domain's material and place each item, narrating out loud what it is and why it goes where it goes. Speaking the reason is the drill. It forces the decision out of your hands and into language, and language is where you find out whether you actually have a rule or just a feeling. If you cannot say the reason in one sentence, the item is red-tagged and you move on.

Concept. A structure you have not migrated is a structure you have only agreed with.

Verse. Haggai 1:5–7 — "Now this is what the LORD Almighty says: 'Give careful thought to your ways. You have planted much, but harvested little...'"

Teaching. Haggai is writing to people who had returned from exile and built themselves panelled houses while the temple lay in ruins, and his diagnosis is not laziness. It is misplaced attention: they were working hard on the wrong structure, and the evidence was that their effort did not produce a harvest. That is an unnervingly precise description of a person who has spent four years improving a filing system for material they never use. The instruction is give careful thought to your ways — not "work harder," and not "feel bad," but examine the actual shape of what you are doing. Migrating one domain is how you examine it. You will find items you have moved four times and never opened, which is a fact about them. You will find three projects that are actually the same project under different names, which is a fact about you. Both findings are cheap this week and expensive in October.

Week 8 · Set in order, and the archive day

Build task. Write your naming convention on one page and apply it to fifty files. Then set the depth rule: no more than three levels inside any container. Anything deeper is a folder that has become a hiding place. Finally, run an archive day: take every project that is finished or dead and move it to Archives, with the completion date in the title or the file name.

Drill — "Three Levels Deep." Ten minutes, timer running. Find every folder or notebook path in your system that is four or more levels deep, and flatten it to three. Each time you flatten one, say out loud what you expected to find in its place. The expectation is the interesting part: people build deep trees because they believe they will remember the path, and they never do, and the depth was never the retrieval mechanism — the name was.

Concept. Order is a set of decisions made once so they never have to be made again.

Verse. 1 Corinthians 14:40 — "But everything should be done in a fitting and orderly way."

Teaching. This instruction comes at the end of a long passage about chaos in a church gathering — everybody speaking at once, nobody understood, visitors concluding that the group is mad. Paul's remedy is not less enthusiasm. It is arrangement: one at a time, in turn, so that the thing can actually be received. The parallel is exact. Order in a knowledge system is not tidiness for its own sake and it is not aesthetics. It is the condition under which anything can be received by a person who arrives later, including you. Note also the word fitting — the order has to suit the thing being ordered, which is why the three-level rule is a rule and the specific names are not. Your convention should fit your material; it should not fit somebody else's example. Close this block by writing your convention on one page and dating it, because in twelve months you will want to know what you actually decided in February, and you will not remember.

Block 3 — Clarify: the decision, not the pile

"Let your 'Yes' be 'Yes,' and your 'No,' 'No.'" — Matthew 5:37

"If you don't pay attention to what has your attention, it will take more of your attention than it deserves." — David Allen

Weeks 9–12 · Winter · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Sorting without deciding only moves the problem into a nicer box. This is the hardest block in the year and the one most people quietly skip, and the reason is simple: it requires a decision on every single item, and decisions are expensive in a way that movement is not.

An open loop is a decision you have not made. That is the whole theory. It is not an item you have not filed and it is not a task you have not done — it is a question you have not answered, and until you answer it, your mind will keep re-asking it at the worst possible moments, which is what people mean when they say they cannot stop thinking about work.

There are only six answers, and there is no seventh. Every item that has ever entered your life resolves to one of these:

1. Do it — if it takes less than two minutes.
2. Delegate it — and then it goes on a waiting-for list with a date.
3. Defer it — a next action with a specific physical first step, on a list you actually look at.
4. Calendar it — if it must happen on a particular day or at a particular time, it is not a task, it is an appointment.
5. File it — reference, no action, into Resources.
6. Discard it — the answer nobody chooses and everybody needs.

The failure mode in this block is not confusion. It is the sixth answer, which people avoid because discarding feels like losing, and so items that should be discarded get deferred instead, and deferred items metastasize into a list of three hundred things, and a list of three hundred things is not a plan — it is a monument to indecision.

What a newcomer should expect. You will be slower than you want to be and slower than the group. A beginner processes twenty items an hour; a person who has done this for a year processes eighty. That ratio is not talent, it is familiarity. You will also feel an urge to stop and do the items you are processing, and you must not, and this is why the drill has a timer: the timer is there to keep you processing rather than executing.

See also. A next action with no calendar slot is a wish, which is block 4. A next action with no limit on how many exist is a fantasy, which is block 10. The question of what should never have entered the funnel at all is block 6.

Week 9 · The processing question

Build task. Process your capture pile from the bottom up, one item at a time, out loud, using the six answers. Bottom-up means oldest first, and the reason is not sentiment: the oldest item is the one that has been costing you attention longest, and clearing it produces the first genuine relief of the year.

Drill — "What Is It?" Ten minutes, timer running, one item every thirty seconds. For each item, say out loud, in this order: "This is a ___." (an action, an appointment, a piece of reference, a project, or nothing) — "The next physical action is ___." — "It goes to ___." If you cannot complete the second sentence, the item is not a task, it is a project, and it goes on the project list instead. The physical word matters. Not "handle the insurance." Find the policy number and write it on the form. If the sentence does not begin with a verb your hands can perform, you have written a topic, not an action.

Concept. An item you have not decided about is charging you rent every day.

Verse. Matthew 5:37 — "All you need to say is simply 'Yes' or 'No'; anything beyond this comes from the evil one."

Teaching. That last clause is startling, and it is not hyperbole about grammar. Jesus is saying that the habit of hedging — the elaborate maybe, the deferred answer, the soft commitment that is not quite a commitment — has a moral character, and it is not neutral. Hedging is how a person avoids the cost of a decision while keeping the benefits of being asked. In a knowledge system, the hedge has a physical form: the item that sits in the inbox because deciding feels worse than carrying. And the cost is paid by someone, always, either by you in attention or by the person who is waiting for an answer they think is coming. So today, be blunt with yourself in a way you would not be with another person. Yes, no, or a specific date. The pile does not need to be organized. It needs to be answered. Note what happens in your body when you say a clear no to something you have been carrying for eight months — that sensation is the thing this whole block is trying to teach, and it cannot be learned by reading about it.

Week 10 · The four lists

Build task. Create four lists and no more: Next Actions, Waiting For, Someday/Maybe, and Reference. Move everything you processed last week into one of them. Then delete the other lists you have been keeping — the ones with names like Misc, Later, Important, and To Think About, which are not lists, they are places to hide.

Drill — "The Tree." Ten minutes, timer running, one hundred items if you have them. Run each item through the tree in order and stop at the first match: Is it actionable? No → trash, reference, or someday. Yes → will it take under two minutes? Yes → do it now. No → is it mine? No → delegate, and it goes on Waiting For with a date. Yes → is there a specific day it must happen? Yes → calendar. No → Next Actions, with a physical verb. The order is not arbitrary. The two-minute test comes before delegation because delegating a two-minute task costs more than doing it, and the calendar test comes last because calendars fill up and lists do not.

Concept. A list that mixes commitments with possibilities will be trusted for neither.

Verse. Exodus 18:17–18 — "What you are doing is not good. You and these people who come to you will only wear yourselves out. The work is too heavy for you; you cannot handle it alone."

Teaching. Jethro's advice to Moses is the first recorded management consultation, and its structure is worth studying because it is not "work harder" and it is not "prioritize." It is change the shape of the work: handle the hard cases yourself, delegate the rest to capable people, and let them carry it. The diagnosis is precise — you will only wear yourself out — and it identifies the actual mechanism of collapse, which is not volume but the attempt to hold all volume personally. There is a second detail that is easy to miss. Jethro says to appoint people who will hate dishonest gain, which is to say he cares about their character and not only their competence. Applied to a list: a Waiting For list is a delegation list, and it only works if you name a specific person and a specific date, because "someone should look into this" is not delegation and will never come back to you. Today, count how many items on your list are somebody else's to do, and notice how long they have been yours instead.

Week 11 · The two-minute rule and the four criteria

Build task. Process one hundred items in a single sitting, applying the two-minute rule strictly. Then, on the resulting Next Actions list, apply the four criteria to choose what to do next: context (where am I, what do I have), time available (three minutes or ninety), energy (sharp or spent), priority (what actually matters today). Write the four criteria on a card and put it where you will see it when you are deciding what to do at nine in the morning.

Drill — "Ten Before Two." Ten minutes, timer running. Take ten items and process them in under two minutes each, with a visible stopwatch. Anything that takes longer is not permitted; it goes to Next Actions. The drill is not about speed, it is about discovering how many of the things you have been carrying are actually four-minute jobs. In most groups the answer is between forty and sixty percent, and that number is the single most demoralizing and encouraging fact of the year, because it means the pile was never as large as it felt — it was only undecided.

Concept. Energy is a resource with a shape, and most people schedule their day as though it were flat.

Verse. John 9:4 — "As long as it is day, we must do the works of him who sent me. Night is coming, when no one can work."

Teaching. There is a version of this verse that gets used to justify frantic busyness, and that version is wrong. Read what it actually says: there is a limit, the limit is not in your control, and the appropriate response to a limit is to work in the light while there is light rather than to pretend the limit is not there. That is a time-management statement, and it is also a humility statement. The four criteria exist because a flat list of actions implies that all actions are interchangeable and all hours are the same, and neither is true. Reading and deciding are not the same task and cannot be swapped. A difficult conversation at four in the afternoon after eight meetings is not the same act as the same conversation at nine in the morning, and pretending otherwise is how people conclude they are bad at their job. Today, apply the criteria and write down what you actually chose and why. Next week, when you are tired, you will not have to make that decision again, because you will have made it once, in writing, while you were not tired.

Week 12 · Empty as a state, not an achievement

Build task. Count what remains in the inbox and put the number on the wall chart. Then write a single page, by hand, that answers three questions: what is this inbox actually for? what does it mean when it is full? and what is the honest count of items I decided to discard this month? Keep the page. It goes into the front of the system and stays there for the year.

Drill — "The Honest Count." Ten minutes, timer running. Go through everything you deferred in the last three weeks and mark each item: still real, was never real, or was real once. Count the three. The middle category is the medicine. It is the number of things you carried for months that were, on inspection, never going to happen, and seeing the number written down is what changes the behavior permanently — not resolve, not a new tool, and not a better method.

Concept. An inbox is a device for producing clarity, not a record of diligence, and the two get confused most often by people who are good at capture.

Verse. Ecclesiastes 12:12 — "Of making many books there is no end, and much study wearies the body."

Teaching. That line is written by somebody who loved books, and it is not anti-intellectual; it is anti-infinite. The Preacher's whole argument is that every human activity, pursued without a stopping condition, eats the person pursuing it, and that this is not a flaw in the activity but a property of created things. A knowledge system has an infinite surface. There will always be one more article, one more folder, one more tag, one more book to summarize, and the appetite for it is unbounded in a way that is flattering, because it feels like virtue. So the block ends here, with a written statement of what the inbox is for, because that sentence is the stopping condition. Without it, you will do this work forever and finish nothing, and you will be able to explain, in impressive detail, exactly why finishing nothing was the correct decision. Close the winter by being honest about the discard count. It is the first number in this document that measures judgment rather than motion.

Block 4 — Time: the only non-renewable inventory

"Teach us to number our days, that we may gain a heart of wisdom." — Psalm 90:12

"You can do anything, but not everything." — attributed to David Allen

Weeks 13–16 · Spring · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Spring is when the calendar fills and the list stops functioning as a plan. The list was never a plan; it was a menu. A menu is fine in January. By April, everything on the menu is due at once, and the person discovers that they have been using a list as a substitute for a schedule, which is roughly like using a shopping list as a substitute for going to the shop.

Nothing in this block is about doing more. It is about one distinction, held firmly for four weeks: a thing that has a time goes on the calendar; a thing that has a place goes on a list. Most people's chaos is a category error between those two. An appointment kept on a task list is a task that will be missed. A task put on a calendar as a two-hour block is a fiction that will be deleted at four in the afternoon when the day has eaten it.

The second idea is that time is the only inventory that cannot be replenished, borrowed, or manufactured. You can buy more paper and more storage. You cannot buy more Tuesday. Every argument in this block follows from that, and so does every argument in block 11 about scope, because scope is just a claim on time that somebody has already made before checking the stock.

What a newcomer should expect. You will time-block a week, it will fall apart by Wednesday, and that is the intended outcome of the first attempt. A time-blocked week that survives intact is not a sign of skill; it is a sign you blocked for too little. The skill is in the recovery, not the plan.

See also. The calendar is where the weekly review lives, and a review that is not on the calendar will not happen. The limit on how much may be in flight is block 10, and it is the same argument applied to projects instead of hours.

Week 13 · Calendar versus list

Build task. Take next week and put everything that has a time onto the calendar first — appointments, meetings, calls, deadlines, recurring obligations, travel. Then look at what is left. That remainder is your actual discretionary capacity for the week, and most people have never seen it, because they have never subtracted the fixed from the whole.

Drill — "Time or Place." Ten minutes, timer running. Take fifty items from your Next Actions and calendar and sort each one with a single question: does this have a time, or a place? A time means it must happen at a particular hour, usually because another person is involved. A place means it can happen whenever you are in the right context. Anything with a time goes to the calendar immediately. Anything with a place stays on a list. Items that have neither are not actions yet, and they go back for clarification.

Concept. A task list is a menu; a calendar is the meal. Confusing them is the most common cause of a week that disappears.

Verse. Psalm 31:15 — "My times are in your hands."

Teaching. That is a short sentence and it is doing a lot of work. It is not fatalism, and it is not the sort of resignation that produces a passive person. It is a statement about ownership: the hours are not mine to allocate as I please, they are held by someone else, and I am a steward of a schedule I did not write. That changes the emotional texture of a full calendar considerably. If the hours are mine, then a full week is a personal failure and every interruption is a theft. If the hours are held, then a full week is a fact about the season and the correct response is to find out what is actually being asked of me in it. Practically, this is why the calendar comes first and the list second. The calendar is the honest record of what has already been promised to other people, and a list that ignores it is not a plan, it is a set of intentions with no arithmetic behind them. Today, look at your next week before you look at your list, and see how much of the list was never going to happen.

Week 14 · Time blocking and the Eisenhower sort

Build task. Build one week as thirty blocks, colored or marked by type rather than by project: deep work, meetings, admin, errands, rest, margin. Leave at least five blocks empty on purpose. Then put the Eisenhower sort on one page — urgent and important, important not urgent, urgent not important, neither — and place your own recurring commitments into the four quadrants. If you have never done this, the answer will be uncomfortable and it is not a trick.

Drill — "Thirty Blocks." Ten minutes, timer running. Sketch next week on paper as thirty blocks, half-day by half-day. Fill only the blocks you are confident about. Then count the empty ones. Fewer than three empty half-days means the week has no capacity for anything unexpected, and a week with no capacity for the unexpected will be spent entirely on the unexpected, because the unexpected is the only thing with authority.

Concept. A plan with no slack is not a tight plan; it is a plan that will be rewritten by whoever interrupts you first.

Verse. Ephesians 5:15–16 — "Be very careful, then, how you live — not as unwise but as wise, making the most of every opportunity, because the days are evil."

Teaching. The phrase usually translated making the most of every opportunity is a commercial word: it means buying up the whole of something while the market is there, the way a trader takes a position before the price moves. It assumes scarcity and it assumes urgency, and it assumes that the opportunity will not be available later. That is a strange thing to say about time, and it is exactly right. The days are evil, in the sense of being unreliable, hostile to good intentions, and full of other people's emergencies. The wise response is not to fill every hour — that is the response of a person who believes they control the market. The wise response is to buy the hours that matter while they are cheap, which means protecting a few blocks with real force, and leaving the rest available for whatever the day actually brings. Note also that the passage is addressed to a community and not to an individual, which means the blocks you protect are partly protected for other people. Margin is not selfishness. It is the only thing that lets you say yes when someone needs you on a Wednesday.

