Street Design & Complete Streets

Begin with a distinction that reorganises everything that follows. A road connects two places; a street is a place. A road's single job is efficient movement between origin and destination, and by that logic every extra second of delay is a defect to be engineered away. A street does something richer: it moves people and goods, but it also hosts commerce, conversation, protest, procession, play, and the simple business of being seen in public. Léon Krier put it sharply when he argued that the modern city confused the two, turning streets into roads and stranding buildings along traffic sewers. Jane Jacobs saw the consequence at eye level in The Death and Life of Great American Cities: the safety and vitality of a district rests on its sidewalks and the 'eyes on the street' that active frontages provide. When you widen a carriageway to shave travel time, you are quietly deciding that this stretch is a road and not a place, whether or not shopfronts, homes and temples still line it. Most Indian arterials were designed as roads while functioning, in daily life, as intense streets, which is the root of their danger and their disorder. Think of a stretch like Bengaluru's Old Madras Road or any state-highway segment that a town has grown around: engineered for through-speed, it is used for crossing to the vegetable market, waiting for a bus, and children walking to school, and the mismatch is measured in casualties. Camillo Sitte, writing in 1889, already warned that treating urban space as a purely engineering problem, all width and radius and flow, would strip it of the enclosure and incident that make a place feel human. Naming the type honestly is the first act of design, because the type dictates the standard: you do not hold a place to a road's speed, nor a road to a street's frontage expectations.

Everything a street does is negotiated across its cross-section, the slice you would see if you cut the right-of-way and looked at the cut face. Read from building to building, a well-made urban street organises into zones. Against the buildings sits the frontage zone, where doors swing, goods spill onto steps and cafes place chairs. Next comes the clear pedestrian through-zone, an unobstructed walking path, typically 1.8 to 2.5 metres, that a wheelchair or a person with a cane can traverse without detour. Then the furniture-and-multi-use zone absorbs trees, streetlights, bus shelters, vending, parking and, crucially in India, the utility boxes and hawker stalls that otherwise colonise the walkway. Only after all of this comes the carriageway for vehicles, ideally with a dedicated cycle track buffered from moving traffic, and a median where the street is wide. The discipline of the cross-section is that width is finite and every zone competes for it. Complete-street design makes that competition explicit and settles it in favour of the most vulnerable and most numerous users first. A useful discipline is to size the through-zone before anything else: fix a continuous, level walking path of at least 1.8 metres, enough for two people or a wheelchair to pass, and only then distribute what remains. Two further rules keep a section honest. First, the through-zone must be continuous, not interrupted every few metres by a pole, a ramp or a tree pit; discontinuity is what makes so many Indian footpaths technically present but practically useless. Second, every fixed object, from the transformer to the bus shelter to the vendor, belongs in the furniture zone by design, so it never migrates into the walking path. The figure shows a balanced arrangement you can adapt to widths from a 12-metre local street to a 30-metre arterial.

A complete street is one designed and operated to be safe and comfortable for everyone who uses it, regardless of age, ability, or mode: the eight-year-old walking to school, the eighty-year-old with a walker, the vegetable vendor, the cyclist, the bus passenger and, yes, the car. It is not a fixed template but a policy commitment that reframes the default. The conventional default asks 'how many vehicles can we push through here?' and treats footpaths as an afterthought. The complete-street default asks 'who needs to use this street, and how do we serve each of them adequately?' In practice this means continuous, obstacle-free footpaths, safe frequent crossings, protected cycling where speeds or volumes demand it, orderly on-street parking and vending, shade, drainage and lighting, all designed together rather than bolted on. The National Association of City Transportation Officials codified this thinking for practitioners in its Urban Street Design Guide, and the Global Designing Cities Initiative extended it worldwide in the Global Street Design Guide, which pointedly draws its examples from Mumbai, Addis Ababa and Sao Paulo as much as from New York.

India did not import complete streets so much as rediscover the necessity. ITDP India has worked with cities including Chennai, Pune and Coimbatore to redesign streets around the majority who walk, cycle and take the bus. Chennai's Corporation adopted a Non-Motorised Transport policy committing a share of the transport budget to footpaths and cycle tracks, and redesigned stretches of streets such as those around T. Nagar with continuous walkways, tabletop crossings that raise the crossing to footpath level and force vehicles to slow, and organised vending. The realities are specific and stubborn. Indian footpaths are routinely lost to parked two-wheelers, transformer boxes, debris and encroachment; a footpath that exists on the drawing but is unwalkable in practice is not a footpath. Good Indian street design therefore designs for the vendor and the parked scooter rather than pretending they will vanish, giving each a defined place so the through-zone stays clear. It also designs for the monsoon: a footpath that floods or a carriageway that ponds is a public-health problem, not merely a comfort one. The lesson is that a complete street in Chennai looks different from one in Copenhagen, and should.

How do you decide how much movement versus how much place a given street should serve? The 'link and place' framework, developed by Peter Jones and colleagues and now embedded in the Global Street Design Guide's thinking, gives you a two-axis matrix. The 'link' axis measures a street's importance for movement through the network; the 'place' axis measures its importance as a destination in its own right. A motorway is high link, low place. A quiet residential lane is low link, high place. A high street or a bazaar road is high on both, which is exactly why it is hard, and why treating it as a pure link ruins it. Plotting your street on this matrix tells you what to optimise: where place is high, prioritise slow speeds, wide footpaths and crossings; where link dominates and place is low, movement efficiency can lead. Crucially, it lets a city hold different streets to different standards on purpose rather than applying one arterial template everywhere. The figure maps common street types onto the matrix.

If you change only one thing about a street, change its design speed, because speed governs both safety and sociability. The physics are unforgiving: a pedestrian struck at 30 km/h has a high chance of survival, while at 50 km/h the odds invert sharply, which is why 30 km/h zones have become standard practice for streets with heavy pedestrian activity. Speed is set less by the posted limit than by the geometry: wide lanes, long sight-lines and generous corner radii invite fast driving, while narrower lanes, tight corners, on-street parking, trees and frequent crossings naturally slow it. This is the craft of traffic calming, done through design rather than enforcement. Tabletop crossings, chicanes, kerb extensions at corners that shorten the crossing distance, and raised intersections all trade a little vehicular speed for a large gain in safety and comfort. The Woonerf, or shared street, pioneered in the Netherlands, takes this to its conclusion by removing the kerb line entirely and letting people and slow vehicles share one surface, with the environment signalling that this is a place to move carefully. Barcelona's superblocks apply the same instinct at neighbourhood scale, calming interiors and reclaiming asphalt as squares.

The most powerful fact about street design is that the land already belongs to the public. A complete street rarely requires buying land; it requires redistributing a right-of-way the city already controls. This is politically fraught because it is visibly zero-sum: a metre added to the footpath is a metre taken from parking or a traffic lane, and someone will notice. Tactical urbanism has become the pragmatic answer, testing a redesign in paint, planters and bollards before pouring concrete, so the public experiences the benefit before committing budget. New York's conversion of Times Square from traffic to plaza famously began with lawn chairs. Pune and Chennai have run pop-up cycle-track and plaza demonstrations to build support. The deeper point is that street design is not a technical exercise appended to planning; it