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 |
|---|---|---|---|
| 5.50 mIf 20% out: 1.23 m | Interpolated | Radius where the thatch meets the ground. Because the roof comes all the way down, this figure is also the plan of the house: there is no wall outside the roof. | |
| 15.20 mIf 20% out: 3.40 m | Interpolated | Height of the cone’s apex above the ground. The form is far taller than it is wide, and that proportion is what makes it read from a distance. | |
| 0.10If 20% out: 0.05 m | Interpolated | How far the cone’s side bellies outward from a straight line between the ground and the apex. The previous figure, 0.055, was barely visible and the form read as a geometric cone; thatch laid course on course bows the side out noticeably. The author’s reading of photographs, not anyone’s measurement. | |
| 0.94If 20% out: 0.41 m | Interpolated | How blunt the cone’s shoulder is near the top. A value of one gives a straight cone tapering to a mathematical point; less than one widens the upper part. Kept close to one: lower values fatten the whole upper half and the form reads as a bell rather than a cone. | |
| 0.30If 20% out: 0.09 m | Interpolated | Where the swell is greatest, as a share of the cone’s height. Not the middle: the courses low down are longer and there are more of them, so the material builds thickest in the lower third and the upper part stays nearly straight. Putting it at mid-height makes the building’s waist its fattest part and the silhouette reads as a bell. | |
| 4.00If 20% out: 0.04 m | Interpolated | How concentrated the swell is about its peak. Larger values pull it into a narrow bulge; smaller ones spread it along the cone. It has to be large enough that the outline leaves the ground with its radius already shrinking — otherwise the roof widens before it narrows, and the lowest course is pushed below grade. | |
| 24Moves no measurement | Interpolated | Points along the cone’s outline. This is drawing resolution rather than a dimension of the building. | |
| 48Moves no measurement | Interpolated | Target facets around the cone. Drawing resolution rather than a dimension of the building — but the actual count is rounded to a multiple of the household count, because a mesh that does not respect the building’s symmetry makes the model claim a symmetry it does not have. | |
| 1.25 mIf 20% out: 0.25 m | Interpolated | Height of the living floor above the ground. The house is raised, but the thatch closes the space beneath it, so from outside it does not read as a house on stilts. | |
| 2.45 mIf 20% out: 1.96 m | Interpolated | Distance between floors. The same for all four above the lutur, so each floor’s radius is set by the cone rather than declared on its own. | |
| 0.07 mMoves no measurement | Interpolated | Thickness of a board floor. | |
| 0.22 mMoves no measurement | Interpolated | Width of one floor board. | |
| 0.30 mMoves no measurement | Interpolated | Section of the centre post, which stands from the ground to the apex and is the axis the whole house is set out from. | |
| 0.19 mMoves no measurement | Interpolated | Section of a ring post. | |
| 0.62If 20% out: 0.76 m | Interpolated | Where the ring of posts stands, as a share of the base radius. | |
| 0.24 mMoves no measurement | Interpolated | Height of the stone a post stands on. | |
| 0.40 mMoves no measurement | Interpolated | Width of the stone a post stands on. | |
| 0.13 mMoves no measurement | Interpolated | Depth of a tie ring, the hoop that holds the frame round. | |
| 0.09 mMoves no measurement | Interpolated | Width of a tie ring. | |
| 4Moves no measurement | Interpolated | Rafters over each household’s segment. The total is this number times the household count rather than a figure of its own — so that every household stands under the same amount of roof, and so that the building actually repeats every one segment. | |
| 0.05 mMoves no measurement | Interpolated | Radius of a rafter. | |
| 0.06 mMoves no measurement | Interpolated | Thickness of a partition between households. | |
| 1.90 mMoves no measurement | Interpolated | Height of a partition on the lutur floor. It does not reach the floor above: the division marks each household’s place rather than shutting it in. | |
| 0.55 mMoves no measurement | Interpolated | Radius of the hearth at the centre of the lutur floor. | |
| 0.30 mMoves no measurement | Interpolated | Exposed depth of one course of thatch. | |
| 0.10 mMoves no measurement | Interpolated | How far a course stands proud of the one below it. | |
| 0.40Moves no measurement | Interpolated | The share of a course that the course above laps over. | |
