What Does a Commission Earn?
At eleven on Saturday, Gu Ning put the D17 goal hierarchy on the table. Tang Ke said the next step was to compare whether the job would make money. If the money from a new commission exceeded materials and labor, continuing would make sense. Ye Cheng asked where the empty half day would be recorded if they reserved a bay and did not win the job.
Tang Ke paused. She had planned to record failure to win the job as zero income, without recording the bay that had been occupied and could not be used for other work. If they did win it, she intended to record all receipts as gains before deducting the materials and labor needed to fulfill it.
Chapter 2 showed that there is more than one goal. This chapter further organizes consequences. Gain is not a collective name for every benefit, and loss does not occur only with final failure. Direct consumption, opportunity occupation, rework, waiting, and recovery may occur at different times and be borne by different subjects.
We remain in Chengwan's fictional Saturday discussion. The client has sent no new message, and Version 2, price, materials, and delivery conditions do not exist. Quantities below demonstrate how to record consequences. They do not enter the actual R17 storyline or represent costs in the repair industry.
Give “Gain” a Limited Meaning First
Gu Ning proposed calling the portion expressible in working-capital units accounting contribution. If the consideration received for a task exceeds the added materials, external services, and clearly attributable labor consumed for it, the remainder can enter the accounting-contribution column.
This is a simplification within the case, not a complete accounting definition. Equipment depreciation, taxes, long-term maintenance, and other real-world items do not disappear because of one formula. Actual use must follow the relevant systems and business records; the book's working-capital unit is not a real accounting rule.
Accounting contribution is not the whole value of an action. A clarification may generate no income yet obtain a critical condition. A task may make a positive contribution yet damage existing fulfillment or exhaust recovery capacity. These results can be retained separately without forcing every value into a money column.
The qualifier prevents expansion in two directions: total receipts cannot be called net gains, while the inability of money to express all value does not make an accounting difference meaningless. It answers one specific question; other questions have their own places.
Total Receipts Differ from Added Contribution
Consider an independent demonstration. Suppose a completed task brings ten working-capital units, consumes four added units of materials, one unit of added external processing, and two units of labor explicitly chargeable under the case rule. Its simplified accounting contribution is three units.
Reporting that “the task brought ten units” may accurately state total receipts, but must not imply that all ten are available for other goals. Reporting only three should state which costs have been deducted so the consequence table does not deduct them twice.
The example does not establish how labor converts into working-capital units. It directly stipulates that two units have already been counted under the case rule to show the structure, not to supply a real conversion. When conversion is unreliable, labor can remain in a separate column rather than being forcibly priced to produce a net number.
R17 has no new quotation, so ten, four, one, two, and three cannot be entered in D17's current fields. The figures only tell the three people to distinguish total inflow from added consumption after a price arrives, before deciding whether a corresponding contribution exists.
Cash Timing May Cross a Boundary First
Even when final accounting contribution is positive, usable working capital may run short during the process. Suppose a task requires four units for materials before seven arrive after completion, producing a simplified final contribution of three. If the station currently has only two deployable units, the positive final figure cannot make the present payment.
Borrowing, deferred payment, or staged receipts can change the path, but each is an added condition with possible costs and responsibilities. The consequence table cannot assume money will always arrive in time merely because the endpoint shows a surplus.
Chengwan presently has four explicitly deployable units and eight protected units. A plan requiring five at the start touches the protection line. Unless a new reliable resource appears or the protected need is reassessed, the plan cannot be adopted under present conditions.
Cash inflows and outflows must therefore be preserved by time. A total answers the difference over a period; timing answers whether action can continue throughout it. Both are real, and neither substitutes for the other.
Labor Occupation Has Two Consequences
Using a labor slot first creates actual work and fatigue. Second, the same slot cannot perform another incompatible task at the same time, creating opportunity occupation. The two consequences sometimes overlap and sometimes do not.
If an otherwise idle slot had no feasible alternative use, reserving it and leaving it empty may have a small opportunity loss, though administration and waiting may still consume something. If another definite task was refused because of the reservation, the consequence of occupation can be traced back to that opportunity.
