Skip to content

Sources

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

DimensionValueClassCitation
Radius of the guard disc. A plank is cut to its own size and the post through it is whatever it is — so the overhang that stops a rat is the difference between the two rather than a figure of its own. Written as an overhang first, which made the disc grow with its post and the check on it a restatement that could not fail. The sources agree the disc is there and that it stops rats; none gives a size. So a defence whose whole efficacy is a difference, with both figures invented.guardRadius0.36 mIf 20% out: 0.07 mInterpolatedNo source
Thickness of the guard disc.guardThickness0.07 mMoves no measurementInterpolatedNo source
How far the disc sits below the granary floor. It has to be lower than the hood’s eave: a guard hidden inside the roof cannot be inspected by anyone, and more to the point, a hood falling past it becomes the way round that it exists to close. Written as 0.34 m first, which put it inside the hood.guardDrop0.55 mMoves no measurementInterpolatedNo source
Facets on the guard disc. Drawing resolution rather than a dimension of the building.guardFacets20Moves no measurementInterpolatedNo source
Section of a post. Slim — and that is part of the defence: a stout post shrinks the overhang of the disc above it until it is no longer any use.postSection0.16 mIf 20% out: 0.02 mInterpolatedNo source
Spacing between posts.postSpacing1.85 mIf 20% out: 0.37 mInterpolatedNo source
Height of the stone a post foot stands on.stoneHeight0.24 mIf 20% out: 0.05 mInterpolatedNo source
Width of that stone.stoneWidth0.40 mMoves no measurementInterpolatedNo source
Height of the granary floor above the ground. High enough for a person to sit and work beneath it, which is what happens.floorHeight2.00 mIf 20% out: 0.40 mInterpolatedNo source
Thickness of a board floor.floorThickness0.07 mMoves no measurementInterpolatedNo source
Height of the store. Nobody can stand up in it, and nobody needs to: what is kept there is rice, and rice goes in from above.storeHeight1.25 mIf 20% out: 0.25 mInterpolatedNo source
How far the store’s walls sit inside the line of the posts.storeInset0.10 mMoves no measurementInterpolatedNo source
Thickness of a board wall.wallThickness0.05 mMoves no measurementInterpolatedNo source
Width of the opening at the end of the store, where rice goes in and comes out.hatchWidth0.60 mMoves no measurementInterpolatedNo source
Rise of the ridge above the store floor.ridgeRise2.50 mIf 20% out: 0.50 mInterpolatedNo source
How far the eave falls below the store floor. An eave that drops past its own floor is what makes this form read as a hood rather than as a roof — but it has to stop above the rat guards, because a hood falling past them becomes a way round.eaveDrop0.28 mIf 20% out: 0.06 mInterpolatedNo source
Reach of the eave from the axis, measured horizontally.eaveReach1.55 mMoves no measurementInterpolatedNo source
How far the hood’s side bellies outward from a straight line between the ridge and the eave. Zero gives an ordinary hip; this figure is what makes the curve.hoodBelly0.70Moves no measurementInterpolatedNo source
How many levels are used to approximate that curve. Drawing resolution rather than a dimension of the building — but it has to be large enough that the curve reads as a curve rather than as a stair, and the check requires each step to be steeper than the one below it.hoodSteps9Moves no measurementInterpolatedNo source
Half-length of the ridge at the top of the hood. Not zero: the top is a short line rather than a point.ridgeHalf0.22 mMoves no measurementInterpolatedNo source
Section of a rafter.rafterSection0.07 mMoves no measurementInterpolatedNo source
Rafters on each face.raftersPerSide5Moves no measurementInterpolatedNo source