| 0.05 mMoves no measurement | Interpolated | Clearance between the frame and the first course of thatch. | |
| 1.10 mMoves no measurement | Interpolated | Height of the apex ornament above the point of the cone. The object is real and has a name; the name is not printed here because the author is not confident enough of the word. | |
| 0.95 mMoves no measurement | Interpolated | Width of the door. There is only one, and it is the only thing on this round building that fixes a direction. | |
| 1.75 mMoves no measurement | Interpolated | Height of the door, measured from the ground. | |
| 0.13 mMoves no measurement | Interpolated | Section of a door jamb. | |
| 0.55Moves no measurement | Interpolated | Waist of the apex ornament relative to its foot. | |
| 0.45Moves no measurement | Interpolated | Height of that waist, as a share of the ornament’s height. | |
| 0.12Moves no measurement | Interpolated | Height of the ring round the foot of the apex ornament. | |
| 0.72Moves no measurement | Interpolated | Radius of the centre post’s stone relative to the width of a ring post’s. Round rather than square, because a stone on the axis has to look the same after the house is turned by one segment. | |
| 1.40Moves no measurement | Interpolated | Depth of the hearth relative to the thickness of a floor. | |
| 0.80Moves no measurement | Interpolated | How high the drum hangs, as a share of the partition height. | |
| 0.34 mMoves no measurement | Interpolated | Width of the apex ornament. | |
| 1.12Moves no measurement | Interpolated | Size of the niang gendang relative to an ordinary house. That the drum house is the principal one in the village is canon; how much larger it is, is the author’s. | |
| 0.42 mMoves no measurement | Interpolated | Radius of the drum hung in the drum house. | |
| 1.15 mMoves no measurement | Interpolated | Length of the drum. | |
| 0.30Moves no measurement | Interpolated | Depth of the dish in the stone that the post foot seats into, as a share of stone height. | |
| 0.30Moves no measurement | Interpolated | Depth of the tested joint engagement, as a share of the smaller member. | |
| 1Moves no measurement | Canon | One door, facing the compang — the ceremonial stone at the centre of the village. A round form holds no direction in itself, so the only orientation this house has comes from where it stands, and the door is what states it. | |
| 1.00Moves no measurement | Canon | The plan is round and the roof is a cone. There is no ridge, no face and no corner — and so no orientation that can be read off the form itself. | |
| 1.00Moves no measurement | Canon | The thatch comes down to the ground. From outside the whole building is roof: there is no wall and no eave standing above the ground. | |
| 5Moves no measurement | Canon | Five stacked floors, each named and each for something — from the living floor up to the loft where offerings are kept. It is not a choice, which is why this pack has only two rules. | |
| 1Moves no measurement | Canon | The lutur floor is divided into segments, one for each household living in it. The number of households is read off that division. | |
| 1Moves no measurement | Canon | A village has one niang gendang. It holds the drum and the ceremonial life of the village; the rest are dwellings. | |
| 1Moves no measurement | Canon | A single centre post stands from the ground to the apex and passes through all five floors. The whole house is set out from it. | |
| 1.00Moves no measurement | Canon | Pegged and lashed joints; the frame goes up without nails. | |
| 1Moves no measurement | Canon | The posts stand on stones; they are not buried. | |
| 1Moves no measurement | Canon | Mbaru niang stand around a circular plaza with the compang — a ceremonial stone platform — at its centre, and every house faces that centre. This house has no front until there is a circle to give it one. | |
| 14.00 mMoves no measurement | Interpolated | Radius of the village plaza, measured from the compang. | |
| 0.70 mMoves no measurement | Interpolated | Height of the stone platform at the centre of the village. A compang is a stone dais where offerings are made; that it exists and is central is canon, its height and shape are the author’s. | |
| 0.40 mMoves no measurement | Interpolated | Width of one upright stone on the compang. | |
| 1.10 mMoves no measurement | Interpolated | Height of an upright stone standing on it. | |
| 2.20 mMoves no measurement | Interpolated | Radius of the stone platform at the centre. That it exists and is central is canon; its size is the author’s. |
Bibliography
antar-2010 · reference
Antar, Y., Pesan dari Wae Rebo: Kelahiran Kembali Arsitektur Nusantara (Gramedia Pustaka Utama, Jakarta, 2010).