Week 15 · Capacity: a real deadline

Build task. Run one day under a genuine deadline with one protected block and the phone physically in another room. Record three numbers: how many times you left the block, how many times you were interrupted, and how long the block survived. Those three numbers go on the wall chart.

Drill — "The Protected Block." Ten minutes, timer running, and then a real ninety-minute block if the room allows. Before it starts, write on paper: what the block is for, what the first physical action is, and what you will do when you are tempted to leave. The third item is the one that works. Everyone leaves their own block; the difference between people who finish things and people who do not is that the former have decided in advance what they will do at the moment of leaving, and the decision is usually "I will write one more sentence first."

Concept. Focus is not a personality trait; it is a property of the environment you have arranged.

Verse. John 11:9 — "Are there not twelve hours of daylight? Anyone who walks in the daytime will not stumble, for they see by this world's light."

Teaching. Jesus is answering a question about safety with an answer about working conditions, which is a very practical move. The point is that there is a fixed amount of usable light, it is not negotiable, and the correct behavior is to walk while you can see rather than to insist on walking in the dark and blaming the road. Applied to a day: you have roughly a fixed number of good hours, they are not evenly distributed, and the entire discipline consists of putting the demanding work inside them instead of inside whatever is left after the meetings. This is why the drill measures interruptions rather than hours worked. Hours worked is a number that can always be inflated by staying up later. Interruptions per block is a number about the environment, and the environment is the thing you can actually change. If your answer is fourteen interruptions, the fault is not your willpower and the fix is not discipline. The fix is a door, a phone in another room, and a written sentence about what happens when you want to leave.

Week 16 · Energy, not hours, and the circle

Build task. Track your own energy for seven days, three times a day, on a scale of one to five, and note what you had eaten, how you had slept, and what kind of work you had just done. At the end of the week, mark your two best hours of the day on a card and put the card on your wall. Then defend those two hours for a month and see what happens.

Drill — "Two Hours." Ten minutes, timer running. Write down, from memory and from the week's log, the two hours of the day when you are most capable of difficult thought. Then write down what currently occupies those two hours. The gap between the two answers is the most actionable single fact in this entire document, and closing it does more for output than any change to your notes.

Concept. Hard work is not a matter of how many hours you have; it is a matter of whether the right hours were spent on the right thing.

Verse. 1 Kings 19:7 — "Get up and eat, for the journey is too much for you."

Teaching. Elijah has just done the most impressive thing in his life and then collapsed under a broom bush and asked to die, and God's response to a suicidal prophet is not a rebuke, not an exhortation, and not a lecture about perspective. It is food and sleep, twice, and then a very long walk. That sequence is in Scripture deliberately, and it makes two claims that the productivity literature mostly does not make. The first is that some failures of will are failures of the body, and treating them as spiritual or moral problems is both unkind and ineffective. The second is that the correct response to depletion is not more discipline but restoration, and that restoration is a command rather than a reward. Notice also that Elijah's great victory immediately preceded his collapse, which is the pattern most people discover the hard way: the week after the biggest push is the week you cannot think. Plan for it. Close this block by writing down the two hours and the two recoveries, and then doing both for a month.

Block 5 — The weekly review and the horizons

"Let us examine our ways and test them, and let us return to the LORD." — Lamentations 3:40

"The weekly review is the critical success factor." — David Allen

Weeks 17–20 · Spring · Rhythm: teach, repetition, capacity, flow

Why this block, why now. The weekly review is the highest-return habit in this entire document and it is the first thing everyone drops. That combination is not a coincidence; it is a structural property of anything that pays out slowly.

By April you will have four months of capture, four months of sorting, and four months of decisions. None of it will hold together unless something goes back over all of it regularly, closes the loops that have reopened, notices the projects that have quietly died, and answers the question is this still true? That is a review. It is not planning. Planning is deciding what to do; reviewing is finding out what is actually the case. Almost everybody substitutes the first for the second, which is why they spend their lives making excellent plans on top of false premises.

The reason this block treats the review as standard work — a written checklist, executed the same way every time, in the 5S sense — rather than as an act of reflection, is that a review that depends on the mood of the person performing it will only be performed by a person in a good mood. The people who need the review most are the ones least able to invent it on the day. So we write it down.

What a newcomer should expect. The first review will take two hours and will feel like an archaeology project. The second will take an hour. By the sixth it will take forty-five minutes and it will be the most useful forty-five minutes of your week. You will miss weeks. Missing a week is not a failure and does not require a restart; it goes on the chart as a missed week and you resume. The one thing you must not do is replace the review with a longer plan when you have fallen behind, because that is the failure mode of the conscientious, and it is far more destructive than the failure mode of the lazy.

See also. The review is the maintenance schedule, and block 12 is the same document at a coarser resolution. Horizons of focus are the reason block 11 can define "finished" without lying to itself.

Week 17 · The checklist as standard work

Build task. Write your own twelve-item review checklist, by hand, on one page. Not borrowed from this document. Twelve items, in the order you will actually do them, each one a physical instruction rather than a principle. Then run it once, badly, and mark every item you skipped or fudged. The marks are the design notes for version two.

Drill — "Write Your Own Twelve." Ten minutes, timer running. Look at a month of your capture pile and ask: what would I have needed to look at, once a week, to have caught the things I actually dropped? Write one checklist line per answer, up to twelve. A reasonable starting set is: empty the physical tray; empty the phone inbox; review the calendar two weeks back and four weeks forward; review the project list for anything dead; review the waiting-for list for anything overdue; review someday/maybe for anything now possible; review the areas for anything neglected; check the wall chart; choose the week's three commitments; write one sentence about what happened last week. But your list should be yours, because a checklist you did not write is a procedure you will not follow.

Concept. A standard is not a rule imposed on you; it is a decision you made once, while thinking clearly, for the benefit of a version of you who is not.

Verse. Luke 16:10 — "Whoever can be trusted with very little can also be trusted with much, and whoever is dishonest with very little will also be dishonest with much."

Teaching. The small thing in this verse is not small in importance; it is small in scale, and the claim is that scale is the wrong variable to look at. Faithfulness is a property of the person rather than of the task, so the way you handle a fourteen-item checklist on a Thursday tells you something about how you would handle a much larger responsibility, and it tells you truthfully, because there is no audience. That is uncomfortable, and it is also the most hopeful thing in this block, because it means the fourteen-item checklist is not a trivial habit. It is the same muscle that everything else uses. The promise here is not that you will become a disciplined person by doing a checklist. The promise is narrower and better: the checklist is where a person who is not naturally disciplined can be disciplined on purpose, in one bounded hour, on one day of the week, and that is enough to hold the rest together.

Week 18 · Repetition: six daily, one weekly

Build task. Run six consecutive daily mini-reviews of ten minutes and one full weekly review of forty-five minutes. Log each one on the wall chart with a single mark: done, shortened, or missed. The chart is the deliverable. Nothing else this week counts.

Drill — "The Ten-Minute Version." Ten minutes, timer running, six days running. The daily version has exactly three questions and nothing else: what came in that I have not decided about? what did I commit to today that is not written down? what is the one thing tomorrow? That is the whole thing. The temptation will be to expand it, because a ten-minute review feels insufficiently serious. Resist, because the purpose of the ten-minute version is not to be thorough. It is to keep the system warm on days when the full version is not available, and a warm system costs nothing to restart.

Concept. Consistency of cadence beats completeness of content, every time, in every system.

Verse. Daniel 6:10 — "Three times a day he got down on his knees and prayed, giving thanks to his God, just as he had done before."

Teaching. The interesting clause is the last one: just as he had done before. Daniel had a cadence, it had a fixed number and a fixed times, and he kept it when the policy changed and the penalty was a den. The passage is often read as a story about courage, and it is, but the courage was available because the habit was already in place and did not have to be invented under pressure. That is exactly the argument for the daily ten-minute review. You are not trying to become thorough. You are building the thing that will still be there on the day you do not want to do it. Note also that Daniel's cadence included thanksgiving, which is worth thinking about as a design feature rather than a formality. A review that begins with what went wrong is a review people avoid. A review that begins with what actually happened, including the good, is one people will come back to. Open every review with one sentence about something that worked this week, and notice how much easier the rest of the checklist becomes.

Week 19 · Capacity: the review on the worst week

Build task. Write the five-minute version of your review on an index card and put it in your wallet or phone case. It has three items and no more. Then, this week, deliberately run the five-minute version on a day that has gone badly, and write one sentence about what the full version would have missed that the short one caught.

Drill — "Shrink, Don't Skip." Ten minutes, timer running. Take your twelve-item checklist and cut it to three items by asking, of each one, if I only had five minutes for the rest of the year, would I keep this? The three survivors are your floor. The other nine are your ceiling, and the ceiling is optional. This is the most important design exercise in the block, because it answers, in writing and in advance, the only question that actually decides whether a knowledge system survives: what happens when I cannot do the full version?

Concept. A system without a degraded mode is not a system; it is a fair-weather habit.

Verse. Psalm 73:26 — "My flesh and my heart may fail, but God is the strength of my heart and my portion forever."

Teaching. The psalmist is not describing a bad week. He is describing the collapse of his whole interpretive frame — he had looked at the prosperous and the arrogant and concluded that his own integrity had been pointless, and he says so plainly before he recovers. What is remarkable is that he says it inside the psalm rather than after it. The record of the failure is part of the record of the faithfulness. That is the correct model for a review, and it is the opposite of the model most of us carry, which is that the review is a report card and a bad week is a bad grade. A review that cannot record a bad week honestly will stop being run in the bad weeks, which are exactly the weeks it exists for. So the written floor matters spiritually as well as practically. It is the thing that lets you say this week was a disaster and I still showed up for five minutes, which is a true statement about faithfulness and not a consolation prize.

Week 20 · The horizons of focus

Build task. Answer six questions on paper, one per horizon, in one sitting: runway — what is on my list this week? one year — what am I trying to be true in twelve months? three to five years — what is the shape of the life I am building? purpose — what am I actually for? principles — what will I not do to get there? and the largest frame — who is this all for? Then write, on one page, the single sentence that connects the smallest item on your list to the largest answer.

Drill — "The Fifty-Thousand-Foot Question." Ten minutes, timer running. Pick the project that has been on your list longest without progress. Ask, at altitude: if this project succeeded completely, what would be different about my life in five years? If the answer is nothing, the project is not a project, it is a habit of intention, and it should be killed. If the answer is something you cannot articulate in one sentence, the project is not understood yet and no amount of next actions will save it.

Concept. Most stuck projects are not blocked by a missing action; they are blocked by an unexamined purpose.

Verse. Psalm 8:3–4 — "When I consider your heavens, the work of your fingers, the moon and the stars, which you have set in place, what is mankind that you are mindful of them, human beings that you care for them?"

Teaching. The move in that psalm is a change of altitude, and it is the move this week is trying to teach. The psalmist does not begin with himself. He begins by looking at something enormous, and the effect of looking at something enormous is not that he feels worthless — the passage goes on to say that humanity is crowned with glory and honour and put over the works of God's hands. The effect is proportion, and proportion is the thing that a person with three hundred next actions has entirely lost. This is why the horizons are worth an hour of your year. Not because the answers are reliable, and not because a five-year plan is a real thing. Because the exercise of standing back far enough to see the shape of your own life is the only reliable cure for the specific insanity of a person who is very busy with something that does not matter. Do it once a year. Date the page. Next year you will read it, and the gap between what you wrote and what you did is the most useful document you own.

Block 6 — Input discipline and the attention budget

"Above all else, guard your heart, for everything you do flows from it." — Proverbs 4:23

"Knowledge puffs up, but love builds up." — 1 Corinthians 8:1

Weeks 21–24 · Spring · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Because the year now has a working system, and a working system with no filter is a bigger bucket.

This is the block that the standard literature handles worst. Building a Second Brain is generous about collection and thin about refusal. GTD is fastidious about capture and mostly silent on whether the thing should have been captured at all. Lean is the tradition that actually has an answer — the whole point of a pull system is that nothing enters the line until there is capacity to work on it — and so this block is mostly lean, applied to information instead of parts.

Three ideas carry the whole month.

First: input is not free, because processing is not free. Every item you let in demands a decision, and decisions cost attention. A saved article is not a neutral asset; it is a future obligation with a small recurring charge, and a thousand of them is a debt you will never repay. The reading pile is not a library. It is an accounts payable ledger.

Second: the filter must be positive, not negative. "Be more selective" is useless advice because it is a prohibition with no criterion. The useful version of selectivity is having a small number of live questions that you are actually working on, so that an incoming item is either relevant to one of them or not. Forte's version of this is a list of favourite problems; the research literature's version is a set of standing questions; either works, and what matters is that the number is small, the questions are specific, and they are written down.

Third: attention is a budget with a hard ceiling, and the ceiling is not negotiable. Unlike money, it cannot be borrowed against future capacity in any way that works. You can steal tomorrow's attention by staying up, and tomorrow will present the bill with interest. Every week in this block therefore includes a subtraction rather than an addition: something is removed from the input stream, permanently, and the removal is written down.

What a newcomer should expect. This block produces more guilt than any other in the year, because everyone knows they read too much and finish too little, and knowing it has not helped. The block's answer is not resolve. It is arithmetic and a small number of written rules.

See also. Distillation in block 7 is downstream of this filter; a filter is defined by what it is filtering for. The reason to refuse input is capacity to express. The arithmetic of attention is in Appendix D, and it is the same arithmetic as project scope in block 11.

Week 21 · The door, and what it is guarding

Build task. List every source that can put something into your attention without your explicit invitation: newsletters, feeds, notifications, group chats, subscriptions, auto-play, the phone's home screen. Count them. Write the number on the wall chart. Then turn off every notification that is not a human being asking you something directly.

Drill — "Where It Enters." Ten minutes, timer running. For each source on your list, write the honest answer to one question: did I choose this, or did I accept it? Chosen sources were deliberately subscribed to for a reason you could state out loud. Accepted sources arrived because somebody added you, or because it came with a purchase, or because it seemed useful at the time. The second category is where the volume is, and it is almost always more than half the list.

Concept. You cannot have a filter at the processing stage if you have no gate at the entry stage.

Verse. Proverbs 4:23 — "Above all else, guard your heart, for everything you do flows from it."

Teaching. The instruction is above all else, which is a strange ranking for a book that also commands justice, generosity, and care for the poor. The reason for the ranking is in the second clause: everything flows from it. Whatever is in the well is what comes out of the tap, and no amount of work on the tap changes the well. That is a claim about sources, and it is far more radical than a claim about notifications. What you let in continuously becomes what you think with, and it does so slowly and without your consent. Note also that the verb is guard, which is an active, physical, slightly aggressive word — a gatekeeper's word, not a librarian's. There is a version of this that becomes fearful and small, and it is not the version here: the proverb is not asking you to read nothing, it is asking you to know what is entering and on whose authority. Today, count the sources. Most people find between forty and a hundred, and most people find that they chose fewer than half.

Week 22 · The reading pile is a debt

Build task. Take the entire backlog of things you have saved to read, watch, or listen to later, and make a single honest count. Then, in one session, delete at least half of it without reading it. Not archive. Delete. Write one sentence afterward about what you were hoping those items would do for you.

Drill — "Declare Bankruptcy." Ten minutes, timer running. Take the backlog and apply the brutal filter: if this had arrived today, knowing what I now know, would I save it again? If not, it goes. Then for the survivors, apply the capacity test: do I have, in the next four weeks, the hours to consume what I just kept? If not, you keep the best of what fits and you delete the rest, because a backlog that exceeds capacity is not a queue, it is a lie about your future self.