Exposed depth of one course of thatch.thatchCourseDepth0.20 mMoves no measurementInterpolatedNo source
How far a course stands proud of the one below.thatchThickness0.08 mMoves no measurementInterpolatedNo source
The share of a course the course above laps.thatchLap0.45Moves no measurementInterpolatedNo source
Clearance between the frame and the first course.thatchBed0.04 mMoves no measurementInterpolatedNo source
Height of the sitting platform beneath the granary, where there is one.seatHeight0.42 mMoves no measurementInterpolatedNo source
Thickness of that platform’s boards.seatThickness0.06 mMoves no measurementInterpolatedNo source
Depth of the tested joint engagement, as a share of the smaller member.jointEngagement0.30Moves no measurementInterpolatedNo source
Depth of the dish in the stone the post foot seats into.postSeat0.30Moves no measurementInterpolatedNo source
Size of a village lumbung relative to a household one. That the village’s is larger is sourced; how much larger is the author’s.desaScale1.24Moves no measurementInterpolatedNo source
This building is for rice, not for people. There is no storey in it a person could stand up in, and in a Sasak yard it is routinely the most carefully made thing standing. The other eleven in this project scale their care to the standing of the people inside; this one scales it to the value of what is stored, and the people sit underneath.builtForRice1Moves no measurementCanonSumarni, N. K., “Lumbung Padi Sasak: Bentuk, Fungsi dan Makna”, Jurnal Arsitektur Zonasi (2018).
Every post carries one disc just below the floor, and it has exactly one job: a rat climbing the post cannot get past it. It is the only element in this project aimed at something other than a person — and it only works if there is no other way up, so what the check tests is the absence of one.guardOnEveryPost1Moves no measurementCanonDepartemen Pendidikan dan Kebudayaan, Arsitektur Tradisional Daerah Nusa Tenggara Barat (Proyek Inventarisasi dan Dokumentasi Kebudayaan Daerah, Jakarta).
Zero other paths from the ground to the store floor. No ladder left leaning, no raking brace touching both, no wall reaching down past the disc — and the hood’s own edge stops above it too. A rat guard with a bamboo pole beside it is not a guard, and neither is one covered by its own roof.noOtherWayUp0Moves no measurementCanonDepartemen Pendidikan dan Kebudayaan, Arsitektur Tradisional Daerah Nusa Tenggara Barat (Proyek Inventarisasi dan Dokumentasi Kebudayaan Daerah, Jakarta).
The eave falls past the floor of the store. This form reads as a hood rather than a roof precisely because its edge sits below the plane it protects.hoodFallsPastTheFloor1Moves no measurementCanonDepartemen Pendidikan dan Kebudayaan, Arsitektur Tradisional Daerah Nusa Tenggara Barat (Proyek Inventarisasi dan Dokumentasi Kebudayaan Daerah, Jakarta).
The post feet stand on stones and are not buried — which closes a path to termites as well as to rats.seatedOnStone1Moves no measurementCanonWaterson, R., The Living House: An Anthropology of Architecture in South-East Asia (Oxford University Press, Singapore, 1990).
Lumbung stand in a row in the compound alongside other granaries and the dwellings. The most carefully made building in the yard is not the one people sleep in, and that only reads when the others are in view.granaryRow1Moves no measurementCanonDepartemen Pendidikan dan Kebudayaan, Arsitektur Tradisional Daerah Nusa Tenggara Barat (Proyek Inventarisasi dan Dokumentasi Kebudayaan Daerah, Jakarta).
Spacing between lumbung along their row. The only figure left for the neighbours: everything else is this granary itself, read off the layout being drawn, because a neighbouring lumbung is a lumbung.neighbourSpacing6.00 mMoves no measurementInterpolatedNo source