erb-1999 · ethnography
Erb, M., The Manggaraians: A Guide to Traditional Lifestyles (Times Editions, Singapore, 1999).
depdikbud-ntt · reference
Departemen Pendidikan dan Kebudayaan, Arsitektur Tradisional Daerah Nusa Tenggara Timur (Proyek Inventarisasi dan Dokumentasi Kebudayaan Daerah, Jakarta).
unesco-2012 · reference
UNESCO Asia-Pacific Awards for Cultural Heritage Conservation 2012, Award of Excellence: Wae Rebo Village, Flores.
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.
- 3.40 m
- 1.96 m
- 1.23 m
- 0.76 m
- 0.41 m
- 0.25 m
- 0.09 m
- 0.05 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 a vertical plane through the axis
21374 points; largest mirror deviation 0.00 mm.
- passThe house repeats every one segment, which is every 60°
19903 points (145 of 161 parts); turned 60° the largest deviation is 0.00 mm.
- passEvery tenon contained in its mortise
31 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
161 parts, every one bearing on something already standing or on the ground.
- passMesh integrity: indices, triangles, normals
28200 triangles; indices in range, none degenerate, every normal unit length.
- passThe plan is round: as wide one way as the other
12.42 m × 12.42 m, differing by 0 mm — as round as a 48-sided polygon can be (27 mm). No face, no corner and no ridge.
- passThe thatch reaches the ground; the whole building is roof
The lowest course meets the ground at 0.03 m. No wall, no eave, and nothing holding it up off the ground.
- passNothing stands through the cone, at any height
3577 points tested against the cone at each one's own height; the closest comes within 0.00 m of it.
- passFive named floors, each narrower than the last because the cone narrows
lutur → lobo → lentar → lempa rae → hekang kode; from 5.98 m at 1.25 m to 2.53 m at 11.05 m.
- passOne centre post runs through all five floors to the apex
from 0.17 m to 17.02 m, passing through 5 of 5 floors. No single member in any other house here touches every storey.
- passOne segment per household, and the whole house divides evenly by it
6 households → 6 partitions, 6 ring posts, 24 rafters and 48 mesh facets — every one a multiple of 6.
- passNiang gendang: the drum hangs inside
The drum is present, hung 2.77 m above the lutur floor.
- passOne door that is actually open, and it is the only direction this building has
0.95 m wide on a bearing of 0°, facing the compang; the thatch is cut back 1.20 m to a height of 1.88 m. A round form holds no direction; this door is what states it, and there is one.
- passThatch courses lap with no bare ring, from the ground to the apex
101 courses over 18.18 m of outline; smallest lap 0.8%.
- passEvery part declares the dimensions it came from
161 parts, every one tagged. By least-sourced input: 0 measured, 0 canon, 161 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.
Five floors is canon, and the cone is where all five of them stand — so the two are not independent. Space the storeys further apart and the topmost climbs into a part of the building too narrow to be a floor, or past the point of the cone altogether. Nothing is zeroed and nothing collapses: the building runs out before the floors do. A rule that cannot be carried out rather than a rule disobeyed — and after four houses, that appears to be how these buildings actually fail.
The house as built
Rear prow raised to 4.17 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.