Concept. An unread pile is not potential; it is a record of promises you made to a person you are not going to become.

Verse. 1 Corinthians 8:1 — "We know that 'We all possess knowledge.' But knowledge puffs up while love builds up."

Teaching. Paul is writing about a genuine dispute in a real church, and he makes a move that is devastating for anyone who thinks of learning as an unqualified good. He concedes the knowledge. He says: you are right, you know correctly, and your knowledge is inflating you. The contrast is not between knowledge and ignorance; it is between an inflating thing and a building thing. Knowledge that is not turned into love — into service, into something that bears weight for another person — is gas. That is a hard sentence and it is the theological form of the collector's fallacy, and it explains why a person can read for ten years and become smaller. Practically: your reading pile is not evidence of diligence. It is evidence of appetite, and appetite untranslated is a puff. Today, delete half of it and notice that nothing happens. That nothing is the lesson. The world does not end, the knowledge you needed still arrives when it is needed, and you have recovered a number of hours that you can now spend building something instead of inflating.

Week 23 · The attention budget

Build task. Write a one-page attention budget for a normal week. Total available waking hours, minus sleep, minus fixed obligations, minus commuting and chores. What remains is the real number, and it is always far smaller than people expect. Then allocate it explicitly: deep work, shallow work, input, output, relationships, rest. The allocation must sum to the total. If it does not sum, you are not budgeting, you are listing wishes.

Drill — "The Subtraction." Ten minutes, timer running, arithmetic on paper. Start at 168 hours in a week. Subtract sleep at your actual average, not your ideal. Subtract meals, travel, and the fixed obligations. Subtract the honest overhead of maintaining a household. Write the remainder in large numbers and look at it. Most people have between twenty and thirty-five discretionary hours, and most people are attempting to spend about sixty.

Concept. Attention cannot be borrowed. Every hour taken from tomorrow is repaid with interest, and the interest is paid in judgment rather than in time.

Verse. Mark 1:35 — "Very early in the morning, while it was still dark, Jesus got up, left the house and went off to a solitary place, where he prayed."

Teaching. The detail that matters is left the house. He did not find a quiet corner; he left, because the house had people in it and the people had needs and the needs were all legitimate. This is the pattern in the gospels and it is worth noticing how deliberate it is: he withdraws, the crowd follows, he serves, he withdraws again. He does not appear to have felt guilty about the withdrawal, and he does not appear to have regarded the crowd's demands as the definition of his day. That is a budget, executed by somebody with less time than you have. The arithmetic in today's drill is not a productivity exercise, it is a truth-telling exercise, and its purpose is to make you stop planning against a number that does not exist. A person who thinks they have sixty discretionary hours will say yes to things that require sixty, and will then fail at all of them and conclude something false about their own character. A person who knows the number is twenty-four will say yes to fewer things and finish them, and will feel, oddly, less busy.

Week 24 · Twelve problems, and the filter that makes the door usable

Build task. Write down the twelve questions you are actually trying to answer in your life and work right now. Not goals; questions. Specific enough that an article could be obviously relevant to one of them. Keep the number at twelve or fewer, and put the list where you capture. Then, for the rest of the year, every incoming item is tested against the twelve before it is allowed in.

Drill — "Relevance in Ten Seconds." Ten minutes, timer running. Take twenty items from your input backlog and test each against the twelve problems in ten seconds. Relevant or not relevant, no third option. Count how many pass. In most groups it is fewer than five out of twenty, and the five are usually better than everything else in the pile.

Concept. Selectivity is not a virtue you cultivate; it is a criterion you write down.

Verse. Philippians 1:9–10 — "...that your love may abound more and more in knowledge and depth of insight, so that you may be able to discern what is best."

Teaching. The sequence in that sentence is worth tracing carefully, because it runs against the way we usually think about discernment. It begins with love abounding, and it moves through knowledge and depth of insight, and it arrives at the ability to discern what is best. The goal is not more knowledge. The goal is discrimination — the capacity to tell the best from the merely good, which is far harder and far rarer. Note that the mechanism is love, which sounds sentimental until you notice what it does: love narrows. A person who loves a particular people, a particular place, a particular work has a criterion that a person who is interested in everything does not have. That is why the twelve problems work. They are not a reading list, they are a statement of what you love enough to spend a life on, and they turn an infinite stream into a finite set of relevant things. Write the twelve. Expect to be embarrassed by how vague the first attempt is. Rewrite them in October, and notice that they have become specific, and that the specificity is the year's actual output.

Block 7 — Distill: notes worth re-reading

"For it is precept upon precept, precept upon precept, line upon line, line upon line, here a little, there a little." — Isaiah 28:10

"Compression is a fundamental operation of the mind." — Tiago Forte

Weeks 25–28 · Summer · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Because the vault is now full of things nobody will ever open again, and the reason is not that they were not worth capturing. It is that they are in the same form they arrived in — somebody else's words, at somebody else's length, written for somebody else's purpose.

Forte's contribution here is genuinely useful and genuinely misunderstood. Progressive summarization is not a highlighting technique. It is the observation that a note becomes more valuable the more times it is used, and that each use is an opportunity to compress it a little further for the next use. Five layers:

1. Layer one — the raw capture, exactly as it arrived.
2. Layer two — the bolded passages, chosen on a second pass, when you are looking for something.
3. Layer three — the highlighted few sentences inside the bold, chosen on a third pass.
4. Layer four — an executive summary in your own words, written when you need the note for something.
5. Layer five — a remix, where the note's content is pulled into something new that you are making.

The critical design principle is that you do not do all five layers when the note arrives. You do a layer when you use the note, and only then, because compression without a consumer is guessing. A person who summarizes every capture to layer five on arrival is doing enormous unpaid work for a reader who may never show up.

Two more terms this block uses and the rest of the document borrows. An intermediate packet is a self-contained piece of work — a paragraph, a diagram, a checklist, a half-finished section — that can be assembled into a larger deliverable without being rewritten. And twelve favourite problems, from block 6, is the filter that decides what is worth distilling at all.

What a newcomer should expect. Distillation is the most pleasant block in the year and the least urgent-looking, which is why people spend years here and never leave. Guard against it. The rule for the month is that every distillation must be attached to a specific thing you are trying to make.

See also. Distillation without retrieval is a hobby; the test is block 9. Distillation without expression is the collector's fallacy in a nicer costume, which is block 8.

Week 25 · Progressive summarization, layers one to five

Build task. Take one long document you have actually used twice — not one you might use — and distill it once per day for four days, one layer at a time. At the end, you have a note you could hand to somebody else. Do not do this to anything else this week.

Drill — "Bold, Then Highlight, Then Say It." Ten minutes, timer running. On a single long note, in three passes, each timed: pass one, bold the passages you would need if you were in a hurry. Pass two, within the bolded text, highlight the sentences without which the passage is meaningless. Pass three, close the note and write three sentences from memory saying what it argues. The third pass is the one people skip and the only one that produces learning, because remembering is a different operation from recognizing, and only one of them builds anything.

Concept. A note you have compressed is worth more than a note you have saved, and the difference is proportional to how many times you used it.

Verse. Isaiah 28:10 — "For it is precept upon precept, precept upon precept, line upon line, line upon line, here a little, there a little."

Teaching. This verse is usually quoted about patient instruction, and it is, but look at what it actually describes: an accumulation of small, repeated, layered additions, in a fixed pattern, until something is built that was not there before. That is compression and it is also pedagogy, and the two are the same operation. You do not understand a thing in one pass; you understand it in the fourth pass, when you notice that the sentence you bolded the first time is not the sentence you would bold now. That change is the entire evidence of learning available to you, and it is why the layers are separated by days rather than done in a sitting. Note also the phrase here a little, there a little, which is the antidote to the fantasy of the comprehensive summary. You are not producing the definitive note on the subject. You are adding one line to a structure that will be useful in a hurry, and the structure's value comes from its shape rather than its completeness.

Week 26 · Repetition: bold ten, highlight three, summarize one

Build task. Distill ten items to layer two, three of them to layer three, and one of them to layer four. Everything with the twelve problems open beside you, so that the irrelevant items are dropped rather than distilled. Volume, no perfection.

Drill — "Ten, Three, One." Ten minutes, timer running, three tracks at once. Start ten notes and bold. Move three of them forward and highlight. Take one of those three and write a four-sentence summary in your own words. Working the three layers simultaneously rather than in sequence is the drill, because it forces you to feel the difference in cost between them, and the difference in cost is what tells you which layer a note deserves.

Concept. The right amount of compression is the least that still lets a stranger use the note.

Verse. Proverbs 25:11 — "A word fitly spoken is like apples of gold in settings of silver."

Teaching. The image is of something small, precise, and set deliberately into a context that makes it visible. It is a craft image, and the craft is not in the gold or in the silver; it is in the fitting. That is exactly the standard for a compressed note, and it explains why most compression fails. People compress by cutting, and cutting produces a smaller version of the same thing. Fitting means choosing the few words that carry the weight of the whole and placing them where the reader will actually meet them. The test is not is this shorter? The test is could somebody else act on this? Which means the summary has to include the thing you would otherwise assume: what problem this solves, what it is for, when not to use it. Those are the parts people leave out, because they were obvious at the time. They are never obvious in March.

Week 27 · Capacity: intermediate packets

Build task. Take one live project with a real deadline and cut it into five to eight self-contained packets, each of which could be used without the others. Each packet gets a name, a single-sentence purpose, and enough internal context that it would make sense to somebody who had never seen the project. Then write the list of packets on the project's page and check off the ones you already have from old work.

Drill — "Ship a Fragment." Ten minutes, timer running. Choose one packet and finish it completely — not the project, the packet. A packet is finished when it could be read, used, or sent by somebody else with no further explanation from you. If you cannot achieve that in ten minutes, the packet was too large, and splitting it is the exercise.

Concept. Work that cannot be divided cannot be finished incrementally, and work that cannot be finished incrementally is work that will be done in a panic.

Verse. 1 Kings 6:7 — "In building the temple, only blocks dressed at the quarry were used, and no hammer, chisel or any other iron tool was heard at the temple site."

Teaching. That is a description of prefabrication, three thousand years early, and it is in the text because it was remarkable. Everything difficult and noisy happened somewhere else, in advance, by people who were not at the site, and what arrived at the site was finished and ready to place. The result was that the building went up quietly. This is the strongest available argument for intermediate packets, and it is an argument about where the noise happens. Most projects are loud at the end, because all the cutting and fitting happens when the deadline is close and everybody can hear it. A person who prepares packets in advance moves the noise earlier, into a period when nobody is waiting, and the delivery becomes quiet. Notice also the dependency: the blocks had to be dressed before they were needed, which means somebody had to know the dimensions in advance, which means the plan existed. That is why this week comes after scope and before the project-management block, and why the packet list is written down rather than remembered.

Week 28 · Flow: pruning, and the twelve problems again

Build task. Go through one area of your vault and delete a third of it. Not archive — delete, or move to a single archive container you will not open. The criterion is not quality; it is would I ever use this for one of my twelve problems? Then rewrite the twelve problems, dated, in light of four months of evidence about what you actually reach for.

Drill — "What Earns Its Keep." Ten minutes, timer running. Open the twenty most recently modified notes and ask of each: did I use this, or did I only make it? Mark them used or made. Count both. The ratio is the single most honest number in this document, and it will be embarrassing, and it will be different next year.

Concept. A collection is not improved by addition. It is improved by the removal of everything that dilutes the signal.

Verse. John 15:2 — "He cuts off every branch in me that bears no fruit, while every branch that does bear fruit he prunes so that it will be even more fruitful."

Teaching. The second half of that sentence is the part that gets skipped, and it is the part that has to do with you. The unfruitful branch is cut off. But the fruitful branch is pruned — which is a more careful and more painful operation, performed on something that is working, in order to concentrate its energy into fewer and better fruit. A vine that is not pruned produces a great deal of small fruit and eventually exhausts itself. If you read this as being about what to delete, you have got half of it. The harder question is what to prune: which of your genuinely good collections of notes, which of your genuinely useful projects, is spread so thin across so many branches that none of them will ripen. The answer usually involves giving up something you are good at. Close the block by writing down the one thing you will deliberately do less of, and put it where you will see it in October.

Block 8 — Express: output is the only proof

"Do not merely listen to the word, and so deceive yourselves. Do what it says." — James 1:22

"The value of a note is not in the note; it is in what the note lets you make." — Tiago Forte, paraphrased

Weeks 29–32 · Summer · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Because it is week twenty-nine, and this is the block that the year exists for, and if it were placed in January nobody would still be present to hear it.

The collector's fallacy is the belief that accumulating understanding is equivalent to producing something. It is not a stupid belief. It is the belief of a careful person, and it is reinforced by an environment in which reading feels like work, saving feels like progress, and nobody ever asks you what you made. It is also, in this document's terms, the characteristic sin of the whole enterprise, and it is why section 5 argues that stewardship is measured in what was finished rather than in what was gathered.

The corrective is not to stop learning. It is a ratio, measured and written down: for every unit you take in, what did you put out? In a healthy system the ratio is not one-to-one, because some input precedes output by years. But the ratio should be visible, because an invisible ratio drifts, and it drifts in the direction of comfort.

Two supporting ideas. Reuse over creation: most people believe they must generate new material when what they need is to bring out old material in a new arrangement. And shipping as a discipline: a thing that has gone out the door to a specific person by a specific date is finished; a thing that is nearly ready is not finished and may never be.

What a newcomer should expect. This block makes people angry, and the anger is usually directed at the tasks. That is a good sign. If a week of this month makes you want to argue that your situation is different, the argument is worth writing down, because sometimes your situation is different and sometimes the argument is the collector's fallacy wearing a good suit.

See also. Scope and definition of done, without which "shipped" is undefined, is block 11. The flow limit that makes shipping possible is block 10.

Week 29 · The collector's fallacy, and one thing out the door

Build task. Before the session ends, ship one small thing to one real person. A paragraph, a diagram, a page, a recommendation, a summary of something you read. It must be small enough to finish today and real enough that the recipient is not a hypothetical. Then write the word shipped on the wall chart next to this week's number, and do that every week for the rest of the year.

Drill — "One Page, One Hour." Ten minutes to choose, then a real block if the room allows. Choose a topic from your twelve problems that you already know something about, and produce one page: a claim, three supports, one caveat. No research. The no-research rule is the whole drill, because it forces you to discover how much you already have. Most people discover that they know considerably more than they thought and have never once assembled it, which is a fact worth knowing about yourself.

Concept. The gap between what you have read and what you have made is not a knowledge gap; it is a production gap, and it closes only by producing.

Verse. James 1:22 — "Do not merely listen to the word, and so deceive yourselves. Do what it says."

Teaching. The verb is deceive, and James is very specific about the mechanism: the person who listens and does not act is not lazy, they are deceived, and the deception is self-administered. He goes on to use the image of a man who looks in a mirror and immediately forgets what he looks like — which is an almost exact description of reading something excellent and being unable to say a week later what it argued. The deception is that the hearing felt like the doing. That feeling is the entire problem, and it is why this block insists on something going out the door today rather than at the end of the month. Not because a paragraph is important, but because the act of producing breaks the spell in a way that no amount of additional input can. Notice also that James does not say the law is difficult to understand. He says the problem is forgetting. The remedy for forgetting is not more study; it is doing the thing while you still remember it.

Week 30 · Repetition: reuse, ten times

Build task. Produce ten pieces of output in one week, and none of them may involve new research. Every one is assembled from material you already have: old notes, old drafts, old conversations, old arguments. Ten is deliberately too many, because the number is what forces reuse instead of creation.

