Every residential architect eventually meets the same client brief, phrased more or less identically. They want privacy from the neighbours. They also want the garden to feel open, the side courtyard to receive light, and the house not to feel like it is sitting inside a box.

These requests appear contradictory. They are not. They only become contradictory when the boundary is treated as a binary choice between solid and absent, which is how it is usually specified.

The resolution lies in an old idea that predates modern residential practice by centuries: screening geometry. Vertical arrays, calibrated gaps, layered depth and angular control let a boundary be opaque from one position and transparent from another. Traditional building cultures worked this out long ago, in the jali of Mughal architecture, the mashrabiya of the Arab world, the shoji and sudare of Japan. Contemporary suburban fencing has largely forgotten it.

The geometry of oblique opacity

The principle is straightforward. A vertical array of elements with a gap between them restricts sightlines as a function of viewing angle. Directly perpendicular to the array, you see through the gaps. As the viewing angle increases, the projected width of each element grows relative to the gap until, past a certain angle, the array reads as solid.

The critical angle is determined by the ratio of element depth to gap width. A deeper element with the same gap closes down faster. This gives the designer two independent variables to work with: the face width, which controls the visual rhythm, and the depth, which controls the privacy threshold.

In practice this means a boundary can be specified to be visually solid from every position within the neighbour’s habitable outdoor space, while remaining substantially open to sky, light and air from directly inside the courtyard. The screening is doing exactly what a solid wall would do, without any of the amenity cost.

Getting this right requires knowing where people actually stand. The relevant sightlines are not abstract; they run from specific windows, terraces and paths on the adjoining property to specific positions on yours. Plot them. The critical angle you need is set by the worst of them, not by an average.

Depth as a design instrument

Most residential boundary systems are effectively two-dimensional. The material sits in a single plane, and the design conversation is limited to colour, height and gap spacing.

Introducing depth changes the available vocabulary considerably.

Layered planes. Two arrays offset in depth and horizontally shifted produce a moiré condition where the effective opacity varies dramatically with position. Walk past it and the boundary opens and closes. This is theatrical, sometimes excessively so, but on a short frontage where a single condition would be monotonous it can be exactly right.

Angled elements. Rotating the vertical members off perpendicular biases the privacy in one direction. A boundary can be opaque toward the street and open toward the interior, which is frequently precisely what the brief requires and almost never what gets built.

Varying rhythm. Gap spacing does not have to be uniform along a run. Tightening the spacing where the neighbour’s window looks in and opening it where the boundary faces sky or planting produces a boundary that responds to its context rather than ignoring it. It also reads as designed, which uniform spacing never quite does.

Shadow. Deep elements cast shadow across their own face, and that shadow moves through the day. A boundary with depth is a different object at nine in the morning and four in the afternoon. A flat one is not.

The light question

On narrow urban lots, side setback amenity is often determined entirely by boundary treatment. A three metre setback with solid 1.8 metre boundaries on both sides is a light well with poor prospects. The same setback with permeable screening at the top third can be a usable outdoor room.

A useful strategy is to vary opacity vertically. Solid at the lower level, where privacy demand is highest because that is where people sit and where sightlines are direct. Permeable above, where the demand is lower and where light penetration matters most. This produces boundaries that perform better than a uniform treatment on both counts, and it is a straightforward detail: a solid base panel with a slatted or lattice extension above.

It also solves a common regulatory tension. Many local planning frameworks limit solid boundary heights on street frontages while permitting greater overall height where upper portions are visually permeable. The vertical split satisfies both requirements without compromise.

Material behaviour and climate

Screening exposes far more surface area per linear metre than a solid boundary, and it exposes it on all faces. Every element has four sides weathering simultaneously, plus cut ends. Where a solid panel presents a continuous protected face, a slatted array presents a large surface area of edges and returns.

This is a durability problem before it is an aesthetic one.

In salt-laden coastal environments, the consequence is direct. Coating systems on steel products are graded by exposure category, and the grade appropriate to an inland site is frequently insufficient within a kilometre or two of open ocean. In Perth, where a substantial share of the residential population lives within that band along a north-south coastline exposed to the Indian Ocean, the distinction matters commercially as well as technically. A boundary specified without regard to it will show edge corrosion at cut ends and fixings long before the flat faces show anything at all.

Powder coated aluminium sidesteps most of this, at higher initial cost. Timber screening in high-UV climates requires a maintenance commitment the client should be told about explicitly at design stage, because they will not thank you for discovering it in year three.

Thermal movement also deserves consideration. Long horizontal runs of metal screening in high-temperature environments will move measurably, and fixing details that do not accommodate it will either distort or fail. This is elementary, and it is routinely overlooked in residential work at a scale where the movement is assumed to be negligible. On a twenty metre run in a climate with a fifty degree annual surface temperature swing, it is not negligible.

The substructure nobody draws

Screening elements need something to attach to, and that something needs to be in the ground.

Post spacing is determined by the span capacity of the horizontal rails and the wind load on the screen. Here the permeability that solves the privacy problem helps structurally: a permeable boundary presents substantially lower wind load than a solid one of the same height, which permits wider post spacing and fewer visible structural elements. Designers who assume screening requires more structure than solid panels usually have it backwards.

Below ground, the specification is governed by soil. Reactive clay moves seasonally and demands greater embedment. Unconsolidated sand offers minimal lateral restraint and requires larger footing diameter to develop capacity. Limestone requires excavation methods that add cost and consequently attract shortcuts from installers pricing competitively.

The failure mode of an under-specified footing is not collapse. It is gradual rotation, which appears as a boundary that has gone out of plumb by a few degrees over several years. On a solid fence this is unattractive. On a precisely aligned screening array where the visual interest depends on the regularity of the rhythm, it is ruinous. The element that made the design work is the element that makes the failure obvious.

Specify the footing on the drawing. Do not delegate it. Where a project sits in a market with this much ground variation, it is worth engaging fencing contractors Perth based rather than assuming a national specification will hold, and FencrGatr routinely adjusts embedment across a single site.

Compliance intersections

Where screening approaches a pool, safety standards take over from design intent. AS1926.1 in Australia, and equivalent standards elsewhere, impose gap dimension limits and non-climbable zone requirements that eliminate most horizontal-element screening outright, because horizontals are climbable.

This is not necessarily a loss. Vertical screening with compliant gap spacing produces a legitimate architectural language, and the constraint of a maximum gap dimension imposes a discipline that often improves the result. What causes problems is designing the screening first and discovering the constraint at certification, which produces retrofitted infill panels that destroy the composition.

Similarly, boundaries shared with a neighbour carry legal weight in most jurisdictions. In Western Australia, dividing fences legislation establishes a right to cost contribution only up to a basic standard. Design ambition above that standard is funded unilaterally. This is worth raising with clients early, because a client who expects to split the cost of an architectural boundary and discovers otherwise late in the project will make the cheapest available decision at the worst possible moment.

Drawing the boundary

The practical recommendation is simple and rarely followed. Draw the boundary at the same level of resolution as any other elevation in the set.

Set out the elements. Dimension the gaps. Show the post positions and relate them to the building grid so alignments read as intentional. Detail the material transitions, the top and bottom terminations, and the junctions where the treatment changes. Specify footing depth against actual site conditions. Coordinate the pool barrier with the landscape rather than after it.

The boundary is the element the occupant sees most and stands closest to. It defines the outdoor rooms, controls the light, and forms the building’s presence from the street. It is worth an hour of drawing time.