AGENTS.md

@@ -73,7 +73,7 @@ In [docs/decisions.md](docs/decisions.md):
 - After a row, a path names a column or a linked table
 - A selected row is fixed, not drawn, and an unnamed read beside it is refused
 - A link column after a row steps up to the parent row, and a path may not end on it
-- A selector names a row by its key alone
+- A table selects by one key column, a code, never a free-form name
 - A parent row with no child row is a load error
 - The choice-of-rows fence guards a data file's root, and requires string fields
 - A table is a record of string columns
@@ -94,7 +94,6 @@ In [docs/decisions.md](docs/decisions.md):
 - A title is a table under `sex`
 - A table owns the spelling of a selector on it
 - No builtin reads the clock, so a date is bounded by days, never by an age
-- A name column without a key resolves inside its parent
 - `misc` is what every locale shares
 - A register's canonical spelling loses to the one its domain writes
 - `misc.territory` is the spine, and a `misc` table naming a territory links to it
@@ -103,7 +102,7 @@ In [docs/decisions.md](docs/decisions.md):
 - `misc.port` selects by number, and carries no `name`
 - `misc.car` is one flat table, not a make linked to its models
 - `misc.territory` names its sovereign in a column, and there is no `misc.country` table
-- `misc.loglevel` is a table keyed by the code, rendering POSIX's keyword
+- `misc.loglevel` is a table keyed by POSIX's keyword, carrying the code
 - `misc.tld` keys carry the leading dot, where other tables key on a bare code
 - `misc.tld` is a table of its own, and `misc.territory.tld` stays a column
 - `misc.territory` carries a currency code, it does not link to `misc.currency`

docs/decisions.md

@@ -385,8 +385,8 @@ parent. A row is selected or drawn alike.
 
 - Where a segment follows, the link column's name reads the parent row, one level per
   step, chained: `city[Oslo].country.name` renders `Norway`, `locality.municipality.name`
-  the drawn locality's municipality, and `geo.SE.locality[Lund].municipality.region.name`
-  Lund's region.
+  the drawn locality's municipality, and
+  `geo.US.locality[1714000].municipality.region.name` Chicago's state.
 - A path ending on it is refused at `New`, `NewTemplate` and `Fake` alike, naming the
   parent's key column and the parent's format spelled as paths: with a `country` format
   of `{name} ({alpha2})`, `city[Oslo].country` names `city[Oslo].country.alpha2` and
@@ -404,13 +404,21 @@ Checked against three simulated template writers from the Audience: all three re
 parent row where a segment followed, and they split between the code and the name where
 the path ended.
 
-## A selector names a row by its key alone
+## A table selects by one key column, a code, never a free-form name
 
 Decided 2026-09-27 by the maintainer, for goal 4's one spelling per result; valid while
-every table either has a `key` or names its rows inside a parent. A table with a `key`
-selects by it alone: `country[NO]` loads, and `country[Norway]` is refused naming
-`[NO]`, found by the row whose cell spells the selector. A table with no `key` selects by
-its `name` inside its parent, which is the key it has.
+the key is the code a user writes, as a database keys a table. `country[NO]` loads, and
+`country[Norway]` is refused naming `[NO]`, found by the row whose cell spells the
+selector. The key is the friendliest code the register holds, `misc.territory[SE]` and
+`misc.currency[EUR]`, and only where it holds none an id the import assigns and never
+moves to another row. There is no `name` option.
+
+- `misc.loglevel` keys on the keyword, `[err]`, which a configuration writes.
+- `misc.tld` keys on the ASCII form, `[.xn--p1ai]`, the one DNS carries.
+- `geo.US.locality` keys on the Census GEOID, `[1714000]` for Chicago.
+- `geo.SE.locality` keys on an assigned id: a postort has no code of its own.
+- A name table, `first-name`, `last-name` and `title`, carries no key, so it is drawn
+  from and never selected.
 
 ## A parent row with no child row is a load error
 
@@ -546,8 +554,7 @@ are a curated proportion rather than bearers anyone counted.
 A reference reaches a table by a path that carries no selector — `sv_SE.person.first`
 reads `first-name` through `sex` — so the walk that resolved a name cannot say where a
 reader would type one. The table's own location can, which is why it keeps its path, and
-why an ambiguity error names `sv_SE.sex[f].first-name[Kim]` rather than the table's own
-name.
+why an error names `geo.SE.municipality[1281]` rather than `municipality[1281]`.
 
 ## No builtin reads the clock, so a date is bounded by days, never by an age
 
@@ -555,13 +562,6 @@ An `age(min,max)` would make a seeded fixture change with the day it runs on, wh
 what a seed exists to prevent; a birthdate for someone 20 to 60 is
 `date(1966-01-01,2006-12-31,…)`, re-pinned as any fixture is.
 