Drill — "Storeroom." Ten minutes, timer running, one output every sixty seconds — a bullet, a paragraph, a message, a list, a diagram sketch. The rule: whatever you produce must cite which old note it came from. Nothing from memory without a source note. The constraint feels artificial and it is the entire point, because it makes visible the inventory you have been ignoring.

Concept. Most of what you need to make already exists in your notes, and the reason you cannot see it is that you have never looked at your notes with a purpose.

Verse. Matthew 13:52 — "Therefore every teacher of the law who has become a disciple in the kingdom of heaven is like the owner of a house who brings out of his storeroom new treasures as well as old."

Teaching. The image is a person with a store, and the action is bringing out — the treasure is not the possession, it is the production. Note that the treasures are both new and old, which means the mature version of this is not only reuse and not only novelty, but a mixture. And note who is described: somebody who has become a disciple, which is to say somebody who has been changed by what they have stored. That is the distinction between a storeroom and a landfill. The storeroom's contents have been digested; the landfill's have not. This is the practical test for your own archive. If you can bring out an old note and it is immediately useful, you digested it when you made it. If you bring it out and cannot tell why you saved it, no amount of reorganization will help, and the honest response is deletion rather than another pass. Today, bring out ten things. Count how many were ready to be used.

Week 31 · Capacity: shipping under a deadline

Build task. Set a real deadline, tell a real person you will deliver by it, and deliver. The subject should be something you would normally take three weeks to prepare. You have this week. Write down, in advance and on paper, the three rules you are using to constrain scope so that delivery is possible: what is in, what is explicitly out, and what the minimum acceptable version looks like.

Drill — "The Constraint Write-Down." Ten minutes, timer running. Before you begin, write three sentences: what this is, what this is not, and the smallest version that would still be useful to the recipient. The second sentence is the one that saves the week. Most scope failure is not over-ambition; it is an unstated boundary, so the work quietly expands into every adjacent interesting thing and the deadline becomes the point at which you stop rather than the point at which you deliver.

Concept. A deadline does not make work take less time; it makes you decide what to leave out, and that decision is the work.

Verse. Ecclesiastes 9:10 — "Whatever your hand finds to do, do it with all your might, for in the realm of the dead, where you are going, there is neither working nor planning nor knowledge nor wisdom."

Teaching. The Preacher's argument, here and throughout, is that death is the great clarifier of effort. Not because effort earns anything, but because the time for effort is bounded and everything unfinished at the boundary stays unfinished forever. That produces a strange and useful kind of freedom. If the deadline is real and the time is short, then the question what should I leave out? becomes a moral question rather than an aesthetic one, and the answer is usually the part that only impresses people who already agree with me. Read the verse again and notice that planning is listed among the things that end. Planning is good. Planning is not the work, and there will come a moment when planning is simply a way of not doing it.

Week 32 · Flow: show the work, and the export test

Build task. Two things. First, put one finished thing in front of one real person who did not ask for it, and ask for a specific kind of response. Second, run the export test: take your system and export everything to a plain, open format — plain text, markdown, or paper — and verify that the export is complete and readable without any application. Write down how long the export took and what was lost.

Drill — "Hand It Over." Ten minutes, timer running. Give your finished thing to somebody and watch them use it without your help. Do not explain anything. Do not answer questions until they have finished. What you learn in those ten minutes is worth more than a month of self-assessment, because you will discover the exact sentence you thought was clear and is not.

Concept. A system you cannot leave is not a system; it is a dependency, and the test of ownership is the ability to hand it over intact.

Verse. Matthew 5:16 — "...let your light shine before others, that they may see your good deeds and glorify your Father in heaven."

Teaching. The instruction is let, which implies that the natural condition of light is to be seen and that the work required is removal of the covering rather than manufacture of the light. That is a different posture from self-promotion, and it is worth distinguishing carefully, because the difference is not in the visibility; it is in the destination of the attention. The verse ends with glory going somewhere other than the person doing the good deed, which is the only test that actually works. Practically: showing the work is not optional, because work that is never shown is work that is never corrected, and uncorrected work does not improve with age. But it matters enormously to whom and for what. Show it to somebody who will tell you the truth and use them for the rest of your life. Then run the export test, which is the other half of the same virtue: your system should be able to leave you, and you should be able to leave it. A vault that only opens in one application has quietly taken ownership of something you thought you owned.

Block 9 — Retrieval: the system's only real test

"Ask and it will be given to you; seek and you will find; knock and the door will be opened to you." — Matthew 7:7

"The purpose of a filing system is not to store things. It is to find them." — unattributed, and universally forgotten

Weeks 33–36 · Summer · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Because everything else in the year is a hypothesis until retrieval is tested, and the test is embarrassingly concrete: can you find a specific thing you know you have, in thirty seconds, on a phone, from a room you are not in?

That is the whole standard. Not "is my system logically organized." Not "does my structure reflect my values." Those questions are unanswerable and infinite, which is why people enjoy them. The retrieval question has an answer, it takes thirty seconds to measure, and it will tell you in one afternoon whether the last eight months produced a system or a hobby.

There are three mechanisms, and they are listed in order of how much they actually contribute:

Naming. Almost all retrieval failure is a naming failure. The thing is in the system; it is called something you would not have thought of. This is why week eight's convention and week thirty-three exist, and why the pedantry about names in this document is deliberate and not a personality flaw. A name is the primary key. Everything else is a fallback.

Indexes. A short note that lists where things are, written for one purpose, updated as things change. Forte calls the general version a map of content; project managers call it a register; hospitals call it a chart. The value is that it is small and hand-written, which is why it can be trusted in a way that an automatic search cannot.

Memory. Spaced repetition and retrieval practice are the only two learning techniques with a very large and boring evidence base, and they are almost entirely absent from the note-taking literature. If you want to have something rather than be able to look it up, these are the only methods that work.

What a newcomer should expect. This is the month where people discover that they have been building a system for an imaginary version of themselves — one who searches neatly, remembers why things were saved, and never has to look something up in a hurry with cold hands.

See also. Naming conventions were decided in week eight and are stress-tested here. The defect taxonomy in block 12 is mostly built out of retrieval failures observed this month.

Week 33 · Search beats structure, and names are the structure

Build task. Write your naming convention on one page, in the form of a rule, not a list: what goes first, what goes second, what is never included. Then rename fifty files or notes to comply. Then, and this is the part people skip, search for ten of them the way you would search in a hurry — three words, no path, no memory of where you put them.

Drill — "The Primary Key." Ten minutes, timer running. Take ten notes and, for each, write the name a stranger would use when looking for it, and the name you actually gave it. Where they differ, rename to the stranger's version. Your own names encode context you had at the time and will never have again. The stranger's names encode what the thing is.

Concept. Structure is what a system looks like from the inside; naming is what it looks like from the outside, and only one of those matters in a hurry.

Verse. Matthew 7:7 — "Ask and it will be given to you; seek and you will find; knock and the door will be opened to you."

Teaching. The three verbs escalate: asking is verbal, seeking is effortful, knocking is physical and local. The progression is a decent description of how retrieval actually fails, too. You ask a system for something and get nothing, so you start seeking — opening folders, reading titles — and the failure happens at the third stage, where you are standing in front of the right door and knocking on the wrong one. Notice that the promise is not that everything will be found; it is that the seeking is not futile, which is a different and better claim. Applied to a vault: the promise is that a well-named thing can be found by somebody who does not know where it is. That is the standard this week is asking you to meet, and it is much more demanding than it sounds, because it requires you to give up the private shorthand that makes you feel clever and to write, instead, the boring name that works in a hurry with cold hands.

Week 34 · Repetition: find-it drills

Build task. Twenty find-it drills over the week, each timed, each logged: the question, the seconds, whether you found it, and — if not — why. Then make the misses into a list and put the list at the top of your system. The misses are the curriculum for the rest of this block.

Drill — "The Thirty-Second Twenty." Ten minutes, timer running, four to six questions maximum. Someone else should write the questions if you can arrange it, because you will unconsciously choose questions you know the answers to. Ask things like: the name of the person who recommended the book; the number you were given in March; the paragraph where you wrote down what you decided about the roof; the diagram you made when you were working out the schedule. Real questions from your own life, not abstractions.

Concept. A retrieval system is not tested by whether it contains the thing; it is tested by whether it returns the thing under time pressure.

Verse. Psalm 119:11 — "I have hidden your word in my heart that I might not sin against you."

Teaching. The psalmist's retrieval strategy is memorization, which is unfashionable and is the only one that works at three in the morning without a device. Note the stated purpose: that I might not sin against you — the retrieval is not for information but for behavior, which is why it must be available without looking it up. That distinction is worth carrying into your own system. Some things you want to be able to find; some things you want to have. A person who has to look up their own principles in a meeting has not yet internalized them, and no search engine fixes that. This is why the drill has a written answer and a spoken answer. Write down what you found, and then say aloud, from memory, the one thing you decided last month that you would do differently. If you cannot, you have found the boundary between what you have stored and what you have actually learned, and it is usually further in than people expect.

Week 35 · Capacity: the minimum viable vault, on a phone

Build task. Build the smallest version of your system that still works, and document it on one page. It must fit on a phone; it must be usable standing up; it must contain your twelve problems, your projects, your next actions, and the five documents you actually need. Then use only that for a week, and write down what you missed.

Drill — "Phone Only." Ten minutes, timer running, device only. Do a complete capture, a complete processing pass, and one retrieval, all on the phone, standing up, with one hand. No desk, no second screen, no sitting down. The constraint is the design brief, and the design brief is the summer rule: if it cannot be done in ten minutes on a phone, it will not be done at all in July.

Concept. The version of your system that you can run in bad conditions is your actual system; everything else is your ideal system.

Verse. Jeremiah 29:4–7 — "...Build houses and settle down; plant gardens and eat what they produce... seek the peace and prosperity of the city to which I have carried you into exile."

Teaching. The instruction is remarkable because the situation was supposed to be temporary. The people had been told they would be in exile for seventy years and were, quite reasonably, still behaving like people waiting to go home — not building, not planting, not investing. The command is to build anyway, in the wrong place, under bad conditions, because the conditions were not going to improve on the schedule they hoped for. This is the correct posture toward a system that does not fit your life. Most people are waiting for the version of their life in which they will have a desk, a block of quiet, and a proper setup, and they will be waiting for years. The person who builds the phone version now is the one who still has a working system when the good conditions finally arrive. Today, build the exile version. It is worse, and it is the one that will actually be there in July.

Week 36 · Flow: indexes and memory, and the circle

Build task. Write three index notes: one for your work, one for your household or family, one for the problem you are most actively thinking about. Each is one page, hand-written, and each lists where things are rather than what things are. Then choose five things you want to hold in your head rather than look up, and set up a simple spaced repetition practice: paper cards are sufficient and are better than an app you will abandon.

Drill — "The One-Page Map." Ten minutes, timer running. Without opening anything, draw from memory the structure of your own system: the containers, the current projects, where the three most important documents live. Then check it against reality and note the discrepancies. The discrepancies are the parts of your system you do not actually believe in, and they are the parts that will fail under pressure.

Concept. An index is worth more than a search box, because an index tells you what exists.

Verse. 1 Chronicles 12:32 — "...men of Issachar, who understood the times and knew what Israel should do."

Teaching. Two things are named in that verse, and they are not the same thing. They understood the times — which is analysis, comprehension, knowing what is going on — and they knew what Israel should do — which is judgment, decision, and command. A large part of the modern knowledge apparatus trains the first and produces very little of the second, which is why a person can be extremely well informed and permanently paralyzed. An index is a small instrument for the second thing. It does not tell you what is true; it tells you what you have, where it is, and therefore what you can decide right now. That is why the index is written by hand and kept short. A generated list of four thousand notes is comprehension. A hand-written page of eleven things you can act on is judgment. Close the block by writing the eleven, dating the page, and putting it where you will see it in the autumn.

Block 10 — Flow and WIP: kanban for one

"Better a handful with quietness than both hands full, together with toil and grasping for the wind." — Ecclesiastes 4:6

"Stop starting, start finishing." — the entire kanban literature, in four words

Weeks 37–40 · Autumn · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Autumn arrives with more due than there are weeks, and the only useful question in that situation is not what is most important — everything is important, that is why it is due — but how much can be in flight at once before all of it goes slowly.

That is what work-in-progress limits are, and the reason they work is arithmetic rather than psychology. If a person is carrying twelve open projects and can finish roughly two a month, then the mean time to finish any given project is six months, no matter how hard they work, and no amount of urgency changes it. Three of those twelve will be finished in the year. The other nine will be "in progress," which is a state that consumes attention in proportion to time spent, and produces nothing. The limit is what converts activity into throughput.

Four ideas carry the month, and they are the whole of practical kanban:

1. Visualize. Everything you are carrying, visible at once, in columns that reflect what actually happens rather than what you wish happened.
2. Limit. A number in every column except the last. When the column is full, nothing new enters, and the only permitted action is to finish something.
3. Pull, not push. Work is drawn into the next stage when there is capacity, not pushed into it because it is ready.
4. Measure cycle time. The timestamp on a card is the most honest number you will ever have about your own capacity, and it is almost always worse than you believe.

What a newcomer should expect. Your first board will have eleven columns because your process feels complicated. It has three columns: to do, doing (with a limit), done. A board with eleven columns is a chart of a process you have not yet understood.

See also. A limit is a budget, and a budget is a scope decision made once instead of daily — that is block 11. The time arithmetic is block 4. The review that enforces the limit is block 5.

Week 37 · Visualize and limit

Build task. Build a physical board — wall, sticky notes, pencil on paper, anything visible without a device — with three columns and a limit written on the middle one. Put every open project on it. Then mark, on each card, the date it started. One number goes on the wall chart this week: the count of projects in flight.

Drill — "Three Columns and a Number." Ten minutes, timer running. Build the board, then apply the limit: if you have more cards in doing than the limit, you do not get to remove the limit; you get to move cards back to to do until you comply. Choosing which ones is the actual work. Most people find it takes about four minutes and produces an immediate sense of loss and a following sense of relief, in that order.

Concept. A limit you have not broken in front of a witness is a suggestion.

Verse. Proverbs 19:2 — "Desire without knowledge is not good — how much more, one who rushes and misses the way."

Teaching. The phrase rushes and misses the way is doing something interesting. It is not describing a person who is slow or idle; it is describing a person with legitimate desire and insufficient knowledge, whose energy is real and whose direction is wrong, so that the energy makes the error worse. This is the exact pathology that a work-in-progress limit addresses, and note that the proverb's remedy is knowledge rather than restraint. It is not that desire is bad. It is that desire without a map produces speed in the wrong direction, and speed in the wrong direction is more expensive than sitting still. That is why the board has dates on the cards. A card that has been in progress since March is desire without knowledge; the date is the knowledge. Do not look at the date and feel bad. Look at the date and learn the number, because an accurate number about your own capacity is a gift and a flattering estimate is the most expensive friend you have.

Week 38 · Repetition: pull, not push

Build task. Work the board for a week and move twenty cards, obeying the limit every time. Count, honestly, how many times you started something new before finishing something existing. That count goes on the wall chart, and next week you try to beat it.

Drill — "Finish Before Finish." Ten minutes, timer running. Every time you complete a card, the rule is: before you pull a new card, finish or file everything the completed card produced. Not next week, not later today. The output of finished work — the document, the follow-up, the note, the administrative tail — is where most of the real inventory hides, and it hides behind the satisfying act of starting the next thing.

Concept. Starting is more pleasant than finishing, which is why the invisible discipline of a system is refusing to start.

Verse. Exodus 16:17–18 — "The Israelites did as they were told; some gathered much, some little. And when they measured it by the omer, the one who gathered much did not have too much, and the one who gathered little did not have too little."

