Chapter 2. Appendix: The records read

Is the demolition worker who stepped back first at a sign of collapse a shirker?

A demolition site in a redevelopment zone

One afternoon while tearing down a brick multi-household building, a demolition worker of fifteen years stopped work and stepped back, saying the sound from the wall had changed. It was a site paid at a contract rate per pyeong, so work stopped for an hour, and some said 'he's slipping off again'. When the demolition foreman went up to look, one side of a beam was already sagging.

The sound's changed. Dust is falling from above. Let's stop here for today.

You're slipping off again, brother. At this rate the rate gets cut.

He didn't slip off, he saw it first. Put props in, and tomorrow morning we start again from the upper floor.

Whether this worker's feel is always right, this scene does not answer. That the rate presses on judgment is also put as an assumption.

The body that stepped back first is the signal that read the danger first. The contract rate covered that signal with the name 'slipping off'.

Relying on one person's feel makes stops frequent. Only if the standard for stopping is set not by a person's ear but by something measurable, like a crack gauge, will everyone accept it.

The problem is who pays for the hour stopped. Under a rate per pyeong, time stopped safely becomes everyone's loss, and only the person who said to stop gets the hard looks.

In this scene the person who stepped back first was not a shirker but the one who saw first. Yet under a rate per pyeong the hour stopped becomes everyone's loss, and the signal is covered by the words 'slipping off'. If who pays for the value of the time stopped stays as it is, the next signal will not be spoken aloud. (as of 2026-09-25)

If the number of accidents and near misses does not differ between the days stopped and the days not stopped at the same site, this judgment weakens. It is checked by setting the stop entries in the site safety log beside the near-miss reports.

Time stopped for a danger signal is taken out of the contract rate and paid separately by the prime contractor. The person who saw first no longer remains as the person who cut colleagues' share.

Why does telling the customer a bad estimate pass over to the mechanic?

Reception counter of a farm-machinery repair centre

In rice-planting season the counter of the town's farm-machinery repair centre was crowded with farmers wanting to leave their rice transplanters. When an estimate came out with repair costs above half the machine's price, the counter clerk sent the customer to the workshop, saying 'The chief will explain it himself.' The repair chief came out several times a day with oily hands to deliver the bad news, and meanwhile the tiller on the workbench stood still.

Chief, it's the customer for transplanter number 3. When I talk about the engine they keep arguing.

All right. From next time let's go out together holding the part we took off. When they see it with their own eyes they argue less.

While you're out, chief, two tillers are waiting. I'll drain the oil first.

Here what passes over is less the amount of work than the emotional load of taking the customer's disappointment. Whether the counter clerk tried to avoid anger or passed it on for lack of grounds to explain, this scene does not decide. How much the stopped workbench lengthened the wait of other farmers is also a share to check separately.

The load of delivering bad news falls on the person who meets customers least. 'The chief will explain' is a channel that sends the emotional load to the workshop.

It is right that a mechanic gives the technical explanation. What the counter passed on is the boundary of expertise, and the only problem is that the time spent explaining comes out of working time.

The counter clerk was not given the grounds to explain the estimate. Neither the old part nor the parts price list is at the counter, so the passing on is the mark of information not reaching the counter.

What the counter passed on was less a gesture of avoiding work than a seat where one has to take anger without grounds. The chief's suggestion to go out holding the old part is a knack for sharing the grounds of explanation with the counter. But if explaining always stops the workbench, the price is paid by other farmers waiting for their tillers. (as of 2026-09-25)

If the counter kept calling the chief for every expensive estimate even after receiving the old part and the parts price list, the cause is not a lack of grounds but a wish to avoid anger. It is confirmed if the number of times the chief was called is written in the reception ledger.

The counter clerk first takes the part that was removed and the parts price list in hand and stands before the customer. The chief comes out only when a technical question remains, and the share of bad news is divided between the counter and the workshop.

What does a staff room where evening self-study supervision is bought and sold reveal?