Bibliography

  • depdikbud-ntb · reference

    Departemen Pendidikan dan Kebudayaan, Arsitektur Tradisional Daerah Nusa Tenggara Barat (Proyek Inventarisasi dan Dokumentasi Kebudayaan Daerah, Jakarta).

  • sumarni-2018 · reference

    Sumarni, N. K., “Lumbung Padi Sasak: Bentuk, Fungsi dan Makna”, Jurnal Arsitektur Zonasi (2018).

  • waterson-1990 · ethnography

    Waterson, R., The Living House: An Anthropology of Architecture in South-East Asia (Oxford University Press, Singapore, 1990).

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. 1ridgeRise0.50 mridge height
  2. 2floorHeight0.40 mheight of the store floor
  3. 3postSpacing0.37 mpost spacing across
  4. 4storeHeight0.25 mheight of the store
  5. 5guardRadius0.07 moverhang of the guard
  6. 6eaveDrop0.06 meave height
  7. 7stoneHeight0.05 mridge height
  8. 8postSection0.02 moverhang of the guard

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.

  • pass
    Bilateral symmetry about Symmetric about the ridge plane, x = 0

    1720 points; largest mirror deviation 0.00 mm.

  • pass
    Every tenon contained in its mortise

    10 pasak, takik and tumpu joints; every one engaged.

  • pass
    No joint spans more than one stage

    Every joint engages two parts from the same stage or from consecutive ones.

  • pass
    Stage ordering, and nothing below ground

    152 parts, every one bearing on something already standing or on the ground.

  • pass
    Mesh integrity: indices, triangles, normals

    568 triangles; indices in range, none degenerate, every normal unit length.

  • pass
    Every post carries a disc, and every disc overhangs its post

    4 discs of 360 mm radius on posts of 160 mm, leaving 280 mm of overhang. The entire efficacy of this defence is that difference — and both figures are the author’s, because no source gives either.

  • pass
    Nothing else reaches from the ground to the store floor

    Zero parts cross the guard height at 1.69 m except the posts themselves; the sitting platform stops short of every post; and the hood’s edge stops 0.27 m above them, so the discs hang clear and can be seen. A rat guard with a bamboo pole beside it is not a guard — so half of this defence is about everything that is not the disc.

  • pass
    There is no storey in it a person could stand up in

    A store 1.25 m high. This building is for rice, and in a Sasak yard it is often the most carefully made thing standing — its care follows the value of what is stored rather than the standing of whoever stores it. The other eleven in this project scale it the second way.

  • pass
    Four posts or six, and nothing between

    4 posts, a lumbung keluarga. A lumbung on five posts is not a smaller one; it is a building this tradition does not make.

  • pass
    The eave falls past the floor of the store

    An eave at 1.96 m against a store floor at 2.24 m — 0.28 m below it. That is what makes this form read as a hood pulled over a box rather than a roof set on top of one.

  • pass
    The hood curves outward: every band is shallower than the one below it

    9 bands, from 79° at the eave to 56° at the top. Made from the same primitive as five earlier roofs with nothing but more levels — a curve is many steps, and what separates it from a faceted cone is that the pitch keeps changing, which a check that only counted levels would not know.

  • pass
    Thatch courses lap with no bare band, the whole way round the curve

    16 courses over 3.11 m.

  • pass
    Every part declares the dimensions it came from

    152 parts, every one tagged. By least-sourced input: 0 measured, 0 canon, 152 interpolated.

  • skipped
    Checked 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 disc does not change. Its radius is the size of a plank, so a stouter post eats that overhang from the inside until there is none — and the rat guard becomes a collar flush with the timber it was meant to project beyond. Not one part of this granary looks weaker; the posts look sturdier. Twelve buildings, twelve rules that cannot be carried out — and what is particular here is the direction: the thing that defeats this defence is the first improvement a builder would reach for.

The house as built

pass160 mm / 280 mmpost section · overhang left

4 discs of 360 mm radius on posts of 160 mm, leaving 280 mm of overhang. The entire efficacy of this defence is that difference — and both figures are the author’s, because no source gives either.

Rear prow raised to 0.72 m

fail720 mm / 0 mmpost section · overhang left

the guard on tiang-0-0 does not overhang its post; the guard on tiang-0-1 does not overhang its post; the guard on tiang-1-0 does not overhang its post; the guard on tiang-1-1 does not overhang its post

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.