feat: rule Q58 and close Q57; §4 lists all six areas; docs/03's checklist cites §4 instead of restating it

Four rulings from Pouya, plus what implementing them turned up.

Q58 — RULED, and he attributed the ambiguity to his own document: "3.5 was meant
as the TOTAL time committed, of which 2 is preparation — leaving 1.5 hours in the
room. Your arithmetic caught it: if prep sat inside, 3.5 and 7 wouldn't be
exactly 2x, because preparation doesn't scale with session length." The card now
reads: half day up to 3 hours of session, fee includes up to 2 hours of
preparation, $2,000; full day up to 6 hours, up to 3 hours of preparation,
$4,000. docs/07's own research table corroborates 3 and 6 — Patey and Zuber both
publish those hours, and ADR Chambers' roster rate counts preparation separately
from "up to three hours of mediation". One provenance note under R14: he recalled
"all or part of 3 hours" as their wording; the committed extract carries the
hours but not the phrase, so docs/07 cites the hours and attributes the phrase to
nobody.

Two things fell out of the ruling that the instruction did not name, and both
were defects. docs/07 §All parameters confirmed was itself prescribing the flat
"including 2 hours of preparation" — the sentence /for-parties/ was built
against, so the spec was generating the defect. And the cap had to reach the
copy: "including up to 2 hours". FEES.mediation.*.hours is corrected 3.5 -> 3 and
7 -> 6; it had no consumer in src/ while the question was open, which is the only
reason no page was ever wrong. /fees/ is unblocked for step 9 on the question Q58
asked.

Q57 — CLOSED with no seventh undertaking. "A reader assumes the outcome, and the
obvious undertaking adds nothing a reader doesn't already infer." The TODO(pouya)
is replaced by the ruling where the question was; src/ now carries zero live
TODO(pouya) markers.

§4's mediation row lists all six published areas. Q56's ruling had named five,
which was four areas plus the word "commercial" — a scope descriptor, not a
seventh area. The hedge is struck on his instruction; the clause saying the six
are not the authorised subject-matter list is restored, because his ruling
supplied a correct value and did not close Q35(c)'s class. Split-stamped.

docs/03's compliance checklist now names what to look for on a page and which §4
row decides it, never the bar's own wording. 12 items before, 12 after — a
structural fix, not a coverage change.

Thirteen review findings across two rounds, all applied, none declined. Three
were mine to own. The capped-form rule was written and then applied to one
surface: /mediation/ shipped an uncapped form in words no barred-string grep
could reach, site.ts quoted a docs/07 sentence Q58 had just deleted, and §9's
Q15/Q16/Q17 row prescribed the flat form — which is what a later implementer
building /fees/ reads. A derived fee term was asserted as applied fact in the
document that is the authority on money: "overtime begins after 3 h and 6 h" is
in no ruling. Struck, and opened as Q59.

And round 2 caught the arithmetic in round 1's own fix. The full-day route is
flat $4,000 until hour 6, so generalising it as 500n+1000 for all n>=3 was valid
only from 6 h, and "cheaper by $500 at every length" was wrong across the whole
3-6 h band. The real spread is $2,000 at three hours narrowing to $500 from six
on — up to four times larger, and largest exactly where a half-day booking
overruns. Written into docs/07 §Recorded dissent and §12's R5 row, which is where
the 12-month fee review will read it. Round 1's fix for the missing consequence
also published the overtime rate on a page that now states an unambiguous cap,
defining the trigger by adjacency with no other quantity for it to attach to; the
rate came off the page.

R11 at the step 6 -> 7 boundary: 13 of 14 pins current. §7's TypeScript hold
named one gate and there are two — typescript-eslint requires <6.1.0, tighter
than @astrojs/check, so the recorded removal trigger was unreachable. Both are
now named.

