WHY THESE NOTES EXIST
A useful tool makes disagreement possible.
A calculator is not valuable merely because it produces a number. A reader should be able to identify the inputs, repeat the arithmetic, find the public source behind a factual claim, and see which missing information could change the conclusion.
The examples below are deliberately concrete. They show the intermediate reasoning that a result card cannot always fit, while avoiding invented live-market data or false precision.
WORKED DECISION NOTES
Three ways the headline answer can mislead.
APT
A purchase price is not a monthly housing cost
Setup: Illustrative inputs: an ₩800,000,000 apartment, 40% cash, a 30-year loan at 4.0%, and an estimated ₩270,000 monthly management fee.
Calculation: The loan principal is ₩480,000,000. Using equal principal-and-interest payments, the monthly payment is about ₩2,292,000. Adding the management-fee estimate produces about ₩2,562,000 of recurring monthly outflow before property tax, insurance, repairs, moving costs, acquisition tax, and brokerage.
Decision lesson: The useful comparison is not “₩800 million versus a deposit.” It is the recurring cash burden, the capital tied up, and the costs that the public data cannot see. APT keeps those categories separate so an estimate cannot masquerade as a household budget.
MOVE
A $25,000 raise can create much less than $25,000 of room
Setup: Illustrative inputs: a current $100,000 salary with $2,100 rent and $2,000 other monthly spending, compared with a $125,000 offer with $3,200 rent and $2,200 other spending.
Calculation: Before tax, the current case leaves about $4,233 per month after the entered rent and spending. The new case leaves about $5,017. The headline salary rises by about $2,083 per month, but the before-tax monthly room rises by only about $784. Income tax, benefits, insurance, moving costs, and local taxes can narrow it further.
Decision lesson: MOVE therefore exposes the entered housing and spending assumptions and calculates a break-even salary. It does not treat a larger gross salary as proof that the move improves the household.
FINVECTOR
A fresh stocking record is evidence, not a catch forecast
Setup: Illustrative evidence: one water has a recent agency stocking record published only at week-level precision, while a second has a nearby USGS gauge and clearer public-access context but an older stocking record.
Calculation: FINVECTOR preserves the week-level date instead of inventing a day. It also avoids treating a nearby gauge as the exact condition at the fishing reach. The first water has stronger stocking recency; the second has stronger current-condition and access context. Neither fact proves fish remain present or that the trip is legal and safe.
Decision lesson: The recommendation is a research shortlist. The user still verifies regulations, closures, weather alerts, flow, land ownership, and access with the responsible agency before departure.
READER CHECK
Five questions to ask before acting.
- Which values came directly from the user, and which came from a public source?
- Is an output a measured fact, a financial calculation, a proxy, or an editorial ranking?
- What important cost, condition, or legal requirement is outside the model?
- Would a small change in rent, rate, deposit, timing, or source precision reverse the decision?
- What must the user verify with an agency, lender, employer, property manager, or qualified professional?
CALCULATION BOUNDARIES
What these examples intentionally leave out.
The APT example does not include taxes, insurance, repairs, loan fees, eligibility, or a lender's underwriting decision. The MOVE example does not estimate a specific person's final tax return, benefits, debt, childcare, or local tax. The FINVECTOR example does not verify a license, private boundary, closure, fish survival, or catch probability.
Those omissions are not footnotes to hide. They define the boundary between decision support and professional, agency, or household-specific verification.
CONTINUE THE REVIEW
Open the method before opening the tool.
Compare each workspace's inputs, sources, limitations, and update policy before relying on a result.