Remove ratio_circulation: the score is already a ratio

Owner: "I think a corridor could be worth a sixth of a room, this is ok. maybe
we should dump the ratio_circulation altogether if there is already a pressure
in circulation caused by the cost benefit ratio per msq. this is the kind of
thing we want to root out of the scoring model: anything that is double
counting, or using a gaussian where a linear ramp is appropriate, etc."

value_circulation = 50 stands; hxi's rate question is closed.

The duplication argument is stronger than it first looks. score = value/cost is
already a ratio, so the per-m2 economics (50 against a build cost of 200) is not
merely an absolute pressure -- adding corridor moves value/cost by an amount
that depends on how much of the building is already corridor. It is ALREADY
proportional. ratio_circulation said the same thing again as a whole-building
multiplier, on a curve where twice the corridor is far more than twice as bad.

Correction to my own first measurement: I overrode ratio_circulation and got
scores going DOWN when a <=1 multiplier was removed, which is impossible. All
four corpus programmes DECLARE ratio_circulation, so the CONF_DEFAULTS value I
had changed was never in play and the two arms differed only in sigma. Same
trap as value_supported in 39.19.

Against the keep-it case, recorded because it is the one real argument: three of
four declare a POSITIVE target (harbor/maple 0.08, health-centre 0.10), making
the term formally two-sided rather than "less is better". It does not survive
the numbers -- the lower side is worth at most 13.3% on the large programmes
against 99% on the upper side, and "a building needs some circulation" is
enforced structurally by access and connectivity, which no amount of value can
buy off.

Disabled in CONF_DEFAULTS and the four corpus configs, each with the reason
inline and a note that a [target, sigma] pair re-enables it. Fail sets
unchanged; it was always a value multiplier. Scores +42% to +7712%.

39.24 also sweeps every remaining term against the owner's two tests. Verdicts:
perpendicular, proportion, width, crinkliness, access, size's lower side,
ratio_outside, staircase volume and the count/limit fails are all sound.
Filed as homemaker-py-dpt: size's UPPER side (cost already charges area; 82% of
size fails are over-target), the minimum-internal-area factor (a third
statement of "build the rooms"), the 0.5**n_fails curve (a ruling, not a
measurement), and two dead paths -- ratio_public/private_outside, which no
config declares, and the daylight factor pinned to 1.0 since the descope.

Also updated a test I added last turn which asserted ratio_circulation was the
second charge; it now pins that the linear ramp is the ONLY one.

419 passed.

Closes homemaker-py-hxi.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MJ84Feep79Hhm3E4zZJmnB

