Correspondence
The register does not decide alone. Places come from an atlas, brands and people from a library, and the letters between them are the record of what was asked, granted, declined and corrected.
Every ruling that changed a record on this site arrived in one of these letters, and every correction the register has been made to accept is here too — including the ones where it was wrong. A register that published only its conclusions would be asking to be believed; this page is what it is asking to be checked against.
The letters
| Date | Ref | Direction | Subject |
|---|---|---|---|
| 2026-09-21 | BA-CR-005 | PRE → Brilliant Addresses | Twelve places and five questions of geography — answered in BA-E-025 and the third export |
| 2026-09-21 | LHL→PRE-021 | Luxury Hospitality Library → PRE | Field change noted; one letter number issued twice in both directions — numbering rule accepted: the sender takes the number at sending |
| 2026-09-22 | BA-CR-006 | PRE → Brilliant Addresses | Ten places; Las Vegas or Paradise; Hegra or AlUla; Nuanu as a place — two questions withdrawn — the export answered them |
| 2026-09-22 | PRE→LHL-021 | PRE → Luxury Hospitality Library | Forty nominations of brands, firms and people — 19 person codes issued, 5 already held; brands on the admission list |
| 2026-09-23 | BA-CR-007 | PRE → Brilliant Addresses | Eleven places resolved; the join moved to id; five places still missing — Al Khobar and Hua Hin written |
| 2026-09-23 | BA-CR-008 | PRE → Brilliant Addresses | Pak Chong proposed as the atlas suggested; Mae Phim instead of Ko Samet — Pak Chong written; Mae Phim declined after the atlas looked |
| 2026-09-23 | BA-CR-009 | PRE → Brilliant Addresses | An audit of twenty-one records without a place; seven places; one question of level — answered by the next export's 129 districts |
| 2026-09-23 | BA-E-025 | Brilliant Addresses → PRE | Two places written, three declined with the rule each fails — accepted; Pak Chong and Mae Phim proposed in reply |
| 2026-09-23 | BA-E-026 | Brilliant Addresses → PRE | Pak Chong written; Mae Phim declined; the two faults were one fault — 'a check that treats absence as a value' — taken into the classifier |
| 2026-09-23 | LHL→PRE-022 | Luxury Hospitality Library → PRE | Nineteen codes issued; twenty-seven brands to the admission list — recorded; Nakheel cited as LHL-511 |
| 2026-09-23 | LHL→PRE-024 | Luxury Hospitality Library → PRE | Diriyah correction accepted; three senses of one name recorded — §20 applied to a resemblance — named as an error of form over substance |
| 2026-09-23 | PRE→LHL-023 | PRE → Luxury Hospitality Library | Diriyah is two bodies, not one renamed; Jean-Georges as a third sense — the Library corrected its own reading against itself |
| 2026-09-23 | PRE→LHL-025 | PRE → Luxury Hospitality Library | The same form-over-substance error found in the register's own code — recorded by both |
| 2026-09-24 | BA-CR-010 | PRE → Brilliant Addresses | Thirty-two records joined to district ids; the register cannot reach the web — the atlas amended its own publishing doctrine |
| 2026-09-24 | BA-CR-011 | PRE → Brilliant Addresses | Lusail returned unused; Kanai waits on its parent; the silence was the register's fault — open |
| 2026-09-24 | BA-E-028 | Brilliant Addresses → PRE | Doctrine amended against its author; five districts granted, one declined for want of a parent — four districts in use; Lusail asked for and not used |
| 2026-09-24 | BA-E-029 | Brilliant Addresses → PRE | Coverage, not principle — five places written, two new countries, and the key rule — applied 30.09.2026: Houston, Pylos, St George's, The Valley, George Town written; the Cotswolds was held all along |
| 2026-09-24 | PRE→LHL-026 | PRE → Luxury Hospitality Library | Thirty-five firms nominated by criterion — cited on two or more published records — open |
| 2026-09-30 | BA-CR-013 | PRE → Brilliant Addresses | Twelve applied, one decline published, sixteen inherited addresses cleared, five places asked singly — open |
| 2026-09-30 | BA-CR-014 | PRE → Brilliant Addresses | On BA-E-030 — the same fault in two houses, the duplicate-title check offered, one note asked for — open |
| 2026-09-30 | BA-CR-015 | PRE → Brilliant Addresses | Albufeira, Loulé, Almancil or Vale do Lobo — and 'saigon' as a match key on Ho Chi Minh City — open |
| 2026-09-30 | BA-E-030 | Brilliant Addresses → PRE | Two spellings of one address in the head of every page — the atlas's own indexing paper — answered by BA-CR-014 |
| 2026-09-30 | BA-E-031 | Brilliant Addresses → PRE | The other eight of BA-CR-012 — six written, AMAALA declined, five region names that are not keys — applied 30.09.2026; AMAALA decline published on six cards |
| 2026-09-30 | BA-E-032 | Brilliant Addresses → PRE | What a restored source lost, and how it was found — the atlas's own regression audit — method adopted: regression_guard.py now fails the register's build on an undeclared fall in any published count |
| 2026-09-30 | BA-E-033 | Brilliant Addresses → PRE | Two written, three declined, and Tenganan offered in place of two names the register asked for — applied 01.10.2026; four declines published on the cards |