High-school staff room for evening self-study supervision

At this high school, evening self-study supervision came round to each teacher four or five times a month. Teachers with young children passed the supervision to colleagues for some tens of thousands of won a time, and the buyers were usually young teachers who lived alone and were paying off student loans. When the head of academic affairs counted the supervision sheet, a few people were guarding night classrooms more than ten times a month.

I'll stand this Thursday as well. You have to go and pick up your child.

I'm sorry every time. Instead, I'll take one class of your performance-assessment marking.

Three people have stood more than ten times. That a price has been attached means this duty weighs differently for each person.

Whether the lesson preparation of the teachers who supervised often decreased, this scene does not answer.

The price reveals that the same duty weighs differently for each person. It is heavy for someone with care to give, and light for someone in urgent need of money.

Buying and selling is a support that people in different circumstances made for themselves. If both sides were better off, what is revealed is the unreasonableness of one rota that is the same for everyone.

If the buyer is always someone young and in urgent need of money, the price becomes a channel that makes the difference in circumstances be paid in night hours. The lessons of the person who stands more than ten times wear thin first.

The buying and selling in this staff room was a support colleagues made among themselves, and what it revealed is the fact that one rota cannot weigh the same for everyone. The offer to take one class of marking is a sign that each knows the share the price alone falls short of. But if the buyer is always a person in the same circumstances, the support tilts the night hours to one side. (as of 2026-09-25)

If many teachers would stand the same number of times without taking a price, the price is not the difference in circumstances but only a way of recording goodwill. It is checked by comparing a term's supervision sheet with the staff room's ledger of trades.

Weekday night supervision and weekend event duty are put on one sheet, so that people take different kinds to suit their circumstances. Those who buy up nights with money do not gather on a few people.

To whom does the helpdesk's 'try rebooting' pass the work?

In-house IT helpdesk

The in-house IT helpdesk of a manufacturing company was assessed by the number of tickets handled. When the phone rang, the first words were 'Try rebooting, and if it doesn't work, get in touch again', and tickets closed that way were counted as solved.

Try rebooting, and if it doesn't work, get in touch again. I'll close this ticket.

It's the third reboot. The month-end vouchers won't print.

The printers on that floor have all been like that since last week's update. I'll go up now. Let's bundle the queries from the same floor and look at them in one go.

This scene did not rule out the cases where the reboot advice actually solves many problems.

'Try rebooting' is a passing off that returns the work to be done to the person who asked for help. The assessment by number of tickets counts it as solved.

Rebooting is a first move that actually solves many problems. What makes the passing off is not the first words but the count that tallies the same query coming back three times as three tickets.

The price of passing work off is paid by the team that asked for help. Unless the time that team spent is written into the helpdesk record as well, the tilt is not seen.

The reboot advice is not in itself passing work off. The count that tallies the same query coming back three times as three solved tickets makes the first words into passing work off. The senior's bundling of the queries from the same floor and going up was a judgment that looked at the cause rather than the number of tickets, and a judgment that actually loses out by the number of tickets. (as of 2026-09-25)

If, among the tickets closed with reboot advice, few came back within a week, this judgment weakens. It is checked in the helpdesk record by the rate of repeat queries from the same person and the same equipment.

Queries that come back within a week about the same equipment are bundled into one ticket, and a person is named to look at the cause. The share of repeated rebooting returns to inside the helpdesk.

If the person who finds an outlier also takes on the investigation, who will try to find one?

QC testing laboratory of a pharmaceutical plant

In the QC laboratory of the pharmaceutical plant, the rule was that the person who first saw an out-of-specification test value also wrote the deviation investigation report. One report took 2–3 weeks, and meanwhile one's own tests were pushed back. Over a year, outlier reports came only from a few people, and when an ambiguous value came out, running a retest before any investigation became more common.

Of the six outlier reports this year, four are mine. My stability tests are three months behind.

When it's ambiguous I just put the sample on again first. If I write a report, a month goes up in smoke.

I'll keep doing the finding. Let's go to the team leader together and say that someone has to take the investigations separately. Bring the number of retests you ran, too.

This scene did not check whether the retests actually covered up defects, but in pharmaceuticals that blank can reach a patient, so the cost is heavy.

