Every dimension the house is generated from is below, with its class and its citation. This is the project’s honesty layer, and it is given a room of its own because a smooth 3D render implies a precision the sources do not have.
The dimensions
| Dimension | Value | Class | What it means |
|---|---|---|---|
| 0.25If 20% out: 0.93 m | Interpolated | Length of a hasta — elbow to fingertip — relative to a depa. A quarter of the arm span is ordinary anthropometry rather than a figure from the Balinese literature. What is sourced is that a hasta is a measure of the owner’s body; what the ratio is, is the author’s. | |
| 0.07If 20% out: 0.10 m | Interpolated | Length of a musti — the closed fist with the thumb laid across — relative to a depa. The author’s anthropometry, as with the hasta. | |
| 0.01If 20% out: 0.02 m | Interpolated | Length of a useran — one rotation of the thumb — relative to a depa. It is also the size of the pengurip, so this figure is what sets how far the house departs from its own whole multiples. | |
| 0.01If 20% out: 0.03 m | Interpolated | Width of one finger relative to a depa. | |
| 1If 20% out: 1.02 m | Interpolated | Spacing between saka, in depa. One arm span per bay: enough for a person to lie across, which is what a bale is for. That the spacing is measured in depa is canon; that it is one, is the author’s. | |
| 5If 20% out: 0.42 m | Interpolated | Height of a saka, in hasta, from the sendi to the underside of the tie. Five hasta gives clear headroom without making the pavilion read as a house. | |
| 4If 20% out: 0.10 m | Interpolated | Height of the bataran, in musti. The platform is low — something to sit on rather than a storey — and that is what separates it from the three raised houses in this project. | |
| 2If 20% out: 0.10 m | Interpolated | How far the bataran stands outside the saka, in musti. There is a margin for a foot outside the posts, and the roof drops its water outside that margin. | |
| 2If 20% out: 0.17 m | Interpolated | How far the eave reaches past the bataran, in hasta. This is what makes a bale usable in rain: a building with no walls is dry only as far as its roof reaches. | |
| 6If 20% out: 0.51 m | Interpolated | Rise of the ridge above the eave, in hasta. It sets the pitch, and alang-alang needs a steep one so water leaves quickly. Written as three hasta first, which gives a pitch of 24° — this dimension’s own note already said “steep” while its value did not, and the silhouette read as an umbrella rather than a roof. Six hasta gives about 42°. | |
| 2Moves no measurement | Interpolated | Height of the sitting platform inside, in musti, above the bataran paving. | |
| 4Moves no measurement | Interpolated | Depth of the sitting platform, in hasta — one depa, enough for a person to lie across. Two hasta at first, 0.85 m, and in the render it read as a shelf rather than something to sit and sleep on. The bale-bale is what this building is for; making it too narrow to use is building a roof over a ledge. | |
| 2Moves no measurement | Interpolated | Rise of one step up to the bataran, in musti. | |
| 8If 20% out: 0.03 m | Interpolated | Section of a saka, in fingers. This is a size of timber rather than a set-out dimension, so it takes no pengurip — and the module check states how many lengths it excludes for that reason. Given in nyari rather than musti because one musti, 128 mm, makes a 2.1 m post read as a stick in the render; eight fingers gives 150 mm. Stock is measured in fingers — it is the unit a person reaches for when saying how thick a thing is, and its whole multiples are closer together. | |
| 3Moves no measurement | Interpolated | Depth of a tie beam, in nyari. | |
| 2Moves no measurement | Interpolated | Section of a rafter, in nyari. | |
| 1Moves no measurement | Interpolated | Thickness of a board, in nyari. | |
| 2If 20% out: 0.01 m | Interpolated | Height of a sendi, the stone a saka foot seats on, in nyari. | |
| 2Moves no measurement | Interpolated | Width of a sendi relative to the saka section, in musti. | |
| 4Moves no measurement | Interpolated | Width of a step, in musti. | |
| 3Moves no measurement | Interpolated | Going of one step, in musti. | |
| 3Moves no measurement | Interpolated | Width of the ridge finish, in nyari. | |
| 1Moves no measurement | Interpolated | Section of a deck leg, in musti. | |
| 5Moves no measurement | Interpolated | Rafters in each bay. Batten spacing rather than a set-out dimension. | |
| 0.22 mMoves no measurement | Interpolated | Exposed depth of one course of alang-alang. Closer than ijuk because the material is grass rather than fibre. | |
| 0.08 mMoves no measurement | Interpolated | How far a course stands proud of the one below it. | |
| 0.45Moves no measurement | Interpolated | The share of a course that the course above laps over. | |
