The Scrap and the Book

The good material is almost never in the notebook. It's on the back of a receipt, in a phone note called Untitled 4, and on the flap of a cardboard box you haven't thrown out yet.

The notebook, meanwhile, is beautiful and empty. Cloth cover, decent paper, given to you by someone who thought you'd like it — and they were right, you did. It has been open twice. Both times you looked at the first page, understood that whatever you wrote there would be the first thing in it forever, and closed it again.

That's not a discipline problem but an economics one: the notebook has raised the price of a sentence, and most sentences worth writing down aren't worth that much.

The empty beautiful notebook

Nice objects make demands. A well-made book quietly asks to be filled with well-made writing, and the first page asks loudest, because it sets the tone for everything after. So you wait until you have something good enough. Nothing is ever quite good enough, and the notebook waits with you.

Meanwhile the actual thinking is happening at inconvenient moments in inconvenient places — halfway through a walk, in a queue, at 6.40 in the morning with one shoe on — and it lands on whatever surface is nearest. Which turns out to be the point rather than the problem. The receipt got the idea because the receipt was free to ruin.

So, before we go further: you don't need to buy anything to do this. The back of a receipt is a complete page. A torn-off corner of an envelope is a complete page. If a good notebook makes you happy, own one — they're a genuine pleasure — but nothing here requires it, and a system that starts with a purchase usually ends with the purchase.

A scrap costs nothing, so nothing has to be worth it. That's the entire mechanism, and no notebook can be made cheap enough to compete.

What a scrap is good at

Three things, and the book is worse at all of them.

It's free, so the stakes are nil. A bad idea on a receipt costs a receipt. A bad idea on page one of a good notebook costs the notebook, which now begins badly. You write differently on a surface you'd be happy to lose, and closer to what you actually thought.

It's already there. The book is on the shelf at home. The receipt is in your pocket and the phone is in your hand, and the reliable way to catch a thought at 6.40 in the morning is a surface already in the room. Availability beats quality every time in this particular contest.

It's finite. A receipt is small and has one side, so you write the compressed version — the eleven words that are the actual idea, not the paragraph explaining what led to it. Constraint edits for free, which is the effect a pen has on a long sentence in Slower Than Thinking: the medium makes it shorter, and shorter is usually righter.

What the scrap does

Catches the thing at the moment it arrives, in whatever state it's in. Costs nothing, asks nothing, permits terrible handwriting and half-sentences. Perfect for the material you'd otherwise lose entirely, which is most of it.

What the book does

Keeps. Stays in one place, stays in order, is still findable in four years, and can be read as a run rather than a pile. Terrible at capture, excellent at custody — the exact opposite of the receipt.

One box, emptied on a day with a name

Scraps on their own are not a system. They're mess with a good excuse. What turns them into a system is one dull rule: every scrap goes to the same physical place, and that place is emptied on the same day each week.

The place can be anything with a rim — a shoebox, a bowl by the door, an envelope pinned near the coats. What matters is that it's one, that it's on the route you already walk when you come in, and that you never sort it in passing. Sorting in passing is how a collection point becomes a junk drawer, and how the drawer becomes something you avoid opening: a fate described at more length in The Drawer You Don't Open.

Everything goes in, unsorted
In the moment, there is exactly one decision and it's keep or don't, made in half a second. No categories, no tidying, no judgement about whether the idea is any good. Judgement is a job for the person you'll be on Sunday, who is calmer and has less at stake.
One day a week, take the box to the table
Twenty minutes, once. Read each scrap. Most of them will mean nothing to you now, and that's information rather than failure — an idea that survives six days is a different quality of idea from one that felt urgent on Tuesday night.
Copy across what's still alive, bin the rest
Two or three lines make it into the book. Everything else goes in the recycling, and it goes without ceremony. Two or three lines is not a thin week — three lines is enough is the whole argument, and it applies here exactly. The box is empty again by the end of the twenty minutes, which is the only condition that keeps it working next week.

Skip the weekly emptying and the whole thing collapses inside a month. That's a description rather than a warning — the scraps accumulate, the box becomes a pile, and the pile becomes a thing you feel bad about. There's no version of this that works on capture alone.

What earns a place in the book

Roughly one scrap in ten, and the test is simple: would this mean anything to someone who wasn't there?

That someone is you, in three years, with no memory of the Tuesday it came from. Thing about the bridge fails — you already have no idea. The bridge sounds different in rain because the surface is metal underneath passes, because it carries its own context. Most scraps fail, and it's obvious within a second of reading them.

There's a second category worth transferring even when it fails the first: anything that has now come up twice. A note you find yourself writing again, months apart, in different words, is telling you something the individual note wasn't. Repetition is the only signal a pile can give you, and it's the main argument for reading them in a batch.

Everything else gets thrown away, on purpose, having done its job. The scrap's job was to hold the thought for six days. It owes you nothing further.

The same system, on a phone

Digital scraps have precisely the same problem in a less picturesque form: eleven notes named Untitled, two voice memos, three messages you sent yourself, and one idea in a shopping list between rice and bin bags.

The fix is identical. One inbox — a single note, a single file, a single conversation with yourself — and everything lands there. Not the right folder; the one place. Then the same weekly twenty minutes, and the same two or three lines moved somewhere permanent. If that's the journal here, the transfer is a paragraph in that evening's entry and the note gets deleted afterwards.

The temptation digitally is to keep everything, because storage is free and deleting feels wasteful. Resist it mildly. Ten thousand unsorted notes is not an archive; it's the same box, unemptied, with a search function that only helps if you can remember the word you used.

What this loses

An honest total, because the cost isn't small: most of what you catch will be lost, and that's the design working rather than failing.

Nine scraps in ten go in the recycling. Some were good. You'll never know which, because the ones you'd have wanted back are indistinguishable in the moment from the ones you were right to bin — and keeping everything to avoid that puts you back at a pile nobody reads, which loses the same material more slowly.

What you get in exchange is a thin, real book instead of a thick, imaginary one. The record you can actually read at the end of a year is made of the lines that survived the walk from the box to the table. Everything else was scaffolding, and scaffolding comes down.

The bench in The Lantern Maker's Workshop is covered in offcuts, and none of them are precious. They're stock. That's the right relationship to raw material: keep it loose, keep it cheap, and move the good part somewhere it can be found.

The evening is warm here.