-## A name column without a key resolves inside its parent
-
-A given name both sexes carry is a row under each, so `name` cannot be the key; the
-parent's row tells the two apart, `sex[f].first-name[Kim]`, the ambiguity error spells
-each row inside its parent, and a name repeating inside one parent row is refused at
-load, since nothing could then select it.
-
 ## `misc` is what every locale shares
 
 A category whose facts differ by country belongs in that country's locale, read from the
@@ -627,7 +627,7 @@ own family; a test proves every value names a row instead. A territory the regis
 records no state for stands alone, which is `EH` alone, and naming one for it would be a
 claim fejkdata has no business making.
 
-## `misc.loglevel` is a table keyed by the code, rendering POSIX's keyword
+## `misc.loglevel` is a table keyed by POSIX's keyword, carrying the code
 
 A flat list of names carries neither the code a PRI encodes nor a selector reaching it,
 and goal 3 draws the two as one fact. The canonical spelling losing to the one its

todo.md

@@ -178,23 +178,28 @@ by a `data-import/` script, as README goal 12 asks.
   cell `NO`, as `locality.municipality` does, which `List` stops advertising. Replace the
   README's `city[Oslo].country` example with
   `fejkdata -d ./mydata 'city[Oslo].country.name'`, rendering `Norway`.
-- Select a row by its key alone, as the decision "A selector names a row by its key
-  alone" states, once the open question on each table's key is settled:
-  `geo.SE.municipality[Lund]` and `misc.territory[Sweden]` select a row today.
+- Key every table as the decision "A table selects by one key column, a code, never a
+  free-form name" states: `geo.SE.municipality[Lund]` and `misc.territory[Sweden]` select
+  a row today.
   - Have a selector that misses name the key of every row whose cell spells it,
     matched without case, else say the table selects by its key:
     `misc.territory[Sweden]` and `misc.territory[se]` name `[SE]`, `misc.protocol[tcp]`
-    its key, and `misc.loglevel[Error]` the row a column spells `Error` in.
-  - Refuse a `name` option beside `key`, which then selects nothing, drop it from the
-    shipped tables, and delete the load check refusing a name that spells another row's
-    key.
+    its key, and `misc.loglevel[Error]` `[err]`.
+  - Delete the `name` option, and the load checks it carries: a name spelling another
+    row's key, and a name repeating inside its parent.
+  - Key `misc.loglevel` on `keyword`, and `geo.SE.locality` on an id `data-import/geo-se.py`
+    assigns and keeps across imports, linking `street` and `postal-code` by it.
+  - Drop `key` from `last-name` and `sv_SE.title`.
   - Rewrite what selects by name: in README.md, "selected by key or name",
     `misc.currency[Euro]`, "selects by either spelling", `misc.tld[.рф]`,
-    `misc.loglevel[err]` and "the code and the keyword do", the `geo` table's "by code
-    or name", "by USPS abbreviation or name" and "by GEOID or name",
-    `geo.SE.region[Skåne län]`, `geo.US.region[IL].locality[Springfield]`, "selection by
-    name", the `name` option's paragraph and "`[key]` or `[name]`"; in CHANGELOG.md, "a
-    path selects a row by key or name", "a name spelling another row's key" and
+    `misc.loglevel[3]`, `misc.loglevel[err]` and "the code and the keyword do", the `geo`
+    table's "by code or name", "by USPS abbreviation or name", "by GEOID or name" and
+    "by name", `geo.SE.locality[Lund]`, `geo.SE.region[Skåne län]`,
+    `geo.US.region[IL].locality[Springfield]`, "selection by name",
+    `first-name[Taylor]`, `first-name[Kim]`, `sv_SE.title[dr]`, the `name` option's
+    paragraph and "`[key]` or `[name]`"; `locality[Lund]` in docs/decisions.md; in
+    CHANGELOG.md, "a path selects a row by key or name", "a name spelling another row's
+    key", "a name repeating inside one parent row", "A `name` without a `key`" and
     `misc.tld[.рф]`.
 - Render selected rows as the decision "A selected row is fixed, not drawn, and an
   unnamed read beside it is refused" states: `New` refuses
@@ -238,16 +243,6 @@ by a `data-import/` script, as README goal 12 asks.
 
 ### Open questions to settle
 
-- Decide which spelling each shipped table selects by, now a selector names its key
-  alone. Keeping today's key loses `geo.US.region[IL].locality[Springfield]` to a
-  numeric GEOID, `misc.loglevel[err]` to `[3]`, `misc.tld[.рф]` to `[.xn--p1ai]`,
-  `misc.currency[Euro]`, `misc.protocol`'s second spelling and
-  `geo.SE.region[Skåne län]`. The tables carrying both today: `geo.SE` and `geo.US`
-  region, municipality and US locality, `misc` currency, httpstatus, language,
-  loglevel, mimetype, protocol, territory and tld, and both locales' `sex`. Decide too
-  whether the `name` option, left meaning "the key inside a parent", keeps its name.
-  Record each answer in `docs/decisions.md`, amending the `misc.tld` and `misc.loglevel`
-  entries it changes.
 - Decide whether a goal 5 violation joins the comprehension gate's exception list: the
   `{/sel}|{/sel}` panic waits behind the whole round at 5.9, and
   `{/geo.SE.locality}|{/geo.SE.municipality}` mispairs the shipped data meanwhile, though a