Load Calibration

Load Calibration is the procedure by which the conceptual weight borne by a structure within the Tower is measured against a certified standard and adjusted until the two agree, or until the discrepancy is reclassified as intended. It is distinct from Calibration proper — which concerns spatial and dimensional agreement — and from Conceptual Load Testing — which verifies that a wall or node survives its designated unreal subject load. Load Calibration asks a narrower question: not whether the structure holds, but whether it is holding precisely as much as the record says it is.

Responsibility for Load Calibration rests with the Clerks of the Net Load, whose registry of calibration schedules for Load-Bearing Nodes is the authoritative list of what is owing to be calibrated, on which floor, and by whom. The Instruments Office issues the certification stamps. The two have not spoken since the winter of the disagreement, discussed below.

Method

Calibration proceeds by comparison with a Standard Load, held in the Œdicentric Calibration Atrium on floor 1122 and massed at 44 units of conceptual weight. The standard is re-massed annually, because it drifts. Drift is expected, recorded, and forgiven up to a tolerance of 0.3 units under Fourth Concordance, Annex 2, which states plainly that "a standard which did not drift would be a standard no one had measured properly" (Annex 2, Clause 6).

Application of the standard is by Load-Diffusers, per the Principle of Hypothetical Load, which governs diffusion as integral to Load Distribution rather than as a separate operation. Where a diffuser is unavailable or under seal, calibration may be performed by the Alternating Method: the subject load is compared against a freshly calibrated neighbour and the discrepancy halved and attributed alternately to each. The Alternating Method is permitted but not popular. Librarian Ospreth Vance, who audited the practice in 3011, noted that "the method permits a clerk to be wrong twice before being right once, which is generous but slow" (Memoranda of the Third Audit, fol. 88).

History and the Stamp Dispute

Load Calibration was regularised by the Decree of Austere Function, drafted by the Clerks of the Net Load after their survey of load-bearing inefficiencies found that roughly one floor in nine had been carrying a load that no document claimed for it. The Decree fixed a schedule of calibration intervals by node class and has never been fully implemented, on the grounds that implementation would require reopening floors 470 to 619, which no one wishes to do.

The relationship between the Clerk and the Instruments Office deteriorated on floor 702 in the winter of 2846, when a calibration certificate was stamped twice — once by each office, in the confidence that the other had not. The double stamp is not illegal. It is, however, ambiguous, and Fifth Concordance, Annex 9, Clause 3 provides that no load "shall be deemed calibrated where the stamp is ambiguous." A dispute arose as to whether the second stamp voided the first or confirmed it. The Instruments Office argued confirmation; the Clerks argued voidance; the floor in question argued neither, having been calibrated correctly and left out of it entirely.

"Where two offices certify one condition, the condition is certified once and the offices twice, and the surplus is to be entered in the ledger as good will."

— Fifth Concordance, Annex 9, Clause 11

The clause settled the matter without satisfying anyone. The Clerks maintain the register; the Instruments Office maintains the stamp. Neither may proceed without the other, and both have written to the Librarians about it. The Librarians have not replied, in keeping with a practice of long standing.

Scholarly disagreement

Whether Load Calibration is a species of Conceptual Load Testing or a separate instrument has never been decided. The Structuralist School holds that calibration is a preliminary to testing and cannot stand alone; its Annex VII reading places calibration under Guild purview wherever the fixture is non-Euclidean. The Clerks of the Net Load reject the reading, citing the Principle of Hypothetical Load and their own administration of the Calculation Prong, and note that Annex VII says nothing about stamps. Practitioners resolve the question by asking which body is nearer, and recording the answer.

Recorded exceptions

A short inventory of calibrations that departed from the schedule, retained for reference:

• Floor 470, Latticed Spandrel, 2848 — calibrated twice, then untested, then reopened.
• Floor 1122, Atrium Standard, 3011 — re-massed early, on the drift, without prejudice.
• Floor 619, Threshold Node, 2877 — calibrated by the Alternating Method across three offices; all three records survive and do not agree.
• Floor 88, Stairwell Landing, 3020 — calibration suspended pending the return of the portable diffuser taken from it; the diffuser has not been returned.

Calibration may not be performed on a floor while its lamps are being changed. This is a maintenance convention, not a rule, but it has the force of one because the lamps are changed often.