The rule that the finder takes on the investigation as well makes not finding the rational side. The body that holds back has changed places into a blank in quality.

The body that runs a retest first is the side that has done harm. Even if the rule is tilted, putting an out-of-specification value on again without investigation is a line the quality system does not permit.

That the finding is concentrated on a few is the bigger tilt. If one person writes four of six reports, that person's tests are pushed back and another blank appears.

The third-year's way of running a retest first is something the rule taught. As long as the finder takes on 2–3 weeks, not finding becomes the rational side. If that rationality goes into a blank in quality, the price is paid outside the laboratory, so this scene is also one where the body that holds back does harm. (as of 2026-09-25)

If, even after someone takes the investigations separately, outlier reports come only from the same few people, the cause is not the rule but each person's own standard. It is checked by the list of reporters and the number of retests in the year before and after.

The person who finds an outlier writes only the first page of the report. The investigation is taken over by an investigation duty that rotates.

Whose blanks does 'not applicable' on the checklist increase?

Solar installation crew

The solar installation crew filled in a forty-two-line completion checklist on a tablet at every site. Head office counted the team by the rate filled without blanks, and 'not applicable' also counted as a filled cell. When the installation part leader looked at last quarter's records, more than half of the earth resistance cells at rooftop sites were 'not applicable'. Two months later an inverter stopped at one site because of faulty earthing.

There's one earth tester per truck, and it was at the next site. You have to fill the cell to move on.

A cell isn't something you fill, it's something you close. Write what you couldn't measure as 'not measured'. I'll go and measure it the next day.

When we ask for one more tester, we can take that 'half' figure with us.

At a solar site, one blank cell comes back two months later as stopped equipment. Whether the shortage of testers is the whole of 'not applicable', or whether bother was mixed in too, this scene does not decide.

The way of counting makes 'not applicable' the cheapest ending. The body that holds back came from the yardstick.

The single tester per truck comes first. 'Not applicable' was the only way to close a cell on a day when it could not be measured.

This scene is the side where the body that holds back actually did harm. Whoever filled the cell, the price of the equipment stopped two months later is paid by the customer and the team that goes again.

That 'not applicable' passed half was less because people wanted to skip measuring than because there was no other way to close the cell on a day without a tester. The yardstick that counts only the fill rate made that way the cheapest, and the stopped inverter is its price. The part leader's 'not measured' is a choice to bring the missing equipment up onto the record. (as of 2026-09-25)

If 'not applicable' stays at nearly half even after a tester is placed at every site, the cause is not the equipment but the way of counting and bother. It is checked with the checklist records of the quarters before and after the testers were increased.

Put a separate 'not measured' cell on the checklist. As those cells pile up, increase the testers and the revisit days by as much.

Has 'pretending not to hear' on the radio become declining a call in the app?

Motorcycle courier dispatch office

In a 1990s motorcycle courier office the dispatch manager called out jobs by radio, and on rainy days the radio went quiet for a moment at a call to a distant neighbourhood. The manager knew who was pretending not to hear, but squared the account by handing them a good job on the next call. In the apps of the 2020s, declining a call lowers the acceptance rate, and a low acceptance rate brings fewer good calls. Rainy-day calls to the outskirts gather on newly arrived riders who have to fill their acceptance rate.

Nobody taking the one to the outskirts? Fine. The next Yeouido job goes to whoever answers now.

These days you can't pretend not to hear. One decline stays as a number.

I take all the rainy-day calls to the outskirts to fill my acceptance rate.

If it's the same direction I'll take one for you. I can drop it off on the way.

That unwanted calls gather on newly arrived riders is this scene's setting, and actual dispatch rules differ from app to app.

Pretending not to hear on the radio was a decline that colleagues knew of and covered. There was someone to square the account, so declines did not pile up on one person.

The app's acceptance rate sends unwanted calls to the newest arrival. When declining becomes an individual's number, only those who can afford to decline do so.

The radio days were not even either. The manager's accounts turned good calls to those close to the manager, and the app brought in a skew of numbers in place of that favouritism.