Verified: check 0 errors, lint 0, build 0 (14 pages), check:claims 0, npm audit
0, minifier tripwire clean, zero JS shipped, all copy present with JavaScript
disabled. Lighthouse not run — tool unavailable until step 7.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Md3GndFqWPzK78xAoebsg5

docs/07-fees.md

@@ -97,11 +97,95 @@ All figures **plus HST**.
 
 | Item | Fee |
 |---|---|
-| Half day — up to 3.5 h, including 2 h preparation | **$2,000** |
-| Full day — up to 7 h, including 3 h preparation | **$4,000** |
+| Half day — up to 3 hours of session. Fee includes up to 2 hours of preparation | **$2,000** |
+| Full day — up to 6 hours of session. Fee includes up to 3 hours of preparation | **$4,000** |
 | Each party beyond two | **$500** |
 | Overtime, per hour | **$500** |
 
+⚠️ **THE TWO ROWS ABOVE WERE AMBIGUOUS UNTIL 2026-08-31, AND THE AMBIGUITY WAS
+IN THIS DOCUMENT RATHER THAN IN ANY COPY — Q58, ruled by Pouya.** They read
+*"Half day — up to 3.5 h, including 2 h preparation"* and *"Full day — up to
+7 h, including 3 h preparation"*. Read literally that makes 3.5 the **whole
+billed envelope** and the time in the room **1.5 h**. Pouya's ruling: *"3.5 was
+meant as the TOTAL time committed, of which 2 is preparation — leaving 1.5 hours
+in the room… The intended reading is the market's, and my wording obscured
+it."*
+
+**What found it was arithmetic, not reading:** 3.5 and 7 are exactly 2×, which
+they would not be if preparation sat inside them, **because preparation does not
+scale with session length**. Under the literal reading the room time was 1.5 h
+and 4 h, which is not 2× anything.
+
+**The corrected numbers are corroborated by this file's own research table**, one
+section up — which is the reason they are 3 and 6 rather than a round guess:
+
+- **Patey** publishes **3 h** half-day and **6 h** full-day in *both* tiers.
+- **Zuber** publishes **3 h** video half-day and **6 h** full day.
+- **ADR Chambers**' roster rate covers *"one half hour of preparation time per
+  party **and** up to three hours of mediation"* — preparation counted
+  **separately from** a three-hour session, which is the shape this card now
+  has.
+
+Pouya: *"Selling 1.5 hours of room time as a half day would be an outlier nobody
+would recognise."* ⚠️ **One provenance note, because this file is the authority
+on money:** he recalled Patey and Zuber as publishing *"all or part of 3 hours"*.
+The extract above records their **hours** but not that phrase, so **the hours are
+what this file relies on** — do not quote the phrase as theirs.
+
+**THE PREPARATION ALLOWANCE IS CAPPED, AND MUST BE PUBLISHED AS CAPPED.**
+*"Including **up to** 2 hours of preparation."* Never *"including 2 hours"*,
+which reads as a flat entitlement, and never *"preparation included"*, which
+sells an uncapped one. See §All parameters confirmed below.
+
+**Consequences, applied 2026-08-31:** `FEES.mediation.*.hours` in
+`src/data/site.ts` means the **session** and is corrected 3.5 → 3 and 7 → 6, and
+**`/fees/` is unblocked for build step 9** on the question Q58 asked.
+
+⚠️ **BUT WHERE OVERTIME BEGINS IS STILL NOT STATED, AND IT IS NOT SAFE TO INFER
+— §9 Q59, OPEN.** A first pass at this paragraph asserted that *"the overtime
+rate begins after 3 h and 6 h of session respectively"*. **That is a derived fee
+term, not part of the ruling**, and `adversarial-reviewer` was right to strike
+it: this file is the authority on money and `/fees/` is now cleared to publish
+from it. Nothing in the card, in Q58's ruling, or in Q15–Q17's answer says when
+the hourly rate starts.
+
+**And the reason it cannot be quietly chosen is that the choice is visible in the
+arithmetic.** Take the trigger as the session cap. The half-day route costs
+`2000 + 500 × max(0, n − 3)`; the full-day route is **flat $4,000 until hour 6**
+and `4000 + 500 × (n − 6)` after it:
+
+| Session run to | Book a half day + overtime | Book a full day | Gap |
+|---|---|---|---|
+| 3 h | **$2,000** | **$4,000** | **$2,000** |
+| 4 h | $2,000 + 1 × $500 = **$2,500** | **$4,000** | **$1,500** |
+| 5 h | $2,000 + 2 × $500 = **$3,000** | **$4,000** | **$1,000** |
+| 6 h | $2,000 + 3 × $500 = **$3,500** | **$4,000** | **$500** |
+| 7 h | $2,000 + 4 × $500 = **$4,000** | $4,000 + 1 × $500 = **$4,500** | **$500** |
+
+**Booking a half day and paying overtime is cheaper at every length — by $2,000
+at three hours, narrowing to $500 from six hours on — and the full-day rate is
+never the cheaper choice.** The gap is in D14's figures rather than in the
+trigger: the half-to-full step is $2,000 and three hours of overtime is $1,500.
+
+⚠️ **A FIRST PASS AT THIS PARAGRAPH GOT THE ARITHMETIC WRONG, AND WRONG IN THE
+DIRECTION THAT MATTERED.** It generalised the full-day route as `500n + 1000` for
+all `n ≥ 3`, which is **only valid from hour 6** — the route is flat until then —
+and concluded *"$500 cheaper at every length"*. The real spread is **up to four
+times larger and is largest at three to five hours, which is the band a half-day
+booking actually overruns into.** The document's own table contradicted the
+formula three lines above it. Found by `adversarial-reviewer` on round 2, in the
+change set that wrote it. *A measurement is a claim about your instrument, and a
+formula is an instrument.*
+
+**Moving the trigger out to the 5 h / 9 h billed envelope does not fix it and is
+not uniformly better either** — the gap stays at $2,000 through five hours and is
+$1,500 at six, worse than the session-cap trigger there, but it closes to **zero**
+from nine hours on, where the session-cap trigger holds a permanent $500. So the
+two triggers trade one band against another and neither removes the anomaly. **It
+is not a defect this file can fix by picking a trigger, which is why the trigger
+goes to Pouya and the step goes to R5** — see §Recorded dissent below, where it is
+written out for the 12-month review rather than left in this footnote.
+
 ### Arbitration
 