Teaching. The manna story is usually read as being about provision, and it is, but it is also the first recorded work-in-progress limit. The instruction was to gather one day's amount, no more; the people who gathered more found it rotted. The limit was not a punishment for greed — it was a description of the capacity of the system. You could not store it, so storing it destroyed it, so the amount that could be worked was the amount that could be used. That is a kanban board with a limit on the middle column, and the failure mode is two days of manna in the inbox. Note the detail that the one who gathered little did not have too little, which is a claim about sufficiency that is worth testing this month. Most people, when they reduce what is in flight, expect to lose output. What usually happens is that the same output arrives in a different order, sooner, with less dread. Gather what you can use today. Let the rest arrive tomorrow, which it will.

Week 39 · Capacity: cycle time, and Little's Law for one person

Build task. Timestamp ten cards from the moment they enter doing. Measure from entry to done, not from the date you first thought about them. Then apply Little's Law, which is the only equation in this document: average cycle time = average work in progress ÷ average throughput. If you have six items in progress and you finish two per week, your cycle time is three weeks. Write that number on the wall chart and keep it there.

Drill — "The Arithmetic of the Board." Ten minutes, timer running, calculator allowed. Count items in progress. Count items you completed in the last four weeks and divide by four. Divide the first by the second. Then answer one question: is the number on this page acceptable to me? If it is not, the equation says there are exactly two levers — fewer in progress, or more throughput — and only one of them is under your control this month.

Concept. Cycle time is a consequence of how much you are carrying, not of how hard you work.

Verse. James 4:14 — "Why, you do not even know what will happen tomorrow. What is your life? You are a mist that appears for a little while and then vanishes."

Teaching. The verse is usually read as an argument for humility about plans, and it is, but look at what it does to arithmetic. A mist has a duration that is short and not under its own control, which means the only sensible planning question is given that the duration is unknown and short, what is worth putting in it? That is a cycle-time question. It is also the reason the number on your wall chart deserves attention rather than shame. Six weeks to finish a project is not a character flaw; it is a fact about capacity, and facts about capacity are the raw material of any plan that will survive contact with a Tuesday. The man in James's example makes his plans without the number. He says he will go to this city, spend a year there, and make money, and the response is not that commerce is wrong — it is that he does not know the number. Get the number. Write it up. Then plan.

Week 40 · Flow: the bottleneck, and the circle

Build task. Identify your bottleneck in one sentence and write it on the board where you cannot avoid it. Then hold it for a month. The bottleneck is not a metaphor; it is the specific stage where work accumulates and waits. For most people it is not the tool and it is not the research. It is the review — the absence of a fixed weekly point at which decisions are made and things are closed.

Drill — "Where the Pile Is." Ten minutes, timer running. Look at the board and find the column with the oldest cards. That column is your bottleneck, or it is downstream of it. Then ask the question the shop floor asks: what is waiting here, and who is it waiting for? If the answer is you, the fix is a calendar block. If the answer is a decision, the fix is a definition of done — that is, the next block.

Concept. Every system has exactly one constraint at a time, and improving anything else is decoration.

Verse. 1 Corinthians 12:21 — "The eye cannot say to the hand, 'I don't need you!' And the head cannot say to the feet, 'I don't need you!'"

Teaching. Paul's argument is about the body and about status, and the part that translates directly into process work is the claim that the members we are tempted to dismiss as unglamorous are the ones that are indispensable. The bottleneck in a personal system is almost always the unglamorous stage. Nobody wants to be the person whose insight is my problem is the weekly review, because that is a claim about ordinary discipline, and everybody would prefer the answer to be a better tool or a cleverer structure. But the constraint is where it is regardless of what you would prefer it to be. Notice also that Paul's remedy is not to elevate the hand above the eye. It is to acknowledge the dependency, which means the correct response to identifying your bottleneck is not to declare war on it. It is to put the constraint in the calendar first and to stop optimizing everything downstream of it, because improvements made downstream of a constraint do not appear in the output. You will already have noticed this in your own life: the perfect filing system for material that never gets processed.

Block 11 — Scope, sequence, and the honest finish line

"I am carrying on a great project and cannot go down." — Nehemiah 6:3

"Of the three constraints — scope, time, cost — you get to fix two. The mistake is believing you fixed all three." — the whole of project management, in one sentence

Weeks 41–44 · Autumn · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Because in autumn everything is due, and the reason things are due at once is almost never that the person is lazy. It is that the commitments were made without a scope.

Scope management is the most underrated skill in project management and the most useful skill in an ordinary life, because every hard conversation you have ever had about a deadline was, underneath, a conversation about scope that nobody had. "How long will it take?" and "when can you have it?" are the same question, and neither can be answered without first answering "what exactly is it, and what is it not?"

There are four constraints and only four. Scope, which is what is included. Time, which is when it is due. Cost, which is what it consumes — money, people, and in personal work, almost always hours. And quality, which is how good it has to be to count. The rule is not that you can have any three; the rule is that any change to one of the four changes at least one of the others, whether or not anyone says so out loud. A new requirement added in October is paid for in November, and it is paid for in quality or in sleep. There is no fourth option, and pretending otherwise is how people arrive at the deadline with something that is neither on time nor good.

The personal-scale translation is blunt and worth having on a card:

• Fix the date. Negotiate the scope. Never the reverse, because a date is a promise to another person and scope is usually a preference.
• Write down what is out. A scope that lists only inclusions is not a scope; it is a wish.
• Define done before you start. "Finished" must be checkable by somebody else.
• One owner, one next action. Two owners is no owner; three next actions is no next action.
• Kill things on purpose, in public, with a date.

What a newcomer should expect. This is the block where people discover that their project list is a list of areas, and that several of their projects have been "in progress" for so long that the phrase has become a way of describing a mood. Killing them is the relief of the year and it will feel like a loss for about four days.

See also. The limit that makes sequencing possible is block 10. The review that catches scope creep before it eats a month is block 5. Project versus Area, settled in block 2, is the same argument at rest.

Week 41 · Outcome, next action, definition of done

Build task. Rewrite five projects as three lines each: the outcome (what will be true when this is finished, stated so a stranger could check it), the next physical action (a verb your hands can perform), and the definition of done (the specific, checkable condition under which you will declare it complete). Then delete any project for which you cannot write the third line, and note in one sentence why you kept it if you did.

Drill — "Could a Stranger Check It?" Ten minutes, timer running, five projects. For each, read your outcome sentence aloud and ask: could somebody who does not know me verify whether this is true? "Get the house in order" cannot be verified. "The garage floor is clear, the boxes are labelled, and the car fits" can. The test is not pedantry; it is the difference between a project that can end and a project that will follow you into next year wearing a new name.

Concept. A project without a definition of done is not a project; it is an area with a deadline attached, which is the most exhausting object in any system.

Verse. Nehemiah 6:3 — "I am carrying on a great project and cannot go down."

Teaching. Nehemiah's enemies invite him to a meeting four times, and four times he declines with the same sentence, and the sentence is a scope statement. He names what he is doing, he names its size, and he names what going down to the plain would cost — the work would stop. That last part is the whole discipline. He is not being rude and he is not being inflexible; he has a criterion, and the criterion makes the decision for him, which means he does not have to weigh each invitation on its merits at the moment it arrives. Notice also that he does eventually go down, in chapter seven, when the wall is finished and the doors are set. The refusal was not permanent, it was sequenced. That is the difference between a person with a scope and a person who is merely stubborn. Write your three lines this week, and use the third one — the definition of done — as the thing that tells you when you are allowed to go down to the plain.

Week 42 · Repetition: the A3, and five projects on one page each

Build task. Produce five one-page project sheets, one per project, in the A3 format: background (why this exists, two sentences), current situation (where it actually stands, honestly), target (what done looks like, with a date), the gaps (what is between here and there), the plan (three to seven steps, each with an owner and a date), and follow-up (when you will look at this again). One page. Not two.

Drill — "One Page, Six Boxes." Ten minutes, timer running, one page. Draw six boxes and fill them in that order, briefly, for the project that is currently causing you the most anxiety. The box that is hardest to fill tells you which part of the project is actually undefined, and it is usually the gaps — people know where they are and where they want to be and have never written down what is between the two, which is why the plan feels like it should be obvious and is not.

Concept. A plan that does not fit on one page has not been thought through; it has been collected.

Verse. Isaiah 8:1 — "Take a large scroll and write on it with an ordinary pen."

Teaching. The instruction is oddly specific: an ordinary pen, on a large scroll, in plain characters, so that anyone passing could read it. The writing is deliberately public, deliberately legible, and deliberately not precious. That is the A3 discipline exactly. The one-page constraint is not an aesthetic preference; it is a forcing function, because when you cannot use more space you are compelled to decide what matters, and deciding what matters is the thinking you have been avoiding. Note also the destination of the writing: it was to be read by other people. A project sheet that nobody else could understand is a private note, and private notes about projects are where optimism hides. Show one of your five pages to somebody this week and watch what happens when they ask a question you did not anticipate. That question is the value of the exercise, and it costs one page and ten minutes instead of three weeks in November.

Week 43 · Capacity: the project that goes wrong

Build task. Take a project that is late, over budget, or worse than you promised, and write the honest status report. Five lines: what was promised, what is now true, what changed, what it will now take, and what I am asking for. Then deliver it to the person who is waiting, this week, without being asked.

Drill — "The Bad News First." Ten minutes, timer running. Write the report you would want to receive if you were the person waiting. Then write the report you would prefer to send. The difference between the two documents is your entire inventory of avoidance, and it is always smaller than it feels and always larger than you hoped.

Concept. Bad news does not improve with age, and the cost of a delayed report is almost always larger than the cost of the problem being reported.

Verse. Proverbs 28:13 — "Whoever conceals their sins does not prosper, but the one who confesses and renounces them finds mercy."

Teaching. The proverb makes a claim about outcomes that is empirically true in project work and worth taking seriously as a general principle. Concealment does not merely fail to help; it compounds, because the problem grows while the report does not, and the eventual revelation has to account for both. The word renounces is the part people leave out of their status reports. Confession without a change of plan is just disclosure, and disclosure without a plan is what makes people stop trusting you. So the five-line format is deliberate: the fourth line is what it will now take, and the fifth is what you are asking for. Those two lines are the renunciation. They convert a confession into a proposal, and a proposal is something the other person can say yes to. Do it before you are asked. The person who reports a problem before anyone notices is trusted with more, not less, and that is a fact that surprises almost everybody the first time it happens to them.

Week 44 · Flow: kill three projects, in front of witnesses

Build task. Kill three projects on purpose. For each: write the reason in one sentence, remove it from the active board, tell anyone who was promised something, and move it to the archive with the word abandoned and the date. If you cannot kill three, kill one and write down what stopped you on the other two — the obstacle is more instructive than the kill.

Drill — "What Dies So This Lives." Ten minutes, timer running. Choose the one project you most want finished this autumn. Then list everything currently competing with it for your attention, and mark what you will stop doing to make room. Stopping something is the only mechanism that actually creates capacity; starting nothing new creates nothing.

Concept. Capacity is created by subtraction or not at all, and a project you will not finish is a decision you have not made yet.

Verse. John 12:24 — "Very truly I tell you, unless a kernel of wheat falls to the ground and dies, it remains only a single seed. But if it dies, it produces many seeds."

Teaching. The image is not of losing something and being compensated for it; it is of a seed that must cease to be a seed in order to become a plant. Something recognizable ends so that something unforeseeable can exist. That is the correct posture toward killing a project, and it is much harder than the arithmetic version because a project is not only work — it is an identity. The project you have been "working on" for two years is a story about who you are, and killing it kills a small part of that story, which is why people protect dead projects with extraordinary energy and call it commitment. Note that the verse is not a general endorsement of destruction. It is an argument about sequence: fruit requires a burial, and the burial has to be real, which in this case means telling the people who were promised something instead of quietly leaving the project in a folder where it will bother you for another year. Kill three. Say why. Write the date. Then look at what is left and notice how much lighter the board is.

Block 12 — Quality, maintenance, and the re-entry protocol

"Let us examine our ways and test them, and let us return to the LORD." — Lamentations 3:40

"Nobody gives you a medal for firefighting. You get a medal for preventing fires." — the quality tradition, paraphrased

Weeks 45–48 · Autumn · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Because every system accumulates maintenance debt, and a person who has never planned for the debt mistakes it for personal failure.

There are exactly four kinds of defect in a personal knowledge system, and they are worth naming because naming a defect is the difference between being annoyed by it and fixing it.

1. Unfindable. The thing exists and you cannot get to it. This is a naming failure or an index failure, and it is the most expensive defect because it costs you twice — once when you fail to find it, and again when you rebuild it.
2. Undated. The thing exists, but you cannot tell whether it is current. Almost all reference material becomes dangerous rather than useful once its date is lost.
3. Duplicated. The same thing in two places, in two versions, one of which is wrong. This is the only defect this document is genuinely pedantic about, because it is the only one that quietly destroys trust in the entire system.
4. Orphaned. The thing exists and nothing points to it. An orphan is not a defect you notice; it is a defect you notice the absence of, years later, when you needed it.

The quality tradition's core claim is that these are process defects, not character defects. A duplicate exists because the system permitted two doors. An unfindable thing is unfindable because the naming rule did not cover that case. You fix the process, and the defect stops recurring, and you do not have to feel bad in the meantime. That is Deming's contribution to your Saturday, and it is worth more than any productivity technique in this document.

This block also contains the most useful single page in the document: the re-entry protocol. Everyone falls off. The people who recover have written the recovery in advance.

See also. The review in block 5 is where defects are found. The retrieval tests in block 9 are where they are observed. The kaizen habit here is the same instrument as the annual refactor in block 13, at a different scale.

Week 45 · The defect taxonomy, and the fifteen-minute clean

Build task. Write your own defect list — the four categories, in your own words, with one real example of each from your actual system. Then schedule a recurring fifteen-minute weekly clean and put it on the calendar as a repeating event with a name. Not "organize files." A name like Friday 4:00 — clean and close.

Drill — "Five Whys, Once." Ten minutes, timer running. Take the most recent thing you could not find and ask why five times, writing the answer each time. Could not find the invoice → because it was in three places → because two systems were in use → because the second one was adopted in June for one client → because the client required a portal → because there is no rule about portals. The fifth answer is the process defect, and it is almost always a missing rule rather than a missing effort.

Concept. A defect is not evidence about your character; it is evidence about your rules.

Verse. Lamentations 3:40 — "Let us examine our ways and test them, and let us return to the LORD."

Teaching. Three verbs, in order, and each one matters. Examine is looking without defending. Test is checking the examination against something external, because self-assessment without a test is just a mood. Return is the action that follows, and it is worth noticing where it points — not to a better system, but to the LORD. That ordering keeps the whole exercise from becoming self-absorption. The purpose of auditing your own ways is not to achieve an optimized version of yourself; it is to remove the things that are between you and the thing you are for. That is why the defect taxonomy is deliberately boring and deliberately short. A four-item list can be examined honestly in ten minutes. A comprehensive framework of thirty-four categories will be examined once, admired, and never opened again. Examine, test, return. Then do it again next Friday, for fifteen minutes, for the rest of your life, and expect it to be unremarkable.

Week 46 · Repetition: refactor, do not decorate

Build task. Refactor one area of your system: merge what should be merged, split what has grown into two things, rename what is misnamed, and produce a deletion-candidate list of at least ten items. Then delete five of them. Refactoring is not tidying. Tidying moves things; refactoring changes the shape so the problem cannot recur.

Drill — "Merge, Split, Rename, Delete." Ten minutes, timer running, four operations, in that order, until the timer stops. The order is deliberate. Merging first removes work from everything downstream, and deletion last means you have already seen the item three times and cannot claim you did not consider it.