Declining in the radio days was declining that the manager squared with the next call, and declining in the app remains in the acceptance rate, so only those with room can decline. In both eras unwanted calls went to someone, but now that someone hardens into the newest arrival. The manager's accounts could tilt towards those close as well, so the old way cannot be called an even way. (as of 2026-09-25)

If calls to the outskirts gather on newly arrived riders even in an app that does not use the acceptance rate for dispatch, the cause is not the acceptance rate but the position of new riders, who decline less. It is checked by the share of outskirt calls by experience in two apps with different ways of dispatch.

Rainy-day calls to the outskirts are taken out of the acceptance-rate count and counted separately. They are turned in sequence regardless of experience, narrowing the way by which the cost of declining piles up on one newly arrived rider.

Is the shallow sea set aside for the older women divers still a shared portion?

Women divers' common fishing ground

In the 1970s this village's fishing cooperative set aside the deep sea for the sanggun divers and the shallow sea inside the rocks for the older divers. A body that goes in less deep was not a fault but a set share. In the 2020s most of the divers are past seventy, and three sanggun who can go into the deep sea remain. The shallow sea is crowded, and the harvest of the deep sea hangs on three people's breath.

From this year, inside that rock is halmang badang. Young ones, don't go in there.

There are three of us to go into the deep. Our breath is this village's abalone price.

We'll see to the shallows, so you go into the deep only on days with a good tide. If your body breaks, it's all over.

How much strain the deep-sea share puts on the sanggun's bodies, this scene does not answer.

The shallow sea set aside was a wisdom that made the body that does less into an assignment instead of a label. The frame is still right now.

The weight of the assignment has turned upside down. Now that most are old, the body that does less has become the majority, and the load of the deep sea gathers on three people.

The crowded shallow sea is a load too. Only if the sum for sharing the harvest is done again will the sanggun's breath and the crowding of the shallows ease together.

The halmang badang of the 1970s was an assignment that made the body that does less into a share instead of a fault. But now that most are old, the same assignment drives the load of the deep sea onto the breath of three sanggun. The name of the assignment is the same, but its weight has turned upside down. (as of 2026-09-25)

If the sanggun's diving days and injuries have not increased compared with the time when the shallow sea was set aside, the judgment that the load gathered on a few weakens. It is checked by counting the fishing cooperative's diving days per year and the divers' medical visits.

The days the sanggun go into the deep sea are reduced to days with a good tide. The harvest reduced by that much is made up from the fishing cooperative's common share.

Into whose hands does a blank in the auction-hall inspection pass?

Dawn inspection team at a fisheries cooperative's auction hall

Before the dawn auction, the place-of-origin label on one box is missing. The inspection team leader suggests looking at the other boxes first. A colleague points out that the further back that box goes, the shorter the auction staff's time for checking it becomes.

Shall we put this box to the back for a moment?

Then the auction side will have even less time. I'll start by finding someone who can check the origin.

At the auction hall, the starting time of the auction adds to the burden of checking. Actual labelling errors or trading results were not observed.

Sending the blank to the back makes the auction staff's share of checking bigger.

It may be that the inspection team has no authority to complete the label.

Looking at the other boxes first is a reasonable order.

Before the inspection team leader is seen as a person trying to avoid it, who can fill in the place-of-origin label must be checked. If there is no authority, the assessment that the blank is being passed off changes too. (as of 2026-09-25)

If the inspection team kept passing it to the back even with the authority and the data to check the place-of-origin label at once, the reason must be sought in another division of the work.

First connect the person who will check the origin information with the auction staff's time. Keep the burden of checking the boxes with missing labels from piling onto one place just before the deadline.

If colour matching is passed on as "not my turn", what gets mixed up?

Colour-matching room of a paint plant

The paint team leader says the next shift will check the final colour. The next leader, seeing parts already in the big tub, reports that the reference panel is missing. The two leaders agree again on which point in time's colour they compared.

Can the next shift look at the final colour?

We can look at it. Just give us the reference panel with it and we won't be lost again.

In the painting process there is a time lag between the small reference panel and the big tub of parts. No actual loss of quality was set.