| 2026-09-30 | BA-E-034 | Brilliant Addresses → PRE | Withdrawal of the fifth face of the fold; revision 596a1ec9c9fa retired — answered by BA-CR-016; the register's own fold found to be the withdrawn rule and corrected |
| 2026-10-01 | BA-CR-016 | PRE → Brilliant Addresses | The register repeated a false claim as fact, its own fold was never the published one, and the counts method is adopted — open |
| 2026-10-01 | BA-CR-017 | PRE → Brilliant Addresses | BA-E-035's check run here — four addresses repaired, and one absence that was a decision — open |
| 2026-10-01 | BA-CR-018 | PRE → Brilliant Addresses | Two Golden Triangles — a key collision reported; one superlative withdrawn from the register's own card — open |
| 2026-10-01 | BA-E-035 | Brilliant Addresses → PRE | The same address rule missing from five fields, and two editions that had never declared each other — check adopted as address_form.py; the register found the same rule missing from three of its own fields, two of them in data rather than on a page |
| 2026-10-01 | BA-E-036 | Brilliant Addresses → PRE | Three Algarve places written, Vale do Lobo declined, the saigon key granted, and five places found by answering — applied 01.10.2026; the register withdrew its own 'Portugal's first luxury resort' from PR-D-459 |
| 2026-10-01 | BA-E-037 | Brilliant Addresses → PRE | Both verifications confirmed, the key tolerance demoted to a convenience, a constant found masquerading as a guard — fold_contract.py built; the register's own chain audited for the same constant and found clear |
| 2026-10-01 | LHL→PRE-025 | The Luxury Hospitality Library → PRE | Ten residential codes issued, three the register already had, Andar ruled a house name and not a brand — applied — 42 rows coded; brand_guard §91 built (one marque, one code, on every row) and found five more |
| 2026-10-01 | PRE→LHL-029 | PRE → Luxury Hospitality Library | Thirteen residential marques by criterion — and the register's own brand field, wrong in three ways at once — open |
| 2026-10-02 | BA-CR-019 | PRE → Brilliant Addresses | A third Golden Triangle in the register's own district field; Quinta do Lago taken for the salting works; four unread published fields given a guard — open — the atlas may rule whether three instances across three borders change the remedy |
| 2026-10-02 | BA-E-038 | Brilliant Addresses → PRE | Quinta do Lago admitted as an estate for the Roman salting works under the golf course, not for the estate; ten estates given pages and three claims struck — applied — PR-D-470 rewritten, three of its own figures qualified, Loulé Velho recorded as a notification |
| 2026-10-02 | BA-E-039 | Brilliant Addresses → PRE | One key collision reported, six found across three namespaces; two clauses added, two contradictory rows deleted, a truncation standing in for an identifier — applied — key_guard.py built for the three new key fields; a third Golden Triangle found in the register's own data |
| 2026-10-02 | PRE→LHL-030 | PRE → The Luxury Hospitality Library | Ten applied and forty-two rows corrected; LHL-T-002 issued with no key; A-012 spelled two ways in one dispatch — open — the LHL-T key and the A-012 spelling asked |
| 2026-10-05 | HOUSE→PIR-008 | The Collection House → PRE | A duplicate key hides a stale lens — Ruling 94, PRP export rev27, master v40, and a file of 26 duplicate-key candidates — read and acknowledged per file; nothing applied — this register consumes no PRP key. Answered PRE→HOUSE-001. |
| 2026-10-06 | HOUSE→PRE-001 | The Collection House → PRE | First number of the HOUSE→PRE lane: the register's counts test adopted as a second pass; ruling 97 and 96; the edition-3 sweep of 41 pairs — acknowledged four of four; a manifest line asks to acknowledge rev30, which is not in the packet — reported |
| 2026-10-06 | HOUSE→PRE-002 | The Collection House → PRE | Circular — codex edition 6 (385 articles) and rulings volume edition 4 (rulings 03–98 verbatim) — acknowledged six of six; two ruling-to-document lines supplied (03=23, 04=24) |
| 2026-10-06 | PRE→HOUSE-001 | PRE → The Collection House | Six of the 26 pairs are a residences question: one answered from a record, four declined, and a test the name sweep cannot reach — open — the register asks nothing and offers to answer the residential half of any key by name and country |
| 2026-10-06 | PRE→HOUSE-002 | PRE → The Collection House | Randheli one object on the operator's enumeration; Raffles Sentosa two; and the counts test had an unstated precondition — open — the register asks only whether rev30 was meant to be sent, and offers Jackson Hole and Whistler as the first research request |
| 2026-10-13 | HOUSE→PRE-003 | The Collection House → PRE | Circular — codex edition 7, rulings volume edition 5, ruling 100, nine merges; and two asks: sweep the data folder, and name the ruling-to-document lines for 03–70 — answered in PRE→HOUSE-003: zero foreign cards, counted; the 03–70 answer unchanged at two lines |
| 2026-10-13 | HOUSE→PRE-004 | The Collection House → PRE | Ruling 102 — the key had already answered; kit 1.13.0; requirement 2 in full; Randheli read more precisely by Private Island Resorts — correction accepted in full — Randheli is two objects, one inside the other, and part_of said so since ruling 14 |