Opportunity cost cannot be calculated solely from the best option visible afterward. If a high-value task becomes known only after Tuesday, the Saturday actor did not abandon an already visible, certain opportunity. Information acquisition can be reviewed, but the original consequence must distinguish alternatives available at the time from paths learned later.
Nor does the absence of an accounting expense make an empty slot lossless. Its schedulable capacity belonged to the capacity in D17's goal table. Once locked, it is unavailable to other known choices. Whether occupation is worthwhile must connect to a deadline and release condition.
Waiting Occupies Attention and Scheduling Space
Some actions do not consume a full labor slot but preserve an unresolved place. Tang Ke must check for replies, Ye Cheng must hold material, and Gu Ning must review the matter before scheduling. Each act is small; without a deadline, it can repeatedly reclaim attention.
The book does not convert every check into an exact amount of money. It can first record frequency, responsible role, and stopping condition, then determine whether other work has been affected. Lack of monetary conversion neither erases the effect nor permits exaggerating it into a severe loss.
If one clarification specifies waiting until noon Monday, the unresolved occupation has a clear boundary. Without one, everyone may reconsider “wait a little longer?” at every new moment, allowing a small action to accumulate invisibly through repetition.
The consequence table must therefore include duration. Saying “one message costs very little” without specifying the later waiting and reopening arrangement may understate the actual work path created by the action.
Rework Is Not a Synonym for Failure
Rework can occur while a task ultimately succeeds. Remaking material, changing schedules, and rewriting explanations all add inputs. Final delivery does not erase rework under a success label.
Likewise, failing to win a task need not cause rework. If the station sends one clarification and makes nothing, its loss structure differs from a completed task revised several times. Rework cannot be filled in directly from a success/failure split.
Its cause also must be distinguished. Changed conditions, execution deviation, defective raw materials, and omitted judgment may require different improvements. The consequence field first preserves what happened; attribution waits for later material. One extra labor period does not immediately prove who was wrong.
The main case has not begun production and has no actual rework. The field is reserved so later events can enter as they occur, not to imply that R17 must undergo repeated revision.
Recovery Cost Describes the Return to an Actionable State
Resources do not necessarily recover the moment an attempt ends. Specialized material may not transfer to another job, a bay may require clearing, the existing schedule may require renegotiation, and the people involved may need rest. A path still lies between stopping and restored capacity.
If a plan commits two units and recovers one after stopping, net consumption differs from gross commitment; nevertheless, the occupation period may still affect other choices. Recoverable value, recovery time, and uncertainty of recovery should be stated separately.
Recovery cost is especially relevant to sustainability. An action that appears to occupy one slot but needs two more after failure affects future capacity by more than one. Conversely, a probe that resets immediately may preserve more next steps even when its result is adverse.
Gu Ning required every action to state how it ends. A table that describes entry but not exit and restoration may merely hide later losses outside the action's name.
Relationship Consequences Cannot Be Casually Converted into Money
Timely explanation of boundaries may increase trust, overcommitment may cause disappointment, and prolonged silence may affect the entrance to negotiation. These consequences are real, but they may have no reliable conversion into working-capital units.
Forced valuation manufactures an appearance of precision. Saying a courteous clarification is worth two units, without an observation and a rule, merely hides a preference inside a number. Directional descriptions with grounds can be used first; quantification can be studied later if stable records emerge.
Refusing to quantify does not make relationship value infinite. If someone claims the relationship requires crossing the eight-unit protection line, they must explain why it is sufficient to cancel a survival constraint. Vague long-term value cannot automatically override a clear present consequence.
The table allows different units to stand side by side. Its purpose is not immediate summation but visibility: which relations still require tradeoffs, and which have already touched an uncrossable boundary.
Learning Gains Must Answer the Original Question
If the action was intended to discover whether Version 2 would form, and one reply supplies scope, timing, and payment conditions, the information goal has obtained definite material. This learning can be recorded even if the station ultimately declines the job.
If the reply only says “we'll see later,” the original question remains unanswered. New words do not fill the “experience gained” field. Information gain depends on which unknown was reduced and whether the material can change an action.
Learning also has a range of transfer. Material specific to R17 can help the present action without becoming a general rule for every commission. A repeatable method of checking versions may have wider use but still requires testing on later objects.