Concept. Decoration makes a system look maintained; refactoring makes it stay maintained, and only one of those is visible from outside.

Verse. Matthew 9:17 — "Neither do people pour new wine into old wineskins. If they do, the skins will burst; the wine will run out and the wineskins will be ruined. No, they pour new wine into new wineskins, and both are preserved."

Teaching. The point of the image is compatibility rather than novelty. New wine is still fermenting and produces gas; old skins have lost their elasticity and will split. The container has to match the contents, and when the contents change, the container has to change too — not because the old one was bad, but because it was built for a different load. Every knowledge system eventually reaches this point. The structure that worked when you had four projects does not work at twenty, and the response is not to squeeze harder. This is why refactoring is a scheduled activity rather than a response to crisis: the wineskin has to be replaced while the wine is still manageable, and by the time it bursts you are cleaning up rather than building. Look at your one area this week and ask the wineskin question directly. Has the volume changed, or the kind of work, or the number of people involved? If yes, the structure is out of date regardless of how tidy it looks, and tidying it is decoration.

Week 47 · Capacity: the re-entry protocol

Build task. Write your re-entry protocol on one index card: seven steps, in order, each taking under ten minutes, that will bring your system back from a month of neglect without a rebuild. Then test it, on purpose, on something you have genuinely neglected. The test is not optional, because an untested protocol is a hope.

Drill — "The Untested Protocol." Ten minutes, timer running. Deliberately apply your protocol to the messiest, oldest, most abandoned area you have. Time each step. If any step takes over ten minutes, split it. If any step requires a decision you cannot make in the moment, rewrite it as an instruction. The standard is that a person who is tired, discouraged, and three weeks behind can follow it without thinking.

Concept. The measure of a system is not how it performs when you are keeping up; it is how it recovers when you are not.

Verse. Jeremiah 18:4 — "But the pot he was shaping from the clay was marred in his hands; so the potter formed it into another pot, shaping it as seemed best to him."

Teaching. The potter does not throw the clay away, and he does not pretend the marring did not happen. He reworks it into a different vessel. That is the correct model for re-entry, and it is a much better model than the two that people actually use, which are restart from scratch and pretend nothing happened. Restarting throws away the clay. Pretending leaves the marring in place and it will show up again in the same spot. Reworking means accepting that the thing you are coming back to is not the thing you left, and shaping it into what is best for the potter's purpose now. Practically, this is why the protocol has a step that says do not process the backlog; sweep and triage instead. You are not restoring February. You are forming a new vessel out of what is still workable, and the rest goes where clay that has dried out goes, without ceremony. Write the card. Test it. Keep it in your wallet for the rest of your life, because you will need it roughly twice a year.

Week 48 · Flow: kaizen, and the control chart

Build task. Two things. Make one small standard change to your system and write the new standard down, dated, on the same page as the old one — do not replace it, because the history is the evidence. Then put twelve weeks of your wall chart numbers on a single page as a simple line and answer one question: is the variation noise, or is it a trend?

Drill — "One Percent, Written Down." Ten minutes, timer running. Choose one small friction in your system — a step you skip, a place you hesitate, a thing you look up repeatedly — and change it. Then write the standard: as of [date], we do [specific thing] this way. The written standard is the whole drill. An unwritten improvement is a preference, and preferences are renegotiated every time you are tired.

Concept. Improvement without a written standard is a mood that will pass.

Verse. 2 Corinthians 10:12 — "When they measure themselves by themselves and compare themselves with themselves, they are not wise."

Teaching. Paul's target here is a group who have invented their own scoring system and are winning it. That is the exact failure mode of self-measurement, and it is worth naming carefully because this whole document asks you to measure yourself and could easily produce the same disease. There are two guards. The first is that the numbers on the wall chart are about the process, not about you — an inbox count is a fact about doors, not a verdict on a person. The second is that the chart is read for variation, not for improvement. A control chart asks whether a reading is inside the normal range for this process; it does not ask whether you are doing well. That distinction is what prevents the numbers from becoming a grade, and it is why the drill ends with the question noise or trend rather than better or worse. Twelve weeks of a number that wobbles between 40 and 60 without pattern is a stable process, and a stable process is a good thing, even if the number is not the one you wanted.

Block 13 — Archive, teach back, and the next year

"There is a time to keep and a time to throw away." — Ecclesiastes 3:6

"The last step of any system is making yourself unnecessary." — the handover principle

Weeks 49–52 · Autumn · Rhythm: teach, repetition, capacity, flow

Why this block, why now. Because the year should close by giving the system away and then taking it apart.

An archive that is a graveyard is a failure, and an archive that is a resource is the entire point of having one. The difference is not organization and it is not tooling. It is whether anything comes out. A resource is drawn from; a graveyard is visited. By December you should be able to name three things you retrieved from the archive and used, and if you cannot, then the archive is a place you are storing things to avoid deciding about them, which section 9 explicitly forbids.

The other thing this block does is teach back, and it is not a formality. A person who can teach a block without notes owns it. A person who cannot has rented it for a year. The gap between those two states is where almost all of the durable learning in this document happens, and it happens in week fifty-one rather than in any of the preceding fifty.

What a newcomer should expect. This block is emotionally strange. People are reluctant to dismantle a system they spent a year building, and the reluctance is worth naming: it is the same instinct that keeps dead projects alive. The year ends with four things kept and the rest released, on purpose, in writing.

See also. The archive inherits the definition of finished from block 11, and the containers from block 2. The refactor here is block 12's kaizen at annual scale.

Week 49 · The archive as a resource, and retention

Build task. Write a one-page retention policy: what is kept forever, what is kept for a fixed period, what is deleted on a schedule, and who decides. Then apply it to one year of accumulated material and delete what the policy says to delete. Put three things you retrieved from the archive this year and actually used at the top of the archive page, as evidence.

Drill — "Three Out of the Store." Ten minutes, timer running. Find three things in your archive you have used since they were archived. Write down what you used them for. If you cannot find three, write down what that tells you, honestly, and reduce the archive by a third this week.

Concept. Retention is a decision, and a decision made once is a policy while a decision made per item is a tax.

Verse. Ecclesiastes 3:6 — "a time to keep and a time to throw away."

Teaching. The Preacher's list is usually read as poetry about seasons, and it is also a filing system with a deletion clause, which is the part nobody quotes. Notice that keeping and throwing away are given the same grammatical weight — neither is the default, and neither is the virtuous option. That symmetry is the whole of a retention policy. A person who keeps everything has not decided; a person who throws everything away has also not decided, they have merely chosen the faster default. What the verse asks for is a judgment about this thing, made once, on grounds you could state out loud. So write the policy, and put the evidence at the top. The three items you actually used are the argument for keeping an archive at all; the absence of three items is the argument for a smaller one. Both arguments are stronger than any opinion you hold about your own diligence.

Week 50 · Repetition: the handover document

Build task. Write the document that would let a stranger run your system. It must cover: where things live, what the naming rule is, what the review covers and when, what the four containers mean in your life, what is archived and on what principle, and the three things that would break if they were done differently. One page, maybe two. Then hand it to somebody and watch them use it.

Drill — "Explain It to Somebody Who Does Not Care." Ten minutes, timer running. Explain your system to a person with no interest in systems, in five sentences, and stop. If you cannot do it in five, you do not have a system; you have a set of habits and a vocabulary. The five-sentence test is brutal and it is the only honest measure of whether the year produced understanding or fluency.

Concept. A system that cannot be handed over is a personal dependency dressed as a methodology.

Verse. 2 Timothy 2:2 — "And the things you have heard me say in the presence of many witnesses entrust to reliable people who will also be qualified to teach others."

Teaching. There are four generations in that sentence: Paul, Timothy, the reliable people, and the people those people will teach. The instruction is not to preserve the teaching but to entrust it, which is a different act — it implies risk, because the person you entrust something to may do it differently or badly. Paul takes that risk anyway, and the structure of the sentence means he has to. A teaching that is only held by one person dies with them, no matter how carefully it was preserved. That is exactly the argument for the handover document. You are not writing it because you plan to leave. You are writing it because a system that only exists in your head is not a system you can rely on during a bad month, and because somebody else in your household, your team, or your church may need to know where things are, and you may not be in a position to tell them. Write the page. Hand it over. Watch what they cannot understand. That list is your real syllabus for next year, and it is more useful than anything in this document.

Week 51 · Capacity: teach back

Build task. Teach one entire block to whoever will listen — a session, a lunch, a family member, one person — without notes, in fifteen minutes, in your own words. Then write one paragraph on what you could not explain. That paragraph is the actual output of the year.

Drill — "No Notes, Fifteen Minutes." Ten minutes of preparation, then fifteen minutes of teaching with nothing in your hands. If you reach for your notes, stop and start the section again. The standard is not fluency; it is having the structure in your head, which is exactly what the year was for.

Concept. You do not understand a thing until you have taught it badly and survived the questions.

Verse. Psalm 145:4 — "One generation commends your works to another; they tell of your mighty acts."

Teaching. The mechanism in that verse is transmission, and it is worth noticing what is being transmitted: not information about God, but works — what he has actually done, told as something that happened. That is a much stronger form of teaching than instruction, and it is available to anybody regardless of expertise, because everybody has a record of what happened to them. So when you teach back, teach from what you did rather than from what you read. I had three hundred and forty items in my inbox in September and I now have eleven, and here is exactly what I changed is a lesson somebody can use. The four containers of PARA are Projects, Areas, Resources and Archives is the same lesson with all the evidence removed. The second one is what people say when they have learned a vocabulary; the first is what they say when they have learned a practice. Notice which one you produced, and if it was the second, that is not a failure. It is your syllabus for next year.

Week 52 · Flow: tear it down, keep four things, and write the next year

Build task. Four things. Keep four components of your system and name them explicitly. Release everything else, in writing, with a sentence each on why. Write the next fifty-two weeks — a page of fourteen or so lines — using what this year taught you about which blocks were right, which were the wrong size, and which were wrong for you entirely. Date it. Then eat together, and do not talk about systems.

Drill — "Four, Dated." Ten minutes, timer running. Write the four things and date the page. The number four is not symbolic. It is a limit that forces the ranking, and the ranking is the only part of this exercise that will survive February.

Concept. A plan without an expiry date becomes a policy, and a policy is a plan nobody is allowed to revise.

Verse. Psalm 127:1 — "Unless the LORD builds the house, the builders labor in vain. Unless the LORD watches over the city, the guards stand watch in vain."

Teaching. This is the right verse to end on, and it is worth reading both halves. The first says that effort without God is futile. The second says that vigilance without God is futile, which is the harder one, because vigilance is what this entire document has been training. Fifty-two weeks of capture, review, limits, and audits — and the psalm says the guards stand watch in vain. That is not an argument against guarding. The psalm goes on to describe sleep as a gift, and the whole point is that a person who has built and watched faithfully is allowed to stop, because the outcome was never theirs to secure. That is the actual lesson of the year, and it is the one that evaporates if it is not written down. So: keep four things, release the rest, write the next year, date it, and then sleep. The system is a servant. It was always a servant. Next January you will open this document again, and roughly a third of it will be wrong for you, and that is the design working exactly as intended.

Appendix A — Words we use, and what they mean

Nothing in this document assumes prior training. If you are leading and you use a word the group does not know, define it before you use it the second time. Here is the vocabulary, spelled out, grouped by where it came from.

From PARA and Building a Second Brain

• PARA — Four containers that hold everything: Projects, Areas, Resources, Archives. The claim is that these four categories are sufficient for all information in a life, and that any fifth category you invent will be a synonym for one of them.
• Project — A short-term effort with a goal and an end date, or at least an end condition. Move the workshop. File the estate paperwork. Finish the certification. The test is one question: does it end?
• Area — An ongoing standard of performance or responsibility that has no end. Health. Finances. The house. My marriage. My direct reports. An area cannot be completed; it can only be maintained or neglected.
• Resource — Material on a topic you are interested in, with no obligation attached. Resources are where things go when you cannot justify keeping them in Projects or Areas and cannot bring yourself to archive them. This makes Resources the most likely home of the collector's fallacy, and it should be audited more often than anything else.
• Archive — Anything from the other three that is no longer active. Nothing is destroyed; things graduate. An archive with a retrieval record is an asset; an archive without one is a graveyard.
• CODE — Capture, Organize, Distill, Express. Forte's four-verb summary of the whole workflow, and the reason this curriculum is ordered the way it is: the four verbs are the four movements of the year, roughly.
• Progressive summarization — Compressing a note in layers, each layer performed when the note is used rather than when it arrives. Layers: raw capture, bolded passages, highlighted sentences, an executive summary in your own words, and a remix into something new.
• Layer 1 through layer 5 — The five stages above. The discipline is that you do not do all five on arrival, because compression without a consumer is guessing.
• Intermediate packet — A self-contained piece of work that can be assembled into a larger deliverable without being rewritten. A section, a diagram, a checklist, a finished argument. Packets are how work gets done incrementally.
• Twelve favourite problems — A written list of the questions you are actually trying to answer, small enough to hold in mind, used as the filter for what enters and what gets distilled.
• Second brain — An external, personal, trusted store of knowledge and commitments, on the argument that biological memory is for having ideas rather than holding them. The term is Forte's; the idea is much older.
• Map of content (MOC) — A note whose only job is to point at other notes. The smallest useful form is a list of links with a one-line reason for each.

From Getting Things Done

• Open loop — Anything that has your attention and has not been decided about or captured in a trusted place. An open loop is not a task; it is a question you have not answered, and it will keep asking.
• Capture — Getting something out of your head and into a trusted container, with no processing at the point of capture. The separation of capture from processing is the single most important mechanical idea in GTD.
• Clarify (or process) — Deciding, for each captured item, what it is and what, if anything, is to be done about it. This is the step everyone skips and the step where all the value is.
• Next action — The next physical, visible activity that moves something forward. Written as a verb your hands can perform. "Handle the insurance" is not a next action; "find the policy number and write it on the form" is.
• Waiting-for list — Items delegated or promised to somebody else, each with a person's name and a date. Without both, it is not a waiting-for list, it is a hope.
• Someday/maybe — Things you are not committing to and are not refusing. The list has one purpose: to let you say no today without pretending the idea never occurred to you.
• Two-minute rule — If an actionable item takes less than two minutes, do it now rather than tracking it, because the overhead of tracking exceeds the work.
• Four criteria — The four filters for choosing what to do in a given moment: context, time available, energy available, priority. All four, in that order, and the failure mode is using only priority.
• Context — The conditions required to do a task: a place, a tool, a person, a connection. Context lists exist because a list of things you cannot currently do is not a plan.
• Weekly review — A fixed weekly point at which the whole system is brought current: inboxes emptied, lists checked, projects verified, calendar reviewed forward and back.
• Horizons of focus — Six levels of altitude from the runway (current actions) to purpose and principles. Allen's argument is that a mismatch between levels produces either paralysis or frantic activity.
• Trigger list — A list of prompts used during a mind sweep to surface items you have stopped noticing. Its purpose is coverage rather than insight.
• Mind like water — The state a system is meant to produce: complete appropriate response to input, no overreaction, no underreaction. Martial, not mystical.