DESIGN.md

@@ -7971,3 +7971,83 @@ untouched and unruled: `value_circulation` = 50 against `value_inside` = 300, a
 room per square metre is a design judgement, not something measurement settles,
 and the linear ramp the owner describes is only as steep as that number makes
 it. Left open on the bead.
+
+### 39.24 `ratio_circulation` removed, and a sweep for the rest of its kind (`homemaker-py-hxi`)
+
+Owner, closing the rate question and opening a general one:
+
+> I think a corridor could be worth a sixth of a room, this is ok. maybe we
+> should dump the ratio_circulation altogether if there is already a pressure in
+> circulation caused by the cost benefit ratio per msq. this is the kind of
+> thing we want to root out of the scoring model: anything that is double
+> counting, or using a gaussian where a linear ramp is appropriate, etc.
+
+So `value_circulation = 50` stands, and `hxi`'s rate question is closed.
+
+**`ratio_circulation` is a duplicate, and the argument is stronger than it first
+looks.** The score is `value / cost` — already a ratio. So the per-m² economics
+(50 against a build cost of 200) is not merely an *absolute* pressure against
+corridor area: adding corridor moves `value/cost` by an amount that depends on
+how much of the building is already corridor. It is **already proportional**.
+`ratio_circulation` states the same thing a second time, as a whole-building
+multiplier, on a curve where twice the corridor is far more than twice as bad.
+
+**A correction to my own first measurement.** I initially measured this by
+overriding `ratio_circulation` and comparing — and got scores going *down* when
+a ≤1 multiplier was removed, which is impossible. The cause: all four corpus
+programmes **declare** `ratio_circulation` themselves, so the `CONF_DEFAULTS`
+value I had changed was never in play, and my "with" arm was accidentally
+comparing two different sigmas. The same trap as `value_supported` in §39.19.
+Measured properly, disabling it raises every score, by +42% to +7712%
+(programme-house s2, which is 44% circulation).
+
+**And the declared values are not what the default suggests:**
+
+| programme | declared | worst the *lower* side can cost |
+|---|---|---|
+| harbor-house | [0.08, 0.15] | 13.3% |
+| maple-court | [0.08, 0.15] | 13.3% |
+| health-centre | [0.10, 0.12] | 29.3% |
+| programme-house | [0.00, 0.15] | 0% |
+| `CONF_DEFAULTS` | [0.00, 0.20] | 0% |
+
+Three of the four target a *positive* circulation fraction, so the term is
+formally two-sided, not "less is better" — and a two-sided target is exactly
+what a gaussian is for. That is the one argument for keeping it. It does not
+survive the numbers: the lower side ("a building needs some circulation") is
+worth at most 13% on the two large programmes, while the upper side reaches
+99%, and the lower side's job is done structurally anyway by the access and
+connectivity checks, which no amount of value can buy off. What is left is an
+upper-side penalty, and the upper side is the duplicated one.
+
+**Shipped:** `ratio_circulation` disabled — `None` in `CONF_DEFAULTS` and in the
+four corpus `patterns.config` files, each with the reason inline and a note that
+a `[target, sigma]` pair re-enables it. Fail sets are unchanged; it was always a
+value multiplier, never a fail source.
+
+**The sweep.** Every term in the objective, against the two tests the owner
+named — does it double-count, and is a gaussian the right curve for what it
+measures?
+
+| term | measures | shape | verdict |
+|---|---|---|---|
+| perpendicular | corner angle vs 90° | two-sided gaussian | **OK** — a right angle is a genuine optimum |
+| proportion | aspect vs target | clipped, one free side | **OK** |
+| width | narrowest vs minimum | clipped, one free side | **OK** |
+| crinkliness | daylight per unit floor | one-sided since §39.14 | **OK** |
+| access | has a circulation neighbour | binary | **OK** — not a curve |
+| size, *lower* side | area vs programme target | gaussian | **OK** — a room has a right size |
+| size, *upper* side | area over target | gaussian | **SUSPECT** — `cost` already charges floor area, and an oversize room is wasteful in proportion, not quadratically. §39.15 found 82% of size fails are over-target. Not changed: §39.15 also found this upper half is the main brake on growth, so removing it needs the value rates looked at first. |
+| ratio_outside | outdoor fraction vs target | two-sided gaussian | **OK** — and its binding side counters the economics rather than duplicating it, since outdoor space is profitable |
+| ratio_circulation | circulation fraction | gaussian, effectively one-sided | **REMOVED** |
+| min internal area | total room area below 1.2× the programme | one-sided gaussian | **SUSPECT** — "build the rooms you were asked for" is already said by the missing-space fails and again by `quality_size`. A third statement of it. |
+| staircase volume | stair fit | gaussian | **OK** — a stair has a right size |
+| stair count, storey limits | counts vs bounds | binary fails | **OK** |
+| `ratio_public_outside`, `ratio_private_outside` | boundary-length ratios | gaussian | **DEAD** — absent from `CONF_DEFAULTS` and from every corpus config, so the branch never runs |
+| `daylight` factor | pinned to 1.0 | constant | **DEAD** — `URB_NO_OCCLUSION` (§6); it is a factor that can never be anything but 1 |
+| `0.5 ** n_fails` | fail count | exponential | **RULING WANTED** — each failure independently halves the building. Coherent as a product of independent defects, but it is not "twice as many failures is twice as bad", and it is the single largest term in the objective. |
+
+The three SUSPECT entries and the fail multiplier are filed rather than changed
+— each needs a ruling or a rate change behind it, and §39.16 is the standing
+reminder about acting on an inherited constant before understanding what it
+measures. The two DEAD entries are inert and cost nothing but confusion.

examples/harbor-house/patterns.config

@@ -267,9 +267,14 @@ storey_limit: 3
 force_roof_garden: 0
 
 # Circulation ratio (allow more circulation for larger building)
+# homemaker-py-hxi (DESIGN.md §39.24): disabled. The score is value/cost, a
+# ratio, so the per-m2 economics (circulation worth 50 against a build cost of
+# 200) is ALREADY a proportional pressure against circulation -- this said the
+# same thing a second time, and with a curve where twice the corridor is far
+# more than twice as bad. Its other side, "a building needs some circulation",
+# was worth at most 13% here and is enforced structurally anyway by the access
+# and connectivity checks. Set a [target, sigma] pair to re-enable.
 ratio_circulation:
-  - 0.08
-  - 0.15
 
 # Outside space ratio (courtyard requirement)
 ratio_outside:

examples/health-centre/patterns.config

@@ -351,9 +351,14 @@ storey_minimum: 1
 storey_limit: 2
 force_roof_garden: 0
 
+# homemaker-py-hxi (DESIGN.md §39.24): disabled. The score is value/cost, a
+# ratio, so the per-m2 economics (circulation worth 50 against a build cost of
+# 200) is ALREADY a proportional pressure against circulation -- this said the
+# same thing a second time, and with a curve where twice the corridor is far
+# more than twice as bad. Its other side, "a building needs some circulation",
+# was worth at most 13% here and is enforced structurally anyway by the access
+# and connectivity checks. Set a [target, sigma] pair to re-enable.
 ratio_circulation:
-  - 0.10
-  - 0.12
 
 ratio_outside:
   - 0.06

examples/maple-court/patterns.config

@@ -406,9 +406,14 @@ storey_limit: 4
 force_roof_garden: 0
 
 # Circulation ratio (allow more circulation for larger building)
+# homemaker-py-hxi (DESIGN.md §39.24): disabled. The score is value/cost, a
+# ratio, so the per-m2 economics (circulation worth 50 against a build cost of
+# 200) is ALREADY a proportional pressure against circulation -- this said the
+# same thing a second time, and with a curve where twice the corridor is far
+# more than twice as bad. Its other side, "a building needs some circulation",
+# was worth at most 13% here and is enforced structurally anyway by the access
+# and connectivity checks. Set a [target, sigma] pair to re-enable.
 ratio_circulation:
-  - 0.08
-  - 0.15
 