Gu Ning required the learning field to state the original question, material obtained, choices changed, and range of transfer. Where no correspondence exists, the field remains incomplete; intellectual inspiration does not replace an actual information product.
Losses Borne by Different Subjects Cannot Be Directly Netted
Suppose a new task adds three units to the station but imposes a two-day delay on the client of an existing commitment. Three cannot simply be reduced by two unless a justified conversion makes “two days of delay” equal to “two units” and the bearer's position has been addressed.
The same applies inside an organization. Income enters a common account while one person bears added night work. A positive total ledger does not make distribution automatically reasonable. Who decides, who benefits, and who bears the result belong in the consequence table.
This does not mean no action is possible whenever any subject incurs loss. Negotiated compensation, rearrangement, or voluntary assumption can form an acceptable plan, but they must actually occur. Overall gains cannot consent on a person's behalf.
RC's hierarchy of power supplies a direction for checking: when evaluative authority and consequence-bearing remain misaligned, a plan may build its stability on the contraction of someone else's margin. This chapter identifies the concrete subjects; the theoretical term does not declare any character malicious.
An Occurring Result Does Not Guarantee Positive Net Consequences
Consider an independent comparison with two actions, “light clarification” and “deep advance preparation,” and two future results, “a suitable version forms” and “no version forms.” The four action-result combinations have different consequences.
When a suitable version follows light clarification, the station obtains conditions while preserving most resources. When no version follows, the main loss is brief communication and limited waiting. Deep preparation followed by a version may permit an earlier start but has already occupied more money and capacity. Without a version, it leaves specialized inputs and recovery work.
The shared result “a version forms” does not decide which action is better. Whether the speed from deep preparation has value depends on the goal and actual conditions. Nor does “no version forms” impose the same loss on every action.
This four-cell comparison supplies no probabilities and does not yet recommend an action for Chengwan. It shows that consequences belong to combinations of actions and external results, rather than to results marked simply as gains or losses.
Failure May Also Produce Something Positive
When the external result misses the primary goal, a limited probe may still yield reusable information, expose a process gap, or release resources in time. These positive products can be recorded without treating failure as the destruction of all value.
But they cannot erase the primary goal's failure. A commission not won remains not won, and actual inputs remain in the ledger. Learning is evaluated separately; it cannot rename every adverse outcome as another kind of success.
Separate fields protect review. Actors neither reject real gains because the primary result was adverse nor use those gains to evade the original consequence. Each goal receives its own material, allowing feedback to identify the level that must change.
Conversely, a favorable result may leave negative products. A completed project may still have crossed the protected cash line, delayed existing work, or overloaded personnel. A favorable endpoint does not sign for the process.
For Uncertain Consequences, First Record Ranges and Conditions
Because Version 2 is unknown on Saturday, many consequences cannot receive numbers. Gu Ning can write, “If specialized purchasing is required, early occupation occurs,” but cannot enter two units. She can write, “If existing slots must be reassigned, a fulfillment conflict is triggered,” but cannot assume delay.
Conditional consequences and occurred consequences belong in separate places. The former help formulate questions; the latter enter the actual ledger. If a condition never appears, the station has not avoided a certain loss; it simply did not enter that path.
A range also needs a source. “A possible loss of one to ten units,” without an action boundary, is merely a wide number. The most accurate current statement may be “not yet quantifiable,” followed by the price, materials, delivery, and responsibility conditions needed for calculation.
This restraint does not prevent every action. A clarification with explicit boundaries can be considered under limited consequences. Deep commitment lacks grounds because both its object and consequences are unclear. Insufficient information affects different actions differently.
Occurred, Committed, and Possible
The consequence table also separates states. After a message is sent, communication labor has occurred. When cancellable material is reserved, money may be temporarily occupied. When a noncancellable purchase is signed, a future payment is committed. These cannot all wait to appear only when cash finally leaves.
Possible consequences depend on an external path, such as rework or delay. They remain conditional: they do not enter actual totals in advance, but they cannot disappear from the risk discussion. Different states carry different meanings for exit and recovery.
If a commitment is cancellable, its deadline and fee must be stated. “We have not spent the money” may hide an obligation already formed, while a paid expense need not be wholly unrecoverable. Real contracts and legal meanings require checking under the applicable rules; this case uses only explicit fictional stipulations.