From lean, 5S, and the shop floor

• 5S — Five disciplines of workplace organization: seiri (sort), seiton (set in order), seiso (shine), seiketsu (standardize), shitsuke (sustain). Applied to knowledge work, the five steps are: remove what does not belong, give everything a place, keep it clean, write the standard down, hold the standard.
• Red tag — A physical marker placed on an item you cannot decide about, with a date and a disposition rule. The red tag exists because people will not sort honestly if sorting means destroying, and because a decision deferred with a date is a different object from a decision avoided.
• Standard work — A written, repeatable description of the best known way to do a task, which is then improved deliberately rather than drifting. A checklist is standard work; so is a template; so is the fixed shape of a session.
• Kaizen — Continuous small improvement, made by the person doing the work, on a schedule, with the change written down. Not a project; a habit.
• Gemba — The actual place where the work happens. The argument for going there is that the process as documented and the process as performed are always different, and only one of them is real.
• Andon — A signal that something is wrong, which anyone is allowed to raise and which stops the line. The rule that protects it is that raising it never gets you punished, and the reason it fails in most organizations is that it does.
• Poka-yoke — Error-proofing: making the wrong action physically difficult or impossible rather than relying on attention. A template with required fields is a poka-yoke. A single door for input is a poka-yoke.
• Muda — Waste. The classic seven: overproduction, waiting, transport, overprocessing, inventory, motion, defects. In knowledge work the dominant wastes are overproduction (documents nobody asked for), inventory (unread piles), and defects (duplicates).
• Kanban — A visual board showing the stages work passes through, with a limit on how much may be in each stage. The board is not the point; the limit is the point.
• WIP (work in progress) limit — A maximum number of items permitted in a stage. The limit is what converts activity into throughput, because it forces finishing.
• Pull system — Work is drawn into the next stage when capacity exists, rather than pushed into it when it is ready. Pull is the mechanism that makes a WIP limit enforceable.
• Cycle time — The elapsed time from work starting to work finishing. Measured, not estimated, and always longer than the person believes.
• Lead time — The elapsed time from a request arriving to being delivered. Always longer than cycle time, and the number that other people actually experience.
• Throughput — How many items are finished per unit of time.
• Little's Law — Average cycle time equals average WIP divided by average throughput. It is exact, it holds for any stable system, and it means that reducing what is in flight shortens delivery time even if nobody works any faster.
• Bottleneck (constraint) — The one stage that limits the whole system's output. Improving anything that is not the bottleneck produces no additional output, which is why effort applied in the wrong place feels like virtue and produces nothing.
• Flow — Work moving through stages without waiting. The measure is the size of the queues, not the speed of the workers.

From the quality tradition

• Defect — A failure of the output to meet the standard. In a personal system the four defects are: unfindable, undated, duplicated, orphaned.
• Root cause — The process condition that makes a defect recur, as distinct from the occasion on which it appeared.
• Five whys — Asking "why" five times about an observed defect until the answer is a process condition rather than an event. The fifth answer is usually a missing rule.
• Fishbone (Ishikawa) diagram — A structured listing of possible causes grouped into categories. Useful mainly because it forces you to look outside your first guess.
• PDCA (Plan–Do–Check–Act) — The improvement cycle: plan a change, try it small, check the result against a measure, and then standardize it or discard it. The "check" step is the one that gets skipped, which is why most systems accumulate untested changes.
• Control chart — A run of measurements with the normal range marked, read for variation rather than for improvement. A stable process wobbles within a range; a trend is a different thing and requires a different response.
• Common cause versus special cause — Variation that is built into the process versus variation caused by a specific event. The distinction matters because the responses are opposites: common cause is fixed by changing the process, special cause by investigating the event. Treating one as the other is the classic management error in both directions.
• Poka-yoke — See above. Repeated here because it is the most useful single idea from this tradition for personal systems.
• Cost of poor quality — The total cost of defects, including the cost of inspection, rework, and lost trust. In personal systems the dominant term is rework: rebuilding the thing you could not find.

From time management and attention research

• Time blocking — Assigning specific activities to specific blocks of time in advance, on the argument that a list does not allocate anything.
• Eisenhower matrix — A sort by urgency and importance into four quadrants. Genuinely useful once, as a sorting exercise, and nearly useless as a daily practice.
• Deep work — Sustained, distraction-free effort on cognitively demanding tasks. The scarcity is not the time; it is the absence of interruption.
• Ultradian rhythm — The roughly ninety-minute cycle of alertness within a day. Used here to justify block lengths rather than as a physiological doctrine.
• Chronotype — A person's natural pattern of sleep and peak alertness. The practical use is to identify your two best hours and defend them.
• Pomodoro — Work in fixed intervals with short breaks. Useful primarily as a measurement device: it tells you how many interruptions a task actually contains.
• Margin (or slack) — Unallocated capacity. The argument for it is that a system with no slack will be consumed entirely by whatever arrives first.
• Parkinson's Law — Work expands to fill the time available. The corollary used here: the deadline determines what is left out, whether or not anyone decides.
• Attention budget — The total discretionary hours available in a week, calculated by subtraction rather than estimation, and allocated explicitly.
• Spaced repetition — Reviewing material at increasing intervals, timed to the point of forgetting. One of the two or three best-evidenced learning techniques in existence.
• Retrieval practice — Recalling something from memory rather than rereading it. The evidence is strong and the practice is unpleasant, which is why it is rare.

Words about the people involved

• Leader — Whoever is running the session. Rotates. Not an expert, not a coach, not in authority over anyone. First names only.
• Build task — The physical change to your own system, made in the session, before you leave the room.
• Drill — The named, timed, repeatable skill practiced in ten minutes.
• Wall chart — The pencil-and-paper chart of one number per week. Not a dashboard; a record.
• Capacity week — The third week of every block, in which the skill is practiced under a realistic and slightly unpleasant load.
• Honest count — A number that includes the embarrassing items. The opposite of a curated metric.
• Re-entry protocol — The written seven-step procedure for coming back after a month of neglect. Kept on a card, in a wallet, for life.

Appendix B — The forty things, in one page

This page exists for the person who has read the whole document once, understood a third of it, and needs to begin on Monday. Read it as a list of claims rather than a list of instructions. Every item is argued somewhere above, and every item has a heading you can find.

On capture

1. Four doors, one room. Any more than four capture points and you have none.
2. Capture and processing are separate acts, done at separate times, and merging them is why people stop capturing.
3. Write it so a stranger could act on it. You are the stranger.
4. An open loop is a decision you have not made, not a task you have not done.
5. The mind is a workbench, not a filing cabinet.

On structure

6. Does it end? If yes, Project. If no, Area.
7. File by commitment, not by topic. A library has no next action.
8. One home per item. Duplicates are the only true defect.
9. Three levels deep, maximum. Depth is not retrieval.
10. The name is the primary key. Everything else is a fallback.
11. Nothing is deleted before it is archived; nothing is archived to avoid deciding.
12. Red-tag what you cannot decide in ten seconds, and date the tag.

On deciding

13. Six answers and no seventh: do, delegate, defer, calendar, file, discard.
14. Under two minutes, do it now. The overhead exceeds the work.
15. If it has a time, it goes on the calendar. If it has a place, it goes on a list.
16. The next action begins with a verb your hands can perform.
17. Context, time, energy, priority. All four, and priority is last for a reason.
18. Be blunt: yes, no, or a specific date. The hedge has a moral character.
19. Count what you discarded. It is the only measure of judgment.

On time

20. Subtract the fixed from the whole before you plan anything.
21. Protect fewer blocks with real force; leave the rest available.
22. Know your two best hours and defend them against your own calendar.
23. Energy is not flat. Schedule the demanding work inside the good hours.
24. Some failures of will are failures of the body. Food and sleep are commands.

On the review

25. The weekly review is the highest-return habit and the first one dropped.
26. Write your own checklist. Twelve items, physical instructions, dated.
27. Have a five-minute floor. A system with no degraded mode is a fair-weather habit.
28. Shrink, never skip. And record the misses, honestly.
29. Stand back far enough, once a year, to see the shape of your life.

On input

30. Guard the door. Count the sources; most people chose fewer than half.
31. The reading pile is a debt, not an asset. Declare bankruptcy and notice that nothing happens.
32. Selectivity is a written criterion, not a virtue. Twelve problems, specific enough to test.
33. Attention cannot be borrowed. Every hour taken from tomorrow is repaid with interest.

On making

34. Compress a note when you use it, not when you save it.
35. Cut the work into packets small enough to finish and self-contained enough to use.
36. Ratio: for every unit in, what went out? Make it visible or it drifts.
37. Fix the date. Negotiate the scope. Never the reverse.

On flow

38. Visualize, limit, pull, measure. The board is not the point; the limit is.
39. Cycle time equals WIP divided by throughput. Reduce what is in flight.
40. There is one constraint at a time. Improving anything else is decoration.

And the four that hold the rest

• Nothing is finished until it is out the door. A thing that is nearly ready is not finished.
• Capacity is created by subtraction or not at all.
• A defect is evidence about your rules, not your character.
• The system is a servant. When it and the person conflict, the person wins — and the point of the review is to know which one you are obeying.

Appendix C — Input discipline: the full case

Block 6 gives the month. This appendix gives the argument, in one place, because it is the part of the curriculum most likely to be dismissed and the part that most determines whether the rest of it produces anything.

The asymmetry nobody accounts for

Input is nearly free. Processing is not. Reading an article costs nothing marginal; deciding what it means, whether it changes anything, where it goes, and whether it was worth the hour costs attention, and that cost is paid later, by a person who did not choose to incur it.

The consequence is a systematic imbalance. Because capture is frictionless and processing is expensive, every system that does not have a gate at the entrance will accumulate a backlog of unprocessed input, and the backlog will grow faster than it can ever be cleared. This is not a discipline problem. It is a queue with an arrival rate above its service rate, and it will behave like one no matter how motivated the operator is.

There are exactly three responses to a queue where arrivals exceed service capacity: increase capacity, reduce arrivals, or accept a growing queue. Most people attempt the first, cannot, and unconsciously choose the third while telling themselves they have chosen the first. Input discipline is the second option, and it is the only one that works.

Four practices that actually reduce arrivals

First: every source must be chosen rather than accepted. An accepted source is one that arrived by default — someone added you to a list, a purchase included a subscription, an algorithm placed it on a screen. Accepted sources account for most of the volume and almost none of the value, and they can be identified in one sitting by asking, of each, did I choose this, and can I say why?

Second: the number of standing sources must be small enough to read in the time available. This is arithmetic, not taste. If a person has forty sources and twenty minutes a day, the de facto policy is that the first three get read and thirty-seven are ignored, which is why the experience of "keeping up" is always accompanied by low-grade dread. Twelve sources read honestly is a better information diet than forty skimmed.

Third: the backlog must be subject to a capacity test. Not is this good? but will I actually consume this within four weeks, given what I know about my own time? Items that fail the test are deleted rather than archived, because archiving a thing you will not read converts a small honest loss into a large dishonest asset.

Fourth: the filter must be positive. "Be more selective" produces anxiety because it has no criterion. The criterion that works is a small set of written questions you are actually working on. Forte's version is twelve favourite problems; the research literature's version is standing questions; a project team's version is a scope statement. All three do the same thing: they allow an incoming item to be classified in about five seconds as relevant or not, without judgment about quality, entertainment value, or general interestingness.

The three objections, answered

"Isn't this just anti-intellectual?" No, and the accusation is worth taking seriously because it is the most common response to this argument. Nothing here says to read less of what you are working on. What it says is that the unselected stream is not scholarship; it is consumption with a scholarly feeling attached. The reading that actually changes a person is concentrated, repeated, and argued with, and none of those three things happen in a feed.

"What about serendipity?" Serendipity is real and it is not random. It is the byproduct of a prepared mind encountering something adjacent to a live question. A person with twelve written problems encounters serendipity constantly, because they can recognize it. A person with no questions encounters volume, which feels similar and is not. The filter does not eliminate serendipity; it is the condition for it.

"I have to stay informed for my job." Sometimes true, and worth testing. Most jobs require deep knowledge of a small area and shallow awareness of a slightly larger one, and almost none require even awareness of the general news cycle. The honest test is the one from week thirty-four: name the last three things you learned from the stream that changed a decision. If you cannot, the stream is entertainment, which is fine, and should be budgeted as entertainment rather than defended as work.

The spiritual argument, stated plainly

Input discipline is the practical form of a theological claim: you become what you attend to. The tradition is unambiguous about this. The psalmist's question is not what a person knows but what a person has set before their eyes. Paul's instruction is to think about a specified list of things rather than about everything. The proverb about guarding the heart is placed above all else, which is a ranking that makes no sense unless the inflow is treated as the primary variable.

The counter-argument — that a mature person should be able to handle anything, that filtering is avoidance — has a surface plausibility and is usually wrong in practice, because it assumes attention is a neutral instrument that the will directs. It is not. Attention is trained by what it is repeatedly pointed at, and the training happens below the level of decision. A person who spends four hours a day in a stream optimized for outrage will become a person who is easily outraged, regardless of what they believe about outrage, and no amount of good intention changes the mechanism.

So the counsel here is not monastic and it is not censorious. It is arithmetic plus a small number of written rules plus one honest number on a wall chart. But underneath the arithmetic is a claim that deserves to be stated without hedging: the shape of your attention is the shape of your life, and it is being shaped right now, by whatever you opened this morning.

Appendix D — Scope, budget, and the arithmetic of attention

Block 11 gives the month, and block 4 gives the week. This appendix gives the arithmetic, because the arithmetic is what makes the argument non-negotiable.

The four constraints

Every piece of work — a report, a renovation, a ministry, a book — is bounded by four variables.

The rule is not that you may choose any three. The rule is that a change to any one changes at least one other, whether or not anyone says so. This is not a management convention; it is conservation. Hours are finite, so a new requirement in October is paid for in November, and it is paid in quality, in sleep, or in someone else's patience. There is no fourth account.

The practical consequence is the only project-management rule most people need: fix the date and negotiate the scope, or fix the scope and negotiate the date, but never pretend both are fixed while quietly spending quality. Silent quality spend is the most common form of professional dishonesty, and it is almost always unintentional, and it is why people who are competent and well-meaning arrive at deadlines with work they are not proud of and cannot explain.

The attention budget, done properly

The arithmetic is simple and almost nobody does it.

Start with 168 hours in a week. Subtract sleep at your actual average, not your intended average. Subtract meals, travel, personal care, and the fixed obligations you cannot move. Subtract the honest overhead of maintaining a household and a body. What remains is your discretionary attention, and it is usually between twenty and thirty-five hours.

Now compare that with what you have committed to. Add up the hours implied by your open projects — the writing, the meetings, the preparation, the administrative tail. If the total exceeds the discretionary number, the situation is not busy; it is insolvent, and insolvency has only three resolutions: reduce the commitments, increase the hours (which is borrowing against health and against other people), or default. Defaulting on a commitment quietly is what most people do, and it is why trustworthy people develop a reputation for unreliability that they cannot account for.

Write the number down. Write it on the same page as the project list. A person who knows their number says yes to fewer things and finishes them, and appears, from outside, to be unusually reliable and unusually calm.

Why slack is not laziness

A budget with no margin is a budget that will be broken by the first unexpected event, and the first unexpected event is always this week.

The mechanism is worth stating because it is counterintuitive. If every hour is allocated, then any interruption must be paid for by displacing something, and the thing displaced is whatever has the weakest immediate claim — which is almost always the work that is important and not urgent: the review, the maintenance, the planning, the difficult conversation. Over a year, a fully-allocated calendar systematically destroys exactly the activities that keep a system alive, and it does so invisibly, one displaced hour at a time.

So the rule is: three empty half-days per week, minimum. Not because you will rest in them. Because they are the only thing that lets you absorb the week as it actually arrives, and because a calendar with no slack does not produce more output; it produces a person who is always behind and cannot say why.

Scope creep, and how to see it coming

Scope creep is not usually somebody adding requirements. It is usually the gradual discovery that the thing is more interesting than expected, and it happens in increments small enough that each one is defensible.

Three defenses, in order of effectiveness.

One: write down what is out. A scope statement that lists only inclusions is a wish. The exclusions are what make the inclusions meaningful, and they are what you point at when the conversation arrives, which is much easier than arguing from memory.