 # Outside space ratio (courtyard requirement)
 ratio_outside:

examples/programme-house/patterns.config

@@ -82,9 +82,14 @@ storey_limit: 5
 force_roof_garden: 0
 
 # Circulation ratio (penalise excessive corridors)
+# homemaker-py-hxi (DESIGN.md §39.24): disabled. The score is value/cost, a
+# ratio, so the per-m2 economics (circulation worth 50 against a build cost of
+# 200) is ALREADY a proportional pressure against circulation -- this said the
+# same thing a second time, and with a curve where twice the corridor is far
+# more than twice as bad. Its other side, "a building needs some circulation",
+# was worth at most 13% here and is enforced structurally anyway by the access
+# and connectivity checks. Set a [target, sigma] pair to re-enable.
 ratio_circulation:
-  - 0.00
-  - 0.15
 
 # Outside space ratio
 ratio_outside:

src/homemaker_layout/fitness.py

@@ -210,7 +210,25 @@ CONF_DEFAULTS: dict = {
     "door_width": 1.2,
     "plot_ratio": [2.00, 0.50],
     "ratio_outside": [0.33, 0.15],
-    "ratio_circulation": [0.00, 0.20],
+    # homemaker-py-hxi (DESIGN.md §39.24). Was [0.00, 0.20] -- a gaussian on
+    # the circulation FRACTION, targeting zero, applied as a multiplier to the
+    # whole building's value (0.013..0.70 across the corpus). `None` disables
+    # it, because it double-counts and does so with the wrong curve.
+    #
+    # The score is `value / cost`, which is already a ratio, so the per-m2
+    # economics -- circulation worth 50 against a build cost of 200 -- is
+    # already a PROPORTIONAL pressure and not merely an absolute one: adding
+    # corridor moves value/cost by an amount that depends on how much of the
+    # building is already corridor. A second, super-linear term on the same
+    # quantity says the same thing twice, and says it with a curve where twice
+    # the corridor is far more than twice as bad (0.10 -> x0.882, 0.20 ->
+    # x0.607, 0.40 -> x0.135).
+    #
+    # Contrast `ratio_outside`, which is KEPT: its target is 0.33, not zero, so
+    # it is a genuine two-sided requirement rather than "less is better", and
+    # its binding side (do not build too much outdoor space) is not something
+    # the economics expresses -- outdoor space is profitable.
+    "ratio_circulation": None,
     # PROVENANCE (DESIGN.md §39.16). This is Christopher Alexander, A Pattern
     # Language 159, "Light on Two Sides of Every Room" -- not an arbitrary
     # constant. The factor is evaluated at `1/crink = A/(L*h)`: floor area per
@@ -2048,6 +2066,7 @@ class Fitness:
         factor *= self.ratio_o(ratios)
 
         circ_ratio = self.conf("ratio_circulation")
+        if circ_ratio is not None:          # §39.24: off by default
             factor *= self.ratio_type(ratios, "c", circ_ratio[0], circ_ratio[1])
 
         min_required = 0.0

tests/test_circulation_proportion.py

@@ -150,13 +150,22 @@ def test_twice_the_corridor_is_exactly_twice_as_bad():
         "so, §39.23's justification needs revisiting")
 
 
-def test_the_amount_of_circulation_is_still_priced():
-    """Removing the per-leaf cap must not make corridors free. Two charges
-    remain, and they are the ones that do not depend on how it is cut up."""
+def test_the_amount_of_circulation_is_priced_exactly_once():
+    """Removing the per-leaf cap must not make corridors free -- but the charge
+    that remains must also be the ONLY one (§39.24).
+
+    The linear ramp is `value_circulation` against the build cost: every square
+    metre of corridor is worth less than it costs, and because the score is
+    `value / cost` that pressure is already proportional to how much of the
+    building is corridor. `ratio_circulation` said the same thing a second time
+    and is disabled; if it comes back, either it has a justification this test
+    does not know about or the double count has returned.
+    """
     fit = _fit()
     assert fit.conf("value_circulation") < fit.cost("inside"), (
         "a corridor must cost more to build than it is worth, or there is no "
         "linear ramp pushing the search to use less of it")
-    assert fit.conf("ratio_circulation") is not None, (
-        "the building-level circulation fraction check is the other charge"
-    )
+    assert fit.conf("ratio_circulation") is None, (
+        "ratio_circulation duplicates the per-m2 economics (§39.24)")
+    assert fit.conf("size_circulation") is None, "§39.23"
+    assert fit.conf("proportion_circulation") is None, "§39.22"