| 0.04 mMoves no measurement | Interpolated | Clearance between the frame and the first course. | |
| 0.18 mMoves no measurement | Interpolated | Height of the ridge finish above the roof line. | |
| 0.09 mMoves no measurement | Interpolated | Thickness of the paras facing over the brick body of the bataran. | |
| 0.30Moves no measurement | Interpolated | Depth of the tested joint engagement, as a share of the smaller member. | |
| 0.30Moves no measurement | Interpolated | Depth of the dish in the sendi that the saka foot seats into, as a share of its height. | |
| 1Moves no measurement | Canon | The house is measured with its owner’s body: depa, hasta, musti, useran, finger. The unit itself is the social fact — two households of identical standing build different buildings because their bodies differ. The other four houses here select a number; this one selects the unit. | |
| 1Moves no measurement | Canon | A principal measure is a whole number of body units plus a pengurip, because a measure landing exactly on its module is called mati — dead. This house is required not to be exactly its own rule, and that is the only rule of its kind in this project. | |
| 3Moves no measurement | Canon | Three vertical parts: the bataran as foot, the saka as body, the roof as head. The division applies to a building, a compound and a village alike, and it is on the building that it reads most directly. | |
| 1Moves no measurement | Canon | The name of a bale is the number of its saka: sakepat four, sakenem six, sangasari nine, sakaroras twelve. The only rule in this project where the word and the number are the same fact said once. | |
| 0Moves no measurement | Canon | A bale is open on its sides: zero wall planes between the floor and the eave. What makes it usable is the reach of the roof rather than any enclosure — so the dimension that decides its comfort is the depth of the overhang. | |
| 1Moves no measurement | Canon | The building stands on a bataran, a low masonry platform. Not a house on stilts: there is no space beneath it, and the bataran is the foot in the threefold division. | |
| 1Moves no measurement | Canon | A saka foot seats on its sendi and is not buried — so the frame can be taken apart and raised again. | |
| 1Moves no measurement | Canon | Orientation is set by the kaja–kelod axis: kaja toward the mountain, kelod toward the sea. It is not a compass direction. In south Bali kaja is north; in north Bali kaja is south — so the same rule gives opposite bearings depending on which side of the mountain you stand. | |
| 4Moves no measurement | Canon | A hipped roof: four planes falling to an eave that closes all the way round, with a ridge shorter than the building. On a square bale that ridge length becomes zero and the roof becomes a true pyramid — the form follows from the number of saka rather than being declared. | |
| 1Moves no measurement | Canon | A bale stands inside a walled compound with several others around the natah, with the shrine in the kaja-kangin corner. The single building is a part; the compound is the house. | |
| 22.00 mMoves no measurement | Interpolated | Side of the walled compound. That there is a compound is canon; its size is the author’s. | |
| 1.80 mMoves no measurement | Interpolated | Height of the compound wall. A Balinese wall is brick and paras with a gate; this is its massing only, with no carving and no gate form. | |
| 0.35 mMoves no measurement | Interpolated | Thickness of that wall. | |
| 1.00 mMoves no measurement | Interpolated | How far the shrine stands off the wall in the kaja-kangin corner. | |
| 2.60 mMoves no measurement | Interpolated | Height of the shrine massing in the kaja-kangin corner. A real shrine is a tiered, carved structure; this is a stone block of that height and nothing more. | |
| 3.00 mMoves no measurement | Interpolated | Plan side of the shrine in the kaja-kangin corner. Only its footprint is drawn. |
Bibliography
gelebet-1986 · reference
Gelebet, I N., Arsitektur Tradisional Daerah Bali (Departemen Pendidikan dan Kebudayaan, Denpasar, 1986).
budihardjo-1986 · reference
Budihardjo, E., Architectural Conservation in Bali (Gadjah Mada University Press, Yogyakarta, 1986).
eiseman-1990 · ethnography
Eiseman, F. B., Bali: Sekala and Niskala, Volume II — Essays on Society, Tradition and Craft (Periplus, Berkeley, 1990).
dwijendra-2008 · reference
Dwijendra, N. K. A., Arsitektur Rumah Tradisional Bali Berdasarkan Asta Kosala Kosali (Udayana University Press, Denpasar, 2008).
anthropometry · none
Tidak ada sumber Bali. Perbandingan antara satu ukuran tubuh dan ukuran berikutnya adalah antropometri umum yang ditetapkan penulis — bukan angka dari kepustakaan Bali mana pun. Kunci ini dipisahkan dari “tidak ada sumber” supaya perbedaannya terbaca di tabel.
