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

DimensionValueClassWhat it means
bayLength2.40 mIf 20% out: 5.15 mInterpolatedSpacing between posts along the ridge. The plan grows by whole multiples of this and by nothing else.No source
bayDepth2.70 mIf 20% out: 1.98 mInterpolatedSpacing between posts across the building. A little larger than the spacing along it, so the bays beneath are not square and their diagonals have a definite direction.No source
bodyRows4If 20% out: 3.29 mInterpolatedRows of posts across the building. It sets the depth; the number along the building comes from the bay rule.No source
postSection0.34 mIf 20% out: 0.08 mInterpolatedSection of an ehomo. These posts are heavy — not for the load standing on them, but because they are also what the driwa deliver their force into.No source
braceSection0.26 mMoves no measurementInterpolatedSection of a driwa. Smaller than an ehomo but not much smaller: these members work rather than merely prop.No source
stoneHeight0.30 mIf 20% out: 0.07 mInterpolatedHeight of the stone a post foot stands on.No source
stoneWidth0.58 mMoves no measurementInterpolatedWidth of that stone.No source
floorHeight3.10 mIf 20% out: 4.92 mInterpolatedHeight of the floor above the ground. The understorey is tall and open, and it is there that the whole idea of the building is visible: the diagonal frame is not hidden.No source
floorThickness0.09 mMoves no measurementInterpolatedThickness of a board floor.No source
wallHeight2.30 mIf 20% out: 0.56 mInterpolatedHeight of the wall from the floor to the eave.No source
wallLean0.42 mMoves no measurementInterpolatedHow far the head of the wall stands outboard of its foot. It leans outward, so the body widens as it rises — the opposite of the tongkonan, whose walls are left vertical because no source gives an angle.No source
wallThickness0.07 mMoves no measurementInterpolatedThickness of a wall board.No source
windowHeight0.62 mMoves no measurementInterpolatedHeight of the window band across the front. One continuous opening rather than a row of separate holes.No source
windowDrop0.55 mMoves no measurementInterpolatedDistance from the eave down to the head of the window band.No source
windowInset0.90 mMoves no measurementInterpolatedHow far the window band stops short of the corner, at each end.No source
ridgeRise5.60 mIf 20% out: 1.37 mInterpolatedRise of the ridge above the eave. The roof is far larger than the body it covers, and the greatest part of this building is roof.No source
eaveOversail1.10 mMoves no measurementInterpolatedDepth of the overhang outside the wall.No source
rafterSection0.11 mMoves no measurementInterpolatedSection of a rafter.No source
raftersPerBay4Moves no measurementInterpolatedRafters in each bay.No source
thatchCourseDepth0.26 mMoves no measurementInterpolatedExposed depth of one course of sago thatch.No source
thatchThickness0.09 mMoves no measurementInterpolatedHow far a course stands proud of the one below it.No source
thatchLap0.42Moves no measurementInterpolatedThe share of a course the course above laps over.No source
thatchBed0.05 mMoves no measurementInterpolatedClearance between the frame and the first course.No source
loftHeight2.00 mIf 20% out: 0.49 mInterpolatedHeight of the loft above the floor, in a si’ulu’s house. The space inside that large roof is used; it is not a void.No source
behuHeight1.60 mMoves no measurementInterpolatedHeight of a standing stone in front of the house.No source
behuWidth0.44 mMoves no measurementInterpolatedWidth of that standing stone.No source
behuCount4Moves no measurementInterpolatedHow many behu this model raises when the household is entitled to. The real number is a record of the feasts that have been held and is not a figure the author can set; four is an honest placeholder rather than a reading of any one place.No source
behuSpacing1.90 mMoves no measurementInterpolatedSpacing between standing stones.No source
plazaOffset2.60 mMoves no measurementInterpolatedHow far the standing stones sit in front of the house.No source
