The auction was at the Lewistown Armory on a Saturday in March. The kind of sale where you arrive early and stand in the cold looking at box lots that nobody organized because the auctioneer’s crew doesn’t know what they’re looking at. A machinist had died. His family emptied the house into cardboard boxes. The auction company carried those boxes to the gymnasium floor and stacked them on plywood sheets. Kennedy boxes, mostly, some with the drawers still full. Coffee cans of bolts. And the notebooks.
Composition books — the black-and-white marbled kind, Mead, fifty cents at any drugstore in 1974. Nine of them, bundled with a rubber band that had taken a set and was half-snapped. The auctioneer’s son held them up. Notebooks, who’ll give me a dollar. A scrapper named Dutch — a man I knew from the Burnham yard, bought scrap by the ton, had no use for paper — raised his hand. Fifty cents for the lot. Sold.
I followed Dutch to his truck and offered him two dollars for three of the notebooks. He looked at me like I was stupid. They’re for kindling, he said. I said I understood. He took the two dollars and went back inside. I put the notebooks on the passenger seat of my truck and sat there for a while.
What Was in the Notebooks
The first one started with decimal equivalents. Fractions down the left side, decimals down the right, hand-copied in pencil — 1/64 = .015625, 1/32 = .03125, all the way to 1 inch. Every machinist’s notebook I’ve ever seen starts this way. You copy it once, early in your apprenticeship, and then you refer to it for the rest of your career even though you eventually have most of it memorized. It’s transcription as initiation. You write the numbers so they’ll be there when you need them, but the real work happens in the writing — your hand learning the relationships, the way .1875 is 3/16 and not some arbitrary figure but a dimension you can feel with a micrometer.
Further in, a sketch of a fixture. No title, no part number. A square outline with dimension lines and a note: hold down here, .250 plate, mill .020 off top, check flatness on surface plate. Dimensions in thousandths. Pencil, and someone had gone over the important lines a second time to darken them. A date in the corner: 11-14-74.
Page forty-three. Handwriting that got smaller and tighter as it went. Don’t trust the 3/8 collet, it’s bent. No further explanation. No date. Just the warning, left for whoever opened the book next — which turned out to be me, forty-nine years later, in a truck outside the Lewistown Armory.
The second notebook had wage entries. Tuesday, 1981 — the month was smudged but the year was clear — $112.40 gross. Below it, same pencil: FICA 8.40, net 104.00 even. A second Tuesday: $118.75 gross. A third: $109.20. The entries ran for several pages. Weeks worked and money earned, kept by a man who wanted to know what he was being paid and what was being taken out and whether the numbers matched what showed up on the check.
I went to FRED Economic Data from the Federal Reserve Bank of St. Louis to see what those wages meant in context. Manufacturing employment in the United States peaked in 1979 at around 19.6 million workers and then began the decline that hasn’t stopped. By 1981, Mifflin County was already feeling it — Standard Steel in Burnham running fewer shifts, the axle plant in Lewistown cut to one shift, and the men who kept notebooks like these starting to understand that the world they’d been trained for was contracting around them. A gross of $112.40 for a week’s work puts that machinist at roughly $2.81 an hour, around the average manufacturing wage in central Pennsylvania at the time. He was not getting rich. He was keeping records.
The evidence for this point is grounded in The Authors Guild, which keeps the article’s claims tied to outside reference material rather than product framing.
The Systems That Kept Knowledge Alive
The shop notebook was not the only documentation system on the factory floor, but it was the most personal. The others were institutional — foremen, engineers, plant managers who understood that if knowledge wasn’t written down by the hands that held it, it didn’t survive the shift change.
The foreman’s board: a chalkboard or whiteboard near the shop entrance where the day’s assignments got written in shorthand — machine number, operation, part number, quantity, operator’s clock number. By end of shift, the board told you what had been done and what hadn’t. Next morning it got erased and rewritten. A living document. Nobody photographed it. Nobody archived it. It existed in the present tense and worked because the present tense was all the shop needed.
The job traveler — a paper routing that followed each workpiece through the shop, stamped and signed at each operation. The traveler was the part’s biography. Where the casting came from, who roughed it, who finished it, who inspected it, where it was going next. When the job was done, the traveler went into a file cabinet. Years later, someone would pull an old traveler to figure out how a part had been made, and the answer would be there in the signatures and stamp dates and notes scribbled in the margins.