Passing it to the next shift gets the judgment made once more.

Time passed on without the reference piece becomes the next shift's rework.

Who is responsible for keeping the reference must be settled first.

The cause of the next leader's slowed hands lies less in a person's laziness than in the reference panel going missing between shifts. The cost should be judged after checking whether the scope of responsibility is set. (as of 2026-09-25)

If the same rework repeats even though the reference panel is always passed on with it, the process conditions must be looked at again rather than missing information.

Make a place for handing over the reference panel together with the time of the big tub. Do not pass the time for judging colour onto the next leader alone; divide it between both shifts.

Whose time are the twenty minutes in which the team leader left the counter empty?

The team leader's seat in a district office's civil-complaints room

The team leader of the civil-complaints room steps back behind the counter for twenty minutes and sets out why the same complaints keep recurring. The counter clerk faces the queue alone. The team leader comes back, hands what was set out to the colleague, and together they change the order of explanation for the next shift.

The same complaint has come back three times. I'll go and write down the cause.

I'll take the queue. Next time please schedule the time for setting it out as well.

The civil-complaints room is a place where the queue shows at once as a colleague's burden. Actual performance in handling complaints was not measured.

The time the seat was left empty is a cost left to the colleague at the counter.

The time spent setting out repeat complaints helps the next shift.

Both effects are there, so who took the burden of the queue must be weighed.

The team leader's setting-out may be useful, but the burden of those twenty minutes of waiting fell on the colleague. The gain of the setting-out and the cost of the waiting should not be tied to the same person's character. (as of 2026-09-25)

If the next shift's repeated explanations did not fall at all and only the waiting at the counter grew, this time cannot be read as room for improvement.

Place the time for setting out repeat complaints in the duty roster, and decide at the same time who will take the counter meanwhile. Do not pile the queue onto one colleague.

Even when the server comes back by itself, is the moment at night empty-handed?

A data centre night operations room in the 2010s

After the server has restarted automatically, the night operator sits down in a chair for a moment. A colleague skims the alerts that piled up meanwhile and asks which work is not yet finished. The operator tells the restart and the finding of the cause apart as different jobs.

The server is back, but the cause is still sitting where it was.

I'll bundle the alerts. Have a glass of water and we'll look at it together.

It is not an actual outage record.

The load differs depending on who sorted the alerts.

The time sat after the automatic recovery is idle.

A short rest is recovery for the next check.

Before counting the operator's rest as empty hands, the checking that remained after the automatic recovery must be looked at. If the colleague skimmed the alerts in the operator's place, that share must be recognised as well. (as of 2026-09-25)

If no further checking were ever needed after the automatic recovery and no burden of alerts fell on the colleague either, there would be no ground for counting this moment as part of the remaining labour.

Divide the finding of the cause after an automatic recovery and the sorting of alerts evenly between the two on the night shift. Bring out, too, who filled the time of the rest.

Whose share does the paddy the spraying drone skipped become?

A joint rice pest-control team in 2030

An automatic judgment chooses the order of the joint pest-control drone. The farm-support leader says one paddy was skipped because of an irrigation channel. A colleague works out that if that paddy is taken last, the work of the other farms is pushed back too.

The channel cuts across this paddy. Let's look at the order again.

I'll work out the travel time too. So that only the households pushed back don't lose out.

In rice farming, the irrigation channels and the order of joint work are interlocked. It is a supposition about the future, and there is no actual effect of pest control.

The skipped order gathers the joint pest-control team's load on one side.

The order the drone chose may have saved total travel time.

The farm that owns the skipped paddy can deal with it separately.

As the colleague's calculation shows, the cost of skipping one paddy may not stay with one farm alone. The drone's gain in total time and the labour pushed back must be looked at together. (as of 2026-09-25)

If taking the skipped paddy later did not increase the other farms' work or the total travel time at all, the drone's skipping did not add to the shared burden.

The automatic judgment chooses the order of visits again, taking in the channels and the travel time. The share of checking for the delayed paddy is divided evenly between the farm and the leader.