1. A word that keeps showing up
Lao deng. Literally, old fellow.
That is not what it means any more. Used the way it gets used now, it lands closer to dinosaur: a man with seniority and grey hair who has stopped keeping up. Set in his old ways, blind to anything new, and — the part that actually stings — still trying to lecture everyone younger than him.
It turns up in product reviews, in company post-mortems, in verdicts about decisions other people made. It has become the fastest way to end an argument. Whatever the merits, the other guy is a dinosaur.
I want to write about this word, and not out of curiosity. We are in range of it.
A shop with twenty-three years behind it, fifteen of them in CNC precision work, thirty machines on the floor; customers who stay for years; machinists whose hands are all callus. Ten years ago that was a résumé. Today nobody says it out loud, but the word is not far away.
So rather than dodge it, let us take it apart.
2. First, the half they are right about
The charge against the dinosaur comes down to one thing: treating past experience as the answer to a future question.
Manufacturing is unusually good soil for that mistake.
Experience on a shop floor is bought with real money. What spindle speed makes a part chatter. Which heat treater actually holds a tolerance. Which alloy starts rusting in the humid season. You do not get that from a book — you get it from scrapped parts and invoices you would rather forget. Calling it worthless would be dishonest.
The trouble is that experience depreciates. And the rate is not up to you. It is set by how fast the world outside is moving.
A machining window used to hold for ten years. Learn your parameters once, and you could live off them for a long while.
Not any more. New alloys, new geometries, tighter tolerances, arriving in waves. On a robot joint housing, one customer’s outer diameter can be more than double another’s, while the features packed inside sit only a few millimetres apart. Run last year’s parameters on that job and you get scrap. The depreciation period on experience is now measured in months.
The symptom is worse than the cause. It is not that the dinosaur refuses to know what is going on outside. It is that his information channels have quietly failed — same people, same sources — and so new things simply never enter his field of view. Laozi put it well: looked at but not seen; listened to but not heard. Not invisible. Outside the range of what he looks at.
One level further down: refusing to change is usually not laziness.
It is that change hurts. You have to break habits that have become second nature. You have to learn something you are genuinely bad at. You have to accept the indignity of being a beginner again in your forties, and carry the risk of spending two years and ending up with nothing to show for it.
That is an accounting problem. Run the numbers over a short enough horizon and standing still is the rational choice.
I have watched that arithmetic happen plenty of times. I have done it myself.
3. And the half we would rather not lose
Lao deng is a wonderfully lazy word. Lazy enough to throw out a great deal at once.
Because swept out along with “refuses to change” are five other things: patience, perseverance, keeping to your own line, attention to detail, and a fair deal for everyone.
In this trade, those five carry more weight than an outsider would guess.
Patience — a Ø30 bore on an H7 tolerance has twenty-one microns of band in total. You wait for the machine to reach thermal stability before you cut. Rush it, and you have made scrap.
Perseverance — part 500 and part 5 out of the same batch should come off with the same size distribution. There is no trick to that. There is only logging process data, day after day.
Keeping to your own line — if you can make it, say so. If you cannot, say so. Do not take on work you cannot hold, and do not fudge a mating face with close enough.
Attention to detail — hard anodizing at 25 microns shifts the pitch diameter of a thread by roughly 30 microns. An M4 6H thread simply fails. If that is not written on the drawing, you make a whole tray of scrap. Only someone who takes it seriously will notice at all.
A fair deal — not squeezing a supplier until he quietly substitutes material, and not quoting below cost to win a job and then clawing the difference back out of the stock.
Notice that not one of those five is a skill. They are all attitudes.
Skills can be replaced, outsourced, or compressed by software. Attitudes cannot. An attitude decides what a person does when nobody is watching.
And all five share one precondition: you have to stay put long enough.
Nobody grows patience on a three-month project. Perseverance is a function of years. A sense of your own line grows only in someone who intends to eat in this trade for a long time and leave his name in it.
Which is why these qualities tend to live on the people the word was invented for.
4. The two halves come from one root
Here is the sentence this whole piece is built around.
The dinosaur’s defect and the dinosaur’s moat grow out of the same thing.
Spend twenty years on a shop floor and two things grow at once.
One is judgement. You have seen enough failures to know what can be made, what cannot, and roughly what it costs. That is the most durable asset a shop has.
The other is path dependence. Because the old way kept working, you assume it will keep working. That is the most expensive liability a shop has.