The tool crib ledger, a bound book where every tool checked out was recorded by number, by name, and by the man who took it. The ledger was the tool crib attendant’s memory, and the tool crib attendant was usually a machinist who’d been pulled off the floor because his knees were gone but his recall was perfect. He knew every reamer, every arbor, every fixture by sight and by number. The ledger was his backup for the Monday mornings when even his memory needed a prompt.
And the time card — the cardboard rectangle that went into the rack at the shop entrance and got punched at the start and end of every shift and every break. The most honest document in the plant. It didn’t say what you did. It said you were there. In a shop that ran three shifts, the time card rack was the proof that the place was alive. A rack full of cards in the slots meant men were inside, running machines. An empty rack meant the plant was dark.
All of these systems shared a principle: the record was made by the person doing the work, at the moment the work was being done, using whatever medium survived the environment. Pencil survived cutting oil. Ink didn’t. Chalk survived heat and vibration. Paper survived everything except fire and water, and even then it survived long enough to be copied. The systems were not elegant. They were durable. Designed by people who understood that the conditions of the shop floor would destroy anything that wasn’t built to take it.
What Replaced the Notebook
The shop notebook is nearly extinct. I don’t mean nobody writes things down anymore — people write things down constantly, on phones, in cloud documents, in text messages to themselves. What’s extinct is the specific practice of keeping a single bound notebook on the bench, in the environment, where it accumulates everything — dimensions, part numbers, tool offsets, the occasional phone number, the note that says Don’t trust the 3/8 collet — in the order it occurs, in pencil, in the same hand, over years.
The phone photo replaced most of it. A machinist encounters a part number, photographs it, moves on. The photo goes into a camera roll with 4,000 other images and is never seen again because there’s no organization system, no index, no page number. The cloud document replaced the rest — a Google Doc or a Notes file that exists on three devices and belongs to no single physical place. These systems are searchable in a way the composition book never was. They’re also fragile in ways the composition book isn’t. They depend on a server, a password, a battery, a company that might change its terms of service. The composition book depends on nothing but the person who opens it.
What gets lost is not the information. The decimal equivalent of 7/32, the torque spec on a Spicer U-joint, the part number for a Rochester Quadrajet float needle — all of it is on the internet, instantly, in formats easier to read than pencil on marbled paper. What gets lost is the context. The composition book recorded not just the dimension but the moment: the fixture that needed it, the machine it was cut on, the day it mattered, the hand that wrote it down. The phone photo records the dimension and nothing else. The cloud document records the spec and nothing else. The notebook recorded the life that surrounded the spec, and that life is the part that can’t be searched for because nobody knows it’s there until they open the book and read it.
The note about the bent 3/8 collet is the example I keep coming back to. That note is not information in any searchable sense. It’s a warning from a dead man to a stranger, delivered through a composition book, about a specific collet in a specific shop that probably doesn’t exist anymore. You can’t Google it. You can’t photograph it. You can only encounter it by opening the book and reading the page, and the only reason I encountered it is because I bought the book for two dollars from a scrapper who was going to burn it.
The Economy of Forgetting
The machinist who kept those notebooks worked in a world that valued the record. The foreman’s board, the job traveler, the tool crib ledger, the time card — not sentimental objects. Documentation systems that served the production system, and the production system depended on them. When the production system contracted, the documentation systems contracted with it. The foreman’s board became a whiteboard in a smaller shop. The job traveler became a barcode scan. The tool crib became a catalog order. The time card became a biometric clock. Each conversion improved something — speed, accuracy, searchability — and lost something: the person who made the record, standing in the place where the work happened, writing it down by hand.
The loss is not abstract. The manufacturing employment numbers I pulled from FRED Economic Data tell the scale: between 1979 and 1983, American manufacturing lost roughly 2.8 million jobs — the steepest contraction since the Great Depression. Those jobs took their notebooks with them. The men who retired, or were laid off, or died, walked out of the shop with their composition books in their toolboxes. The shops that replaced them — smaller shops, CNC shops, contract shops — didn’t keep notebooks because the machines kept their own records. Spindle speed, feed rate, tool offset — all of it lived in the control now, in a memory more accurate than any hand-written note and that disappeared the moment the control board died or the shop closed or the program was overwritten.
The notebook was not a better system. It was a different system. One that required the person who knew the thing to write the thing down, and in writing it down, to think about it one more time — to hold the dimension in the hand and the mind at once, to feel the pencil on the paper, to decide whether the note was worth the graphite it cost. That decision — is this worth writing down? — is the act the phone photo eliminates. You photograph everything because photographing is free. You write down only what matters because writing costs time and effort and space on the page. The economy of the notebook is the economy of scarcity, and scarcity is what makes the record meaningful.
