area-measurement

Settles the question every project answers three times and differently: which area standard is being used, where its boundary is drawn, and what counts toward it. A building has a zoning floor area, a

SKILL.md

GitHub

Area Measurement

Every project carries several area figures that were measured differently, and almost nobody writes down which rule produced which number. The programme says one thing, the zoning submission another, the lease a third, and each is correct. The problem appears when two of them are compared.

Scope and limits — read first

What this does. Establishes, for each purpose the project has, which standard governs, where the boundary is drawn, and what is included and excluded. Records them side by side and reconciles the resulting figures so the differences are explained rather than argued about.

What this does not do.

  • It carries no standards and no definitions. Which document governs, and what its text says counts, comes from you: the zoning ordinance, the building code, the measurement standard the lease names, the client's own brief. These differ between jurisdictions, between editions, and between standards.
  • It measures nothing. It does not take areas off a drawing or a model. You supply the figures; it establishes what rule each was produced under.
  • It does not track the programme. space-program does, and this establishes the basis that programme is measured on.
  • It does not do the zoning arithmetic. /test-far does, against the jurisdiction's own text.
  • It does not resolve which figure is right. They can all be right.
  • It does not replace the Architect of Record.

What you need before starting. The purposes the project measures for. For each, the governing document, quoted where it defines the measurement. The figures you already have and where each came from. The drawings or the model they were taken from, and by whom.

Workflow

Step 1. List the purposes before any number

Ask, in one message:

  1. What does this project measure area for? Work through: zoning submission, building code (allowable area, occupant load), the client's programme, a lease or sale, a fee based on area, a certification, a cost plan, a facilities or asset record.
  2. For each, what document governs the measurement, and can you quote where it defines it?
  3. What figures exist today, who produced each, from what drawing or model, and on what date?
  4. Is anything already committed: a lease signed, a submission made, a fee agreed on an area?

Question 4 changes the exercise. Where a figure has already been relied on, the job is to explain the difference rather than to correct it, and to find out who has to be told.

Step 2. Record one basis per purpose

One row per purpose. Every cell comes from a quoted document or is marked open.

Purpose Standard Boundary Included Excluded Source quoted

Boundary is the cell that produces most of the disagreement: measured to the outside face, to the centreline, to the inside face, or to a dominant portion. Two figures produced to different boundaries will never reconcile and the difference is not an error.

Work Included and Excluded explicitly rather than assuming. The items that differ between standards, and which one to ask about every time:

  • Shafts, stairs, and lift wells, and whether they count on every floor or once.
  • Structure: columns, walls, and which side of them.
  • Mechanical space, plant rooms and rooftop enclosures.
  • Parking, whether structured, below grade, or at grade.
  • Covered but unenclosed area: balconies, terraces, arcades, porte cochères.
  • Mezzanines, and at what ceiling height they begin to count.
  • Basement and below-grade space, often the largest single difference.
  • Space below a stated headroom.
  • Interstitial and service floors.
  • Loading, refuse and utility rooms.

Step 3. Reconcile the figures

Take the figures you were given and set them against the bases:

Figure Purpose Basis it was measured on Confirmed?

Confirmed? asks whether anyone verified the figure was actually produced under the basis it is attributed to. Usually nobody did, and a figure inherited from an earlier phase is often on a basis nobody recorded.

Then, for each pair that people compare, show the difference and account for it item by item:

<purpose A>   <figure>
<purpose B>   <figure>
difference    <figure>
accounted for by:   <item>  <item>  <item>
unaccounted:        <figure>

The unaccounted line is the finding. A difference fully accounted for by known exclusions is not a problem. An unaccounted remainder means one figure is wrong, or was measured on a basis nobody recorded.

Step 4. Name what depends on each figure

For each purpose, record what is riding on it: a submission already made, a lease, a fee, an allowable area calculation, a certification. Then say what would have to happen if that figure moved.

This is what turns a measurement exercise into a decision. A programme area that drifts is a conversation; a leased area that drifts is an amendment.

Step 5. Report

  • The basis table, with every cell that was assumed rather than quoted, counted.
  • The reconciliation, with the unaccounted remainder for each pair compared.
  • Every figure whose basis nobody confirmed.
  • Every committed figure and what depends on it.
  • What to quote and from where, for the bases still open.

Close by saying that nothing was measured here, and that several figures being different is the normal condition rather than an error.

Step 6. Save

Ask where. One file per project, and revisit it whenever a new figure appears, because the point is that the bases stay written down.

Rules

  • Never supply a standard, a definition, an exclusion, or a factor.
  • One basis per purpose, and every basis names its source or is marked open.
  • Record the boundary explicitly for every basis.
  • Never reconcile two figures into one. Explain the difference and report the unaccounted remainder.
  • Ask, for every basis, about the items in Step 2 rather than assuming.
  • Name what is committed and what would move if a figure did.
  • Say on every run that nothing was measured here.

Anti-patterns

  • Producing one area for the project.
  • Assuming a standard's exclusions from what is usual.
  • Comparing a figure from the programme against one from the lease without asking what boundary each used.
  • Treating an unaccounted difference as a rounding error.
  • Inheriting a figure from an earlier phase without recording its basis.
  • Recalling whether a jurisdiction counts below-grade area.
  • Correcting a committed figure without naming who has to be told.

Resources

None. This skill is one file. Output is written directly at the path you choose.