Two: decide the smallest useful version first. Write one sentence describing the minimum that would still be useful to the recipient. Anything beyond that is an improvement rather than a requirement, and improvements can be cut at the end without destroying the deliverable. Projects that fail usually had no minimum defined, so there was no state short of "perfect" that counted as done.

Three: put the change on the record. When a requirement changes, write it down with its cost. Not as a complaint; as arithmetic. Adding this means the date moves by a week, or something else comes out — which? The question is answerable, and asking it is the entire skill.

The personal version of the triage problem

In an emergency department, triage exists because demand exceeds capacity permanently and the only question is who is treated first. Every person with more than a dozen open commitments is running a triage system without having designed it, and the design that emerges by default is whoever asked most recently and most loudly.

The alternative is a written rule. The rule does not need to be sophisticated. Three categories, applied in order: must happen this week or something breaks; should happen and will be scheduled; will not happen and is now explicitly released. The third category is the one that does all the work, and it requires saying the word no to a real person, which is why it is the one people skip.

Note what the third category is not. It is not a "someday" list, which is a deferral dressed as a decision. It is a release, spoken out loud, with a date. The difference in how it feels is the difference between carrying something and putting it down.

Appendix E — The reading list, and the rule about reading lists

The rule: the list is one page, it does not grow during the year, and reading about systems does not count as doing the work. This is rule nine in section 9 and it exists because the literature on personal knowledge management is large enough to consume a life. If you read all six of these books in a year you have read more than almost anyone who talks about this subject. If you read them twice you have read enough.

Start here, and only here, for the first six months.

• David Allen, Getting Things Done. The mechanics of capture, clarify, and review, and the argument that the mind is for having ideas rather than holding them. Read it once for the parts you will actually use: the processing question, the two-minute rule, the four criteria, and the weekly review. You do not need the horizons chapter in year one.
• Tiago Forte, Building a Second Brain. The four containers, the four verbs, and progressive summarization. Read it for PARA and for the distinction between collecting and making. Read it critically; the book is more confident about the benefits than the evidence supports, and the parts about tools age badly.
• Daniel Kahneman, Thinking, Fast and Slow. Not a productivity book at all, and the most useful thing on this list. Read the sections on the planning fallacy and on what intuition is and is not good for. It explains why your estimates are wrong in a specific direction.
• Cal Newport, Deep Work. The argument for sustained attention as a scarce and trainable capacity, and a set of environmental designs that make it possible. Skip the parts that read as professional advice and keep the parts about interruption cost.
• Eliyahu Goldratt, The Goal. A novel about a factory, and the best available explanation of why improving anything other than the constraint produces nothing. Read it for the idea, not the plot.
• W. Edwards Deming, Out of the Crisis. Dense, dated in places, and the source of the single most humane idea in this document: that most defects are process defects and that blaming people for them is both unkind and ineffective. Read the fourteen points and the chapters on variation. Skip the rest until year three.

For year two, if you are still doing this.

• James P. Womack, Daniel Jones, and Daniel Roos, The Machine That Changed the World. Where lean actually came from, and why it was a moral argument before it was a management technique.
• Masaaki Imai, Gemba Kaizen. The 5S material, at source, with the shop-floor detail that the popular summaries strip out.
• David J. Anderson, Kanban. The WIP limit, cycle time, and Little's Law, written for knowledge work rather than for factories.
• Peter Drucker, The Effective Executive. Short, old, and better than almost everything written since. Particularly the chapters on time and on deciding what not to do.
• Scott Berkun, Making Things Happen. Project management without the ceremony, and unusually honest about the politics of scope.

What is deliberately not on this list. No book that is primarily about note-taking applications. No book whose central claim is that a particular method will change your life. No book published to accompany a course that is being sold. No book about morning routines. If you find that you are reading a fourth book about productivity before you have finished a single project with the system you already have, the correct response is not to finish the book.

Appendix F — Where the design comes from

This is a short note on why the shape is what it is, so that a future reader can change it on purpose rather than by accident.

The four-verb spine — capture, organize, distill, express — is Forte's, and the ordering is the reason this curriculum runs a year rather than a weekend. The temptation is to teach the interesting verb first. Expression is the interesting verb. But a person who has not collected cannot sort, and a person who has not distilled has nothing worth expressing, and the whole sequence collapses if you start where the energy is. That is why the year opens with January's thankless collection and does not reach output until week twenty-nine.

The mechanics are Allen's, and the mechanics are not the point. GTD is often dismissed as a list system, which is like dismissing a workshop as a set of tools. Allen's actual contribution is the observation about open loops and the insistence on a fixed weekly point of review. The weekly review is the single highest-leverage hour in this entire document, and it is Allen's, and it is worth saying so plainly because his book repays reading once.

The discipline is Toyota's, and the objection to importing it is worth answering. The standard objection is that a factory is not a mind. This is true and it does not matter, because the parts of lean this document borrows are not about manufacturing. They are about the conditions under which a person does reliable work: a place for everything, a written standard, a visible signal when something is wrong, a limit on work in progress, and small improvements made continuously by the person doing the job. Those are properties of any workspace, physical or otherwise, and the shop floor learned them expensively over seventy years. Borrowing them is not a metaphor; it is a transfer.

The epistemology is Deming's. The claim that most defects are process defects, that variation must be understood before improvement is attempted, and that blaming the individual is the characteristic error of management — all of it is his, and all of it applies to a person managing themselves. The reason it matters so much here is that the characteristic emotional failure of anyone learning a knowledge system is moral: they conclude they are disorganized, undisciplined, not the type. Deming's answer is that they are running a process with a missing rule, and that the fix is a rule. That is a much better diagnosis and it is also a much kinder one.

The project-management layer is deliberately thin. Triple constraint, definition of done, one owner per action, sequence, buffer, honest status reporting. That is roughly a tenth of what a project manager knows and it is more than an ordinary person needs. The rest of project management exists because organizations have many people in them; a person running their own work has one, and most of the ceremony would be theatre.

The spiritual frame is not an addition to any of this. It is the load-bearing wall described in section 5, and the test of whether it is genuine is whether it changes anything practical. It does, in three specific places: it produces a stopping condition, so the system cannot consume the life it serves; it produces the honesty requirement in the wall chart and the re-entry protocol, because a system you are lying to is worse than none; and it produces rest as a command rather than a reward, so falling off for a month is a known state with a written protocol rather than a moral catastrophe. If the frame were decorative, those three things would not be in the document. They are in the document because they are the only reason a system survives past its second year.

What this design deliberately does not claim. It is not evidence-based, in the sense that no controlled trial has been run on this curriculum and the underlying practices come from contexts that differ from yours. It does not make anyone smarter or more creative. It does not substitute for thinking; it removes a class of friction that prevents thinking. It does not produce career outcomes, and anyone who tells you that capturing notes will advance your standing is selling something. It does not reduce anxiety in any clinical sense. It does not improve with a better tool, and it does not fail because of a worse one.

What it does is claim that a person who can capture reliably, find things quickly, decide about them clearly, finish them honestly, and say out loud how much is in flight — is free, in a small and specific way, to do the thing they are actually for. That is the whole offer. It is small, it is checkable, and it takes a year.

Appendix G — The fifty-two drills, and what each one needs

Every drill in this document is named so that it can be said out loud, taught to somebody else, and recognized when it is being done badly. This table exists so that a leader can run a session without re-reading the week, and so that a person can find a drill they half-remember.

Almost every drill needs the same four things: a timer, something to write on, something to write with, and your own real material. The fourth is not optional and is the most common failure. A drill performed on a downloaded example teaches nothing, because the whole point is that your material is messier than an example.

The three drills worth repeating out of season, when a week has gone wrong. The Ten-Minute Version (week 18) because it is the floor of the whole system. The Untested Protocol (week 47) because it is the recovery. Where the Pile Is (week 40) because it finds the constraint, and the constraint is where all the leverage is.

A note on the timer. The timer is not there to make the drill fast. It is there to prevent the drill from becoming research. Every one of these exercises can expand to fill an afternoon, and the expansion is always the same move: instead of deciding, you begin investigating. Ten minutes is long enough to decide about more items than most people believe, and short enough that investigating is impossible. If you finish early, stop. Do not continue.

Appendix H — Running this as a group

This document is written for one person, and it works better with three to eight. The reasons are the same reasons a weekly study group beats a stack of books, and they are worth stating so that the group form can be defended when it becomes inconvenient.

Accountability is not the main benefit. The main benefit is normalization. A person alone with a backlog of three hundred and forty items concludes something about their character. Three people with the same backlog conclude something about backlog. That reframing is worth more than any amount of encouragement, and it cannot be produced by a book.

A group also produces the honest numbers. People curate for themselves and not for a room. A number that has to be said out loud in front of six people is a different number from a number that is written in a private app, and the difference is the entire value of the wall chart.

The four rules for a group

Rule one: rotate the leader. Any regular participant can lead, using this document, and a person who has led twice should be teaching somebody else to lead. Rotation prevents the group from becoming one person's project, prevents burnout, and produces the only real test of whether anyone learned anything, which is whether they can run it.

Rule two: every session is complete. No session assumes the previous one happened. A person who attends only week thirty-four leaves with a working board, one number, one idea, one verse, and one commitment. The teaching says the prior context out loud in its first two sentences, which is why it is written down rather than summarized.

Rule three: nobody's screen is shown without specific spoken permission. The specific words matter. Can I show this? is not the rule; may I show this one screen to this room right now? is the rule. Inboxes contain other people's names, diagnoses, salaries, and confessions, and a shared screen is a publication. Say this before the first session and again whenever somebody new arrives.

Rule four: the room does not become a support group. This is the hardest one, and it matters because these sessions touch real distress. People will discover during a mind sweep that they have been carrying something serious, and they will want to talk about it. Let them talk, briefly, in the circle, and then get back to the drill. The document is not therapy and the leader is not a counselor. If a person discloses something that needs care, the leader's job is to say that is more important than this and here is who can help — privately, afterward — and then to keep the meeting's shape.

The ninety-minute version

Seventy-five minutes is the base. With more than six people, add fifteen minutes and split them as follows: three minutes to the arrival and dump, seven minutes to the circle, and five minutes to the build, because the leader's circulation time is what scales with headcount and it is the part that cannot be compressed.

What the leader actually does

Before. Read the week's entry once, out loud if possible, and decide which single sentence of the teaching is the one that must land. Not the whole teaching — one sentence. A leader who tries to deliver all of it will deliver none of it.

During the drill. Demonstrate twice, slowly, on your own material rather than an example, and let them see the mess. Then set the timer and be quiet. The most common leadership error in this curriculum is talking during the drill, which converts a practice into a lecture and destroys the ten minutes.

During the build. Circulate, answer one question at a time, and never correct in public. If two people have the same problem, that is a teaching moment for the whole room and should be named as such without naming the person.

At the measure. Write the number without commentary. If a number is bad, say nothing. The chart does the talking over twelve weeks, and a leader who reacts to each reading trains people to curate.

At the verse. Read it twice. Then be silent for two full minutes, which will feel like six. Then invite one sentence each. Do not summarize what people said, do not synthesize, and do not resolve it. The verse is the teacher and the leader's job is to keep the room quiet long enough for it to be heard.

At the close. Every person states a commitment with a time attached. Write nothing down. The act of saying it aloud is the mechanism, and a leader who becomes the record-keeper removes the mechanism.

When the group should not meet

There are three conditions worth writing down in advance, in the same spirit as the weather calls in section 10.

• When the room has become a performance. If people are bringing curated systems rather than real ones, stop the build for a week and run week forty-seven on everyone's messiest area, including the leader's. Nothing restores honesty faster than the leader showing their worst folder.
• When attendance halves twice in a row. That is not a scheduling problem; it is a signal that the content has drifted from what people actually do. Ask, in the circle, what the last four weeks were for, and rewrite the next four based on the answer.
• When the leader is the only one who has read anything. This means the rotation is not working. The fix is not more preparation; it is assigning next month's blocks to four different people and letting them teach badly.

Appendix I — Troubleshooting: the hard cases

The following are the situations that come up in every group, in the order they usually arrive. Each answer assumes the rest of the document, so read the referenced section if the answer seems abrupt.

"I have four hundred items in my inbox and I want to quit." Do not process it. Do a mind sweep (week two) and then a triage (week forty-seven). Processing four hundred items one at a time is how people lose three weekends and their enthusiasm in the same month. Triage means: sort into must, should, and release, in one sitting, with a timer, and accept that the release pile will be large. The backlog was never the problem. The absence of a gate was the problem.

"My employer requires one set of tools and I have another for myself." This is the most common structural problem in the whole document and it has a clean answer: the boundary is a container, not a category. Your work system is an Area. Everything the employer mandates stays inside that Area's container and follows its rules, and you do not fight it. What you own is the seam: one note per project that says what the project is, what the next action is, and where the real material lives. That seam is three sentences and it is the only thing you need to maintain across the boundary.

"I don't have a team, so delegation, waiting-for, and status reports don't apply." Half true. Waiting-for still applies, because you are always waiting on somebody — a reply, a shipment, an approval, a spouse. Most people's stalled items are waiting on another person and are not tracked as such, which is why they sit in a task list looking like failures. Status reports apply too, with the recipient being future-you at the review, and the honest version is still the same five lines.

"I have a diagnosable attention problem." Nothing here is treatment and this document is not qualified to give it. What can be said without qualification: every technique in this document that reduces the number of decisions in front of you is likely to help, and every technique that relies on sustained unaided concentration is likely to need adaptation. The specific adaptations worth trying are the five-minute floor in week nineteen, the phone-only vault in week thirty-five, and the visible board in week thirty-seven — because externalized state does not require working memory. If you are not being treated, that is a different conversation and a more important one.

"I'm a student, or retired, or between jobs. The project/area distinction doesn't map." It maps; you are just calling the areas something else. Being a student is an area. Finish the thesis chapter is a project. Retired is not an area at all — it is a fact, and the areas are what you are now responsible for maintaining, which for most people gets clearer rather than vaguer. Between jobs is genuinely different in one way: your Areas have shrunk, and the temptation is to fill the space with input. Resist it with block six, because a job search is a project with an end and a defined next action, and the reading you are doing instead is not one.

"I did the whole system and then abandoned it for four months. Do I start over?" No. Run the re-entry protocol (week forty-seven). Nothing is lost, the archive is intact, and the only thing that has changed is that you now have evidence about which parts of the system did not survive contact with your real year. That evidence is the most valuable thing you own, and restarting throws it away.

"My problem isn't capture, it's that I never do the things." Then capture was never your bottleneck and neither is your system. Run week forty — find the column with the oldest cards. If the pile is in doing, the constraint is time and the answer is block eleven's scope conversation. If the pile is in to do and the board does not move, the constraint is the absence of a fixed decision point, which is block five. Note what is not on that list: motivation. Motivation is a symptom of an unclear next action far more often than it is a cause.

"I keep reorganizing instead of working." That is the characteristic failure of this curriculum and it is worth naming precisely: reorganization produces the sensation of progress with none of the exposure. The intervention is a rule with a date. Pick one project, define done in one checkable sentence, and forbid yourself any structural change until it ships. If you break the rule, write down what you changed and why, and put that page in the front of the system where you will see it next time. Week forty-four is the week for this, and most people need it twice.

"How do I know when to stop?" Three stopping conditions, and any one of them is sufficient. The system costs less than it returns — you can feel this and you should trust it. The review is happening and the output is not — that is the collector's fallacy in its mature form and block eight is the answer. The work has displaced something it was meant to serve — this is the condition that section five exists for, and when it happens, the correct response is not to optimize. It is to stop, keep four things (week fifty-two), and go and do the thing the system was for.

Soli Deo gloria.