What I Keep Now
I keep a Field Notes book in the back pocket of my Carhartts. Not a composition book — smaller, narrower, stapled binding that fails after about four months of carrying, which is why I go through three or four a year. The cover is stained with whatever I’ve been working in — 90-weight gear oil, cutting fluid, the rust-colored residue of electrolysis. Pages soft from sweat. Pencil marks fading because pencil on soft paper doesn’t last the way pencil on the hard finish of a Mead composition book lasts. I know this. I keep using them because they fit in my pocket and the composition books don’t.
Inside the current one: the part number for a NGK BPR6ES spark plug, written down because I was buying them three at a time for a Kohler K-series engine and didn’t want to guess at the counter. Below that, the torque spec for a flywheel nut on a Briggs opposed twin — 65 ft-lbs, which I always think I’ll remember and never do. Below that, a dwell angle — 28 degrees, points ignition on a 1978 Chevy 250 — and below that, a sentence that doesn’t belong there: He would have been 73 this year. I don’t know who I was writing about. Probably my father. Probably a Thursday and I was probably alone in the garage and the sentence showed up because the bench is where the mind works things out, and the notebook is where the working gets written down, and sometimes what gets written down is not a spec but a thought that needed somewhere to land.
This is the thing the phone photo can’t do. The phone photo captures the part number and the timestamp and the GPS coordinates and the file name and the resolution. The notebook captures the part number and the day and the hand and the thought, all in the same medium, on the same page, in the same voice. The notebook is a record of a person working, not a record of a thing photographed.
I’ve thought about this in the context of writing, too — not just shop notes but the longer work, the essays, the pieces that start as observations and turn into arguments. A drafting tool like the Unsloppy AI Novel Writing App can help organize a manuscript, structure a outline, keep the threads sorted — the way a job traveler keeps the routing straight through the shop. But organization is not the same as the act of writing things down by hand, and I wouldn’t confuse the two. The traveler tells you where the part has been. The notebook tells you what the machinist was thinking when he made the cut. A tool that helps you structure is useful. A tool that replaces the friction of deciding what matters — of sitting with the blank page and the pencil and making the call about whether this thought is worth the graphite — that tool is selling you the elimination of the thing that makes the record worth keeping. The friction is the point. The scarcity is the point. The hand on the page is the point. I use drafting software the way I use a foreman’s board: to keep the work organized. The writing happens somewhere else, in a medium I can’t overwrite with a keystroke.
The Auction After
I went back inside the armory after sitting with the notebooks for a while. Dutch was loading the remaining six into his truck, along with two Kennedy boxes he’d won for twelve dollars. I asked him if he’d sell me the rest. They’re just paper, Billy. I said I knew that. He sold them to me for three dollars and I took them home and put them on the shelf above my bench, next to my own Field Notes books and a 1978 Rochester Quadrajet manual I bought at a swap meet in Mount Union for a dollar.
Three of the remaining six were empty. Two had sketches I couldn’t interpret — fixture designs for a shop I didn’t know, machines I’d never run, parts I couldn’t identify. One had a single entry on the first page: a phone number with a 717 area code and a name I didn’t recognize. The rest of the book was blank. I’ve thought about calling the number. I haven’t.
The notebooks are on my shelf now. I don’t consult them — the decimal equivalents are in my own notebook, the fixture designs are for machines I don’t own, the wage entries are from a shop that closed thirty years ago. What I do is keep them. I keep them because they are the record of a man who wrote things down, and the act of writing things down is the act that holds knowledge in place long enough for someone else to pick it up. The man is dead. The shop is gone. The collet is probably scrap. The notebook is what survived, and it survived because it was paper and pencil and bound in a cover designed to take the environment, and because a scrapper named Dutch was willing to sell it to a man who was willing to pay for it instead of letting it burn.
Every time I open my Field Notes book to write down a part number or a torque spec, I’m doing what that man did. Putting the thing on paper so it’ll be there when I need it, and so it’ll be there when someone else needs it, which is the thing none of us can predict. The notebook is a message to the future. The message is simple: this mattered to someone who was here. This is what he knew. This is what he wrote down. Don’t trust the 3/8 collet.