They are two faces of one experience. You cannot keep the first and refuse the second — any more than you can keep a knife’s edge and throw away its spine.
Which is why lao deng fails as a label. It ties two separable things into a single bundle and throws the bundle away.
The right move is to separate them.
The half to keep is how you treat the work — patience, perseverance, keeping to your own line, attention to detail, a fair deal. These get more valuable with age.
The half to drop is how you think the world works — experience standing in for evidence, path dependence, we have always done it this way. These get more dangerous with age.
One question sorts them: is this statement about how to do a job, or about what the world is like?
The first question, the older generation answers better than we do. The second question, their answer may have expired.
Mixing the two together — that is the actual disease.
5. Stop the mud, and let it clear
There is a line in chapter fifteen of the Daodejing.
The standard Wang Bi text reads: who can make the turbid clear by stillness? Who can make the settled live by motion? The Heshanggong edition carries two extra characters and punctuates the sentence differently:
Who can stop the turbid, and by stillness let it clear by degrees? Who can rest settled long enough, and by motion let life emerge by degrees?
I prefer the Heshanggong version, because those two extra characters turn a description of scenery into a procedure you can follow.
Stop the turbid, and by stillness let it clear by degrees. When water is muddy, stirring makes it worse. The first skill is not filtration. It is stopping. Let it stand, and the sediment settles on its own, and the water clears.
Our information environment is muddy. One wave of enthusiasm after another, each argument more urgent than the last. In that weather the scarcest ability is not judgement. It is not being carried along.
Rest settled long enough, and by motion let life emerge by degrees. But you cannot stand still forever either. Stay comfortable too long and you have to move. Move, and vitality starts coming back.
Read the two together, and the point is this: stopping and moving are not opposing roads. They are two stages of one practice.
And notice the word degrees. Slowly.
Mud clears slowly. Life emerges slowly. There is no overnight transformation in that sentence, and no way to skip past the competition. It says something we would rather not hear: real change has its own speed, and your only option is to work with it.
Which may be the most useful answer there is to the question of what to do about AI. Not a hurried pivot. Not pretending it is not happening. Stop first. Then, slowly.
6. What that looks like in practice
Four things, concretely.
Treat the good qualities as assets, not nostalgia. The machines are on the floor and twenty-three years of process records are in the drawers. The work is converting both of those from something held in a foreman’s hands into something the shop can reuse. Part 500 and part 5 coming off with the same distribution is a hard metric here, not a slogan.
Stay open, and actually learn the new tools. AI in manufacturing is not a concept, it is a tool. We use it to read drawings, to estimate cycle time, to organize process documents. We are clumsy at it and we get things wrong. On that road, using it badly is a skill problem. Not using it is an attitude problem.
Stay humble enough to take things in. Two ways, concretely. With customers: do not turn a job down because it is small — a creator’s first part is usually a single piece, and today’s one-off may be next year’s batch. With peers and suppliers: do not stop listening because you think you already know. Every link in a supply chain is holding information you do not have.
Build the system properly. Equipment can be bought; your competitor can buy the same machine and the same brand of tooling. What actually separates two shops is whether, when a part goes wrong, you can trace it to a specific operation, a specific lot of material, a specific setup. That is a system, not a craft.
7. Immovable — but what, exactly?
Most people know the phrase immovable as a mountain. It comes from Sun Tzu:
Swift as the wind, orderly as the forest, fierce as fire, immovable as the mountain.
What is interesting is that immovability is only one quarter of it. Wind, forest, fire, mountain — fast when speed is called for, ordered when order is called for, fierce when force is called for, steady when steadiness is called for. Immovability is a quarter, not the whole.
That matters, because if immovable gets read as never moves, it lands you straight back in the disease.
So our immovability is specific. What does not move is the standard of judgement — what counts as right work, what counts as a job done well. Those do not change with the market.
As for how we work: when something needs to move, it moves. Machines get replaced, tooling gets upgraded, methods get rewritten. None of that was ever on the list of things we are attached to.
I notice this piece does not really have a conclusion. Because the lao deng problem is not solved by an essay. It is solved one day at a time. The thing you had not learned before — did you actually learn it, or did you pretend to? The part you scrapped — did you trace it to the root, or did you wipe the bench and move on?
Keep your own boat in order, and let the wind blow from wherever it likes.
The tide comes in and the tide goes out. Whoever builds a sound boat does not have to guess the direction every time.