This state table prepares for staged commitment. A stage gate can function only when the transition from possible to committed and from committed to consumed is known.
A Total Net Figure May Hide a Breached Floor
Consider a stipulated path: the station starts with twelve units, pays five, falls briefly to seven, later receives nine, and ends at sixteen. The endpoint has increased by four, but the process crossed the eight-unit protection line.
If the line protects payments due on existing work, later income may not repair a default that occurred earlier. A positive endpoint does not cancel a decisive consequence in the middle. The time series preserves a layer that the single net figure loses.
Another path may receive two, pay four, and receive two, ending with the same net change of zero without ever crossing the line. Equal net totals can conceal different survival states. Final change alone cannot rank the processes.
The five, nine, and other figures in this demonstration do not belong to the D17 storyline. They prove why the consequence table retains opening stock, timed flows, and closing stock without inventing a price for the future task.
Why Small Losses Also Need Records
Recording the small amount of time used by one communication may seem excessive. But repeated similar actions can accumulate, and without any place in the record, people can later rely only on the impression that “it probably cost nothing.”
The record need not count every minute. It can use an agreed granularity such as one communication, one review, or a fraction of a labor slot, together with a stopping condition. The granularity serves the question of whether a choice would change; it does not pursue boundless detail.
Recording small losses is not a reason to refuse every probe. Visibility does not make a cost unacceptable. It connects the cost to information and goals, preventing low cost from being called zero while also preventing every nonzero amount from being called waste.
If recording costs more than the action, the fields should be reduced. The table exists to capture critical flows, not to turn every thought into administrative work.
Keep Loss Distinct from Error
A reasonable action may still create losses. A sample consumes material, clarification consumes time, and an alternative path requires maintenance. RC treats loss as a condition of evolution; it does not call every loss an error.
Error is another level: an input recorded incorrectly, a constraint crossed, execution deviating from plan, or persistence without grounds after adverse material exists. A loss may be a consequence accepted in advance; an error must be examined through concrete relations. They can occur together or separately.
If every loss is classified as failure, actors will hide their inputs. If every loss is called necessary tuition, constraints disappear. The consequence table records faithfully first; goals and rules then assess what was bearable.
Later chapters on repeated failure and attribution continue this division of labor. This chapter does not declare any future cost worthwhile in advance; it creates a place where costs can enter feedback.
Organizing the D17 Consequence Table
Gu Ning divided the table into seven columns: added accounting contribution; timing of cash inflows and outflows; direct labor use; opportunity occupation; rework and recovery; relationship effects; and information products. Each column states the bearer, status, and grounds.
The action rows retain no active contact, one clarification, a bay reserved without purchasing, general preparation, and deep specialized preparation. The result columns provisionally contain four candidates: a suitable version forms, an unsuitable version forms, conditions remain incomplete, and there is no continuation. None receives a probability.
“An unsuitable version forms” matters. Complete conditions are an information result but do not ensure the station should accept them. A table containing only agreement and no agreement would force complete but boundary-crossing terms into either success or failure.
Unknown fields state what must be obtained: specific scope, period, payment, materials, delivery, and responsibility. None has appeared, so occurred fields remain empty. The table permits blanks instead of manufacturing comparability with invented numbers.
Saving the Consequence Interface at 11:40 on Saturday
At 11:40 on Saturday, Gu Ning saved the D17 consequence table. There has been no new external information, cash expenditure, capacity reservation, or purchase commitment. The three people's discussion uses internal time, but the book invents no monetary price for it.
The table separates total inflow from added contribution, closing net change from a protection-line breach during the process, direct consumption from opportunity occupation, and stopping from recovery. Relationship and learning retain their own observation positions rather than being forced into working-capital units.
Conditional consequences remain expressed as “if” and do not enter actual accounts. How each action ends, who bears it, and when it is released become interfaces that later plans must fill. The historical relation among the old 0.5, R17's nonoccurrence, and the proposal to talk next week remains unchanged.
This chapter reaches the following judgment: gains and losses can share a table without first being compressed into one total. Preserving kind, time, subject, and recovery reveals whether later weighting compares the same objects. The next chapter introduces expected value, showing how probability weights these scenarios and why a positive average still cannot answer whether the station can bear a specific failure.