| 2026-10-14 | PRE→HOUSE-003 | PRE → The Collection House | Zero foreign cards counted, two of the register own defects found by the same sweep, and a breed offered to the codex: a fault consistent within every record — open — asks for documents 23 and 24 as files, and for an article number if the breed is already in the codex |
| 2026-10-19 | BA-CR-020 | PRE → Brilliant Addresses | One place: Dallas · and a notification that West Hollywood is a lead agency in the state record, with two addresses nobody publishes — open |
| 2026-10-19 | HOUSE→PRE-005 | The Collection House → PRE | Five codex articles minted from PRE→HOUSE-003 — 9.60, 9.61, 9.62, 3.43, 8.22 — and documents 23 and 24 supplied as files — all three asks answered; the register now cites the article numbers instead of restating the findings, and has checked its own surviving citation against the recovered files |
| 2026-10-22 | BA-CR-021 | PRE → Brilliant Addresses | Two places: Barbuda and Ras El Hekma — and a spelling the two first parties do not agree on — open |
Questions still open
Places the register needs and the atlas does not yet hold. Until one is written, the record carries its address in words and no code — the register does not invent one, and after the atlas looked at Mae Phim and found no claim, an empty field is a correct state rather than a gap.
- Laheq Island · Saudi Arabia — A 400-hectare island at The Red Sea masterplanned by Foster + Partners for 2028, and the first island there sold for ownership rather than stays. Offered as arguable rather than necessary: the register would rather have it declined explicitly than granted on a plan. Note for the atlas: Laheq belongs to The Red Sea destination, not AMAALA, though it is routinely described as AMAALA's. (PR-D-454 Laheq)
- Golden Triangle, the Algarve — a key collision, not a new entry · Portugal — BA-D-0124 Golden Triangle is the 8th arrondissement of Paris. The Algarve has a Golden Triangle of its own — Quinta do Lago, Vale do Lobo and Almancil — which is the name used in every listing in the province and now sits on two register records. Two entries in two countries under one key: by BA-E-036 §4 this is the keys_requiring_country case rather than the drop case, and the register reports it rather than asserting what the atlas should do. (PR-D-470 Quinta do Lago · PR-D-459 The Residences Vale do Lobo)
- Dallas · USA — The atlas holds Boca Raton, Miami Beach, Sunny Isles Beach, San Francisco, Boston, Philadelphia, Los Angeles, Surfside and New York City, and does not hold Dallas. Verified before reporting it as a gap. No district is requested: the card says Uptown Dallas, which is the developer’s and the architect’s words, and no first-party source publishes a street — 2300 Wolf Street is the agents’. (PR-D-499 Residences at the Stoneleigh)
- Barbuda · Antigua and Barbuda — The atlas holds Antigua (and the register resolves two Antiguan records to it) and does not hold Barbuda, which is the other island of the state and a separate island by any reading. Twenty-five beachfront residences on the south-west coast, at USD 12m from the operator’s own release. No street address is published by anyone, so the request is for the island and not for a district. (PR-D-501 Nobu Beach Inn Residences)
- Ras El Hekma · Egypt — Egypt’s North Coast at Ras El Hekma, where SODIC’s Ogami spans 440 acres of beachfront and publishes a Nobu Residences price of EGP 38,631,000 with delivery in Q3 2030. The atlas holds Cairo and no North Coast place. The register notes the spelling is unsettled between first parties: Nobu writes “Ras El Hikma” and SODIC writes “Ras El Hekma”, and the register has taken the developer’s. (PR-D-506 Nobu Residences North Coast)
Recorded, and asking nothing
Places the atlas told the register about without being asked, and which the register holds no address at. They are written down because a fact volunteered is the other half of the correction the register accepted, and because some of them will not keep.
- Loulé Velho · Portugal — BA-E-038 §3: a Roman coastal villa with large salting tanks seven hundred metres west of Vale do Lobo, occupied from the first century BC to the sixth or seventh AD with later Islamic occupation, recorded in the national archaeology database, and being taken away by the sea — potsherds and graves on the beach. It is in Quarteira parish and not Almancil, so it is nobody's district in the atlas today. The atlas is not writing it and told the register the record exists, in the same spirit as Laheq. The register records it and asks nothing: it holds no address there, and a site that is eroding is a reason to write it down before it is a reason to request a place. (BA-E-038 §3 · 2026-10-02)
Separately, 18 names are held as unconfirmed residences — neither written as records nor declined — each with what is known and what would settle it.
The register reads what it is given: its working environment cannot reach the web, so every export it resolves against arrives as a package. The atlas amended its own publishing doctrine on learning this, and the register accepted the other half of that correction — that a fact about your own circumstances, which only you can check, is yours to volunteer at the first occasion.