jointEngagement0.30Moves no measurementInterpolatedDepth of the tested joint engagement, as a share of the smaller member.No source
postSeat0.30Moves no measurementInterpolatedDepth of the dish in the stone the post foot seats into, as a share of stone height.No source
sebuaScale1.22Moves no measurementInterpolatedSize of an omo sebua relative to an omo hada. That a si’ulu’s house is larger and has a loft is sourced; how much larger is the author’s.No source
everyBayTriangulated1Moves no measurementCanonEvery bay of the substructure is crossed by driwa: no rectangle of four posts is left without a diagonal. It is the only canon rule in this project that is not a statement about people but about the ground — South Nias sits on an active margin, a rectangle racks and a triangle does not, and that is why these buildings survive shaking. Note that this pack’s canon list is therefore doing two jobs at once.Gruber, P. and Herbig, U., “Settlements and Housing on Nias Island: Adaptation and Development”, in Proceedings of the International Conference on Vernacular Heritage and Earthen Architecture (2009).
bracingVisible1Moves no measurementCanonThe diagonal frame stands in an open understorey and nothing is put in front of it. The part that makes this house stand is the part easiest to see, which is true of none of the other five here.Feldman, J., The Architecture of Nias, Indonesia, with Special Reference to Bawomataluo Village (PhD dissertation, Columbia University, 1979).
raisedOnPosts1Moves no measurementCanonThe house stands on posts, and the understorey is tall and open.Viaro, A., Urbanisme et architecture traditionnels du sud de l’île de Nias (UNESCO / PNUD, Paris, 1980).
seatedOnStone1Moves no measurementCanonThe post feet stand on stones and are not buried — a footing allowed to move a little is better than one that has to break.Viaro, A., Urbanisme et architecture traditionnels du sud de l’île de Nias (UNESCO / PNUD, Paris, 1980).
windowBand1Moves no measurementCanonThe front is opened by a single continuous window band under the eave rather than by a row of separate holes.Feldman, J., The Architecture of Nias, Indonesia, with Special Reference to Bawomataluo Village (PhD dissertation, Columbia University, 1979).
behuAreEarned1Moves no measurementCanonBehu are raised by si’ulu, and each stone records a feast that was held. They stand outside the house and state something about the household that no part of the building states — the only rule in this project that adds something outside the building.Feldman, J., The Architecture of Nias, Indonesia, with Special Reference to Bawomataluo Village (PhD dissertation, Columbia University, 1979).
loftInRoof1Moves no measurementCanonA si’ulu’s house has a loft inside its roof. A roof that large is too valuable to leave empty.Departemen Pendidikan dan Kebudayaan, Arsitektur Tradisional Daerah Sumatera Utara (Proyek Inventarisasi dan Dokumentasi Kebudayaan Daerah, Jakarta).
streetTerrace1Moves no measurementCanonSouth Nias houses stand in a close terrace along a paved village street, side against side. An omo is not a free-standing object on open ground: it is one unit of a terrace, and the street is what it faces.Viaro, A., Urbanisme et architecture traditionnels du sud de l’île de Nias (UNESCO / PNUD, Paris, 1980).
streetWidth9.00 mMoves no measurementInterpolatedWidth of the paved street in front of the house.No source
neighbourHeight5.50 mMoves no measurementInterpolatedHeight of the neighbouring houses’ massing on either side. A neighbour is another omo and is entitled to a whole model of its own; what is drawn is a block the height of its eave, enough to read that this house is one unit of a terrace without inventing thirteen more houses.No source
pavingDepth0.12 mMoves no measurementInterpolatedThickness of the stone paving of the village street.No source
neighbourGap0.80 mMoves no measurementInterpolatedDistance to the neighbouring house on either side. That the houses stand in a close terrace is canon; a gap this small is the author’s.No source