 Sole, party-appointed and co-arbitration appointments **in commercial matters**
@@ -182,10 +266,16 @@ page is an offer.
 ### All parameters confirmed
 
 Q15, Q16, and Q17 were closed on 2026-08-26. **Preparation time is bundled and
-must be stated on the page** — "including 2 hours of preparation", "including
-3 hours of preparation". Do not quietly fold it into the hours figure. At these
+must be stated on the page, IN HOURS AND AS A CAP** — "including **up to** 2
+hours of preparation", "including **up to** 3 hours of preparation". Do not
+quietly fold it into the session figure, and do not drop the "up to": the
+allowance is capped, so the unqualified form sells an uncapped one. At these
 rates, saying preparation is included is the selling point, not a footnote.
 
+*(The "up to" was added 2026-08-31 with Q58's ruling. This paragraph previously
+prescribed the flat form, and `/for-parties/` shipped it — the one page written
+for a reader with no counsel to catch it.)*
+
 ---
 
 ## Recorded dissent — for the 12-month review (R5)
@@ -227,6 +317,35 @@ unrecoverable, and it is far easier to add a lower tier later than to raise a
 headline rate. Setting the ceiling first and discounting privately preserves
 more optionality than the reverse.
 
+**And a fourth item, added 2026-08-31: the half-day-plus-overtime route is
+strictly cheaper than the full-day rate, at every session length.** This is the
+one item in this section that is arithmetic rather than judgement, so it is the
+one the review can settle without new market data.
+
+| Session run to | Half day + overtime | Full day | Gap |
+|---|---|---|---|
+| 3 h | $2,000 | $4,000 | **$2,000** |
+| 4 h | $2,500 | $4,000 | **$1,500** |
+| 5 h | $3,000 | $4,000 | **$1,000** |
+| 6 h | $3,500 | $4,000 | **$500** |
+| 7 h | $4,000 | $4,500 | **$500** |
+
+*(Session-cap trigger; the trigger itself is `AGENTS.md` §9 **Q59**, open.)* The
+cause is the relationship between two of D14's own numbers rather than anything
+about the trigger: **the half-to-full step is $2,000 and three hours of overtime
+is $1,500.** Any trigger leaves a gap; the envelope trigger closes it only from
+nine hours on.
+
+**What the review has to decide is whether that is a choice.** It is defensible
+as one — a full-day booking buys certainty and a reserved diary, and a client who
+knows they need six hours may prefer to pay $500 for not having to watch the
+clock. It is also exactly the kind of thing counsel comparing published cards
+finds in under a minute, and this practice's whole pricing argument is that the
+card is published in full and means what it says. **The lever, if it is not a
+choice:** either raise the overtime rate so the routes converge, or narrow the
+half-to-full step. Both are rate moves, which is why they belong here and not in
+a footnote to the card.
+
 ---
 
 ## Sources