Which of these matters most
Nearly every dimension on this page is the author’s own. What the provenance bar does not say is whether that matters. Each row below is computed by rebuilding the house: one dimension is pushed 20%, and the house is measured to see how far it follows. Nothing is estimated — the figures are the difference between two runs of the generator.
- 1.02 m
- 0.93 m
- 0.51 m
- 0.42 m
- 0.17 m
- 0.10 m
- 0.10 m
- 0.10 m
If only one dimension could be surveyed, this is the one. This is sensitivity, not uncertainty: it shows what moves, not how likely the value is to be wrong. The 20% is itself the author’s choice.
Invariants
There is no measured drawing yet, so correctness rests on structural truth. These checks fail the build.
- passBilateral symmetry about The building mirrors about z = 0, the plane the ridge runs along
1112 points; largest mirror deviation 0.00 mm.
- passEvery tenon contained in its mortise
50 pasak, takik and tumpu joints; every one engaged.
- passNo joint spans more than one stage
Every joint engages two parts from the same stage or from consecutive ones.
- passStage ordering, and nothing below ground
120 parts, every one bearing on something already standing or on the ground.
- passMesh integrity: indices, triangles, normals
176 triangles; indices in range, none degenerate, every normal unit length.
- passThe number of saka is the name of this building
Bale gede (sakaroras): 12 saka, 3 × 4. The name and the count are one fact said once.
- passEvery principal length is a whole number of its owner’s body
9 principal lengths, in 3 body measures, from a depa of 1700 mm. Not one length in this building is written as a metre; every one is counted off its owner.
- passNo principal length lands exactly on its module: an exact measure is dead
A pengurip of one useran, 21.3 mm, added to 9 principal lengths. This building is required not to be exactly its own rule, and no other rule in this project has that shape.
- passThree vertical parts, in order: bataran, saka, roof
Foot 0 → 0.53 m, body → 2.71 m, head → 5.43 m. The head is the largest part, as it should be.
- passOpen on every side: not one wall plane between the floor and the eave
Zero enclosing planes between 0.53 m and 2.71 m. What makes this bale usable in rain is not a wall but 0.87 m of overhang.
- passFour roof planes at one pitch, and a ridge shorter than the building
A ridge of 1.72 m over an eave of 7.61 × 5.89 m: shorter by exactly half the eave depth at each end.
- passThe eave reaches past the bataran on all four sides
0.87 m across and 0.87 m along. On a building with no walls this figure is what decides which of it stays dry.
- passThe saka feet seat on their sendi and are not buried in the bataran
12 saka standing on their stones. Because none is buried, the frame can be taken apart and raised again.
- passEvery frame member stops where its roof plane runs out
No point of the frame reaches outside the roof surface at its own height. On a hip the four planes narrow as they rise, so a rafter near a corner is shorter — and a fault of this kind is completely hidden by the thatch.
- passThatch courses lap with no bare band, from the eave to the ridge
18 courses over 3.91 m of slope.
- passEvery part declares the dimensions it came from
120 parts, every one tagged. By least-sourced input: 0 measured, 0 canon, 120 interpolated.
- skippedChecked against a measured drawing of a real house
Skipped. No measured drawing has been wired in, so there is nothing to compare against. This check will not be made to pass by weakening it.
What a check is preventing
Ten green rows are not a reason to believe anything. So one check is run against a house built to break it, and what is printed below is that check’s own verdict rather than a description of it.
The pengurip is one useran, and its job is to keep every principal measure off its own module. Grow the useran and nothing bends, nothing collapses and nothing goes to zero — the addition simply climbs until it is a whole unit in itself, and the measure it was keeping inexact lands squarely back on its module: dead again, just one unit larger. Every other counterexample in this project ends with a building that cannot be constructed. This one ends with a building that stands perfectly well and is wrong for a reason no amount of looking would reveal — which is the strongest argument the project has for why the rules live in the model rather than in the caption.
The house as built
Rear prow raised to 0.08 m
The second house is never rendered. It is built during the build, failed, and thrown away.
This bar moves as real surveys are wired in. That movement is the project's progress metric.