Bibliography

  • feldman-1979 · reference

    Feldman, J., The Architecture of Nias, Indonesia, with Special Reference to Bawomataluo Village (PhD dissertation, Columbia University, 1979).

  • viaro-1980 · reference

    Viaro, A., Urbanisme et architecture traditionnels du sud de l’île de Nias (UNESCO / PNUD, Paris, 1980).

  • gruber-herbig-2009 · reference

    Gruber, P. and Herbig, U., “Settlements and Housing on Nias Island: Adaptation and Development”, in Proceedings of the International Conference on Vernacular Heritage and Earthen Architecture (2009).

  • depdikbud-sumut · reference

    Departemen Pendidikan dan Kebudayaan, Arsitektur Tradisional Daerah Sumatera Utara (Proyek Inventarisasi dan Dokumentasi Kebudayaan Daerah, Jakarta).

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. 1bayLength5.15 mangle of the diagonals
  2. 2floorHeight4.92 mangle of the diagonals
  3. 3bodyRows3.29 mdepth of the body
  4. 4bayDepth1.98 mdepth of the body
  5. 5ridgeRise1.37 mridge height
  6. 6wallHeight0.56 mridge height
  7. 7loftHeight0.49 mloft height
  8. 8postSection0.08 mdepth of the body

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 The building mirrors about z = 0; the behu on the plaza are not counted

    2008 points (219 of 223 parts); largest mirror deviation 0.00 mm.

  • pass
    Every tenon contained in its mortise

    84 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

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

  • pass
    Mesh integrity: indices, triangles, normals

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

  • pass
    Every bay of the substructure is crossed by a driwa: no rectangle left a rectangle

    45 bays — 24 along the building and 21 across it — every one triangulated. It is the only check in this project whose source is the ground rather than custom: a rectangle racks, a triangle does not. What is tested is the geometry; the strength of the members is not, and never will be — this project has no material properties, and inventing them would be exactly the kind of plausible number the provenance layer exists to refuse.

  • pass
    The understorey is open: what makes this house stand is what is easiest to see

    Nothing but stones, ehomo and driwa below a floor at 4.15 m.

  • pass
    The post feet stand on stones and are not buried

    28 ehomo seated on their stones. A footing allowed to shift a little is better than one that has to break — the same reasoning as the driwa, stated at the other end.

  • pass
    The front is opened by one continuous band, not a row of holes

    One opening 15.79 m long and 0.76 m high, stopping 0.90 m short of each corner.

  • pass
    The roof is larger than the body it covers

    A body of 2.81 m under a roof of 6.83 m — 2.43 times. The space inside it is too valuable to leave empty, and in a si’ulu’s house it is not.

  • pass
    A loft inside the roof in a si’ulu’s house, and none in an ordinary one

    Omo sebua: a loft at 9.39 m, inside the roof.

  • pass
    Behu stand on the plaza, outside the building, and only where the household is entitled

    4 behu, every one outside the eave line. Each stone records a feast that was held — the only thing in this project a household states without building anything on its house.

  • pass
    Every frame member stops where its roof plane runs out

    No point of the frame reaches outside the roof surface at its own height.

  • pass
    Thatch courses lap with no bare band, from the eave to the ridge

    31 courses over 9.78 m of slope.

  • pass
    Every part declares the dimensions it came from

    223 parts, every one tagged. By least-sourced input: 0 measured, 0 canon, 223 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 strongest check in this pack — that every bay of the substructure is triangulated — cannot be broken by pushing any single dimension, because the driwa are built from the same list of bays the check walks: widen a bay and its diagonal lengthens with it, for ever. That is deliberate, and it is why the claim is tested another way instead — by building a house and taking its cross-braces away, in the test file. What is shown here is a comparison of two genuinely independent numbers: this building’s roof is larger than its body, and shrink it far enough and what is left is a house with a hat rather than an omo.

The house as built

pass2.81 m / 6.83 mbody · roof

A body of 2.81 m under a roof of 6.83 m — 2.43 times. The space inside it is too valuable to leave empty, and in a si’ulu’s house it is not.

Rear prow raised to 2.24 m

fail2.81 m / 2.73 mbody · roof

A body of 2.81 m under a roof of 2.73 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.