This document says how the True Inch Standard is kept. It is referred to by each Part's Foreword, but it is not part of the standard's requirements.
1Stewardship
1.1 The steward. The standard is published by Trueinch, the steward.
1.2 The standard owner is named on the method page. They sign each edition and do not report on revenue.
1.3 Sales stay out of the standard.
No commercial target shall change a curve, a weight or a threshold.
Developers, agents and other parties whose homes are rated sit on no body that decides the standard's content.
1.4 Year two to three. Stewardship is meant to pass to an independent non-profit, or to a balanced consensus body with three interest groups:
| Interest group | Who |
|---|---|
| Producers | Developers and builders |
| Users | Buyers and consumer groups |
| General interest | Architects, scientists, accessibility specialists |
2Comment
2.1 First round. Each new edition opens with a 60-day round of public comment, from its publication.
2.2 Second round. A 30-day round on the revised text follows.
2.3 Synopsis. A public synopsis says how each material comment was handled: accepted, changed or declined, with the reason. Every commenter receives it.
3The advisory board
3.1 Members. Five to seven people from outside Trueinch, named publicly, with their conflicts declared and a fixed honorarium:
architects from India and the US;
a daylight or ventilation scientist;
an accessibility specialist;
a buyers' or consumer advocate;
a valuer.
3.2 Balance. No interest category holds more than a third of the seats.
3.3 Comply or explain. The board's written advice and the steward's written response are both published.
4Continuous maintenance
4.1 Proposals. Anyone may propose a change at any time, through the form on the method page.
4.2 Response. The steward acknowledges a proposal within 10 business days, decides within 90 days, and publishes the decision with its reason.
5Interpretations and rulings
5.1 TIS Interpretations answer how the text applies to a kind of case.
They are numbered and public.
They apply to every report scored after their date.
5.2 Rulings answer one report. They set no precedent.
5.3 Who issues them. Only the steward issues interpretations and rulings.
6Editions
6.1 Three kinds of change:
| Kind | Example | What happens to scores |
|---|---|---|
| Patch, 1.0.x | Errata; code that didn't do what the text says | No intended change. Affected reports are scored again and their owners told |
| Minor, 1.x | New checks; curves calibrated again | Scores can move. Curves are re-anchored so the reference set's average doesn't change. A table of changes is published: the median change, the 90% band, and the homes that changed band |
| Major, x.0 | Structure, weights, aggregation | Scores move. A 60-day comment round, a pilot, a period when both editions run, and an end date for the old one |
6.2 Reports keep their edition.
Every report and certificate names its edition.
A report stays valid as scored under that edition.
An owner may ask for a new score under a newer edition. It is a new report, and the old one remains.
6.3 Review. The whole standard is reviewed at least every two years. The next review date is printed in each Foreword.
6.4 Records. Drafts, comments, synopses and decisions are kept for five years.
7Disputes
There are three tracks:
| Track | What it covers | How it's handled |
|---|---|---|
| Report facts | "That window is 1.5 m, not 1.2 m." | The owner corrects the input in the report; it is scored again at once, and the change is flagged. A free review by a person can be asked for instead, answered in writing with the evidence within 30 days. If a dispute stays unresolved, the owner may attach a statement to their shared links. A seller or developer may ask for a review of a buyer's report, and the outcome goes to the report's owner |
| Method | "This curve is wrong." | Logged as a maintenance proposal (§4), and answered publicly by the next minor edition |
| Procedure | "You didn't follow your own rules." | Decided by a panel of three not involved in the decision: two board members and one outside member. Only procedural grounds count, filed within 15 business days of the decision. The decision is published |
8Quality and gaming
8.1 Random checks. A person checks 2% of reports at random, and every report behind a badge. The defect rate is published.
8.2 Bunching. Each quarter, the distribution of scores is checked for bunching just above 55, 70 and 85, above all among plans developers submit and among plans scored again after edits.
8.3 Fairness. Each edition publishes the distribution of scores and of bands by:
market and climate zone;
bedroom count;
house or flat;
floor;
price band, where known.
It shall confirm that a well-designed one-bedroom or affordable home, in every climate, can reach Excellent.
8.4 Badges (when offered) require all of these:
the lower end of the likely range inside the band shown;
an unedited plan of documented origin;
the whole score shown, with no choosing which numbers appear;
a link to the method.
9Names and marks
9.1 What the standard is not called. It is never styled an "Indian Standard" (BIS Act 2016 §26) or an "American National Standard".
9.2 How it is cited. "True Inch Standard for Homes, Part 2-IN, edition 1.0", or "TIS-IN 1.0".
9.3 How a score is stated. "True Inch Score 74 (TIS-IN 1.0)".
10What is published
The method page carries:
the current Parts, and every earlier edition;
this document;
the bench results of the engines;
the reference set's composition;
the validation protocol and its results;
the tables of changes between editions;
the interpretations;
the advisory board and its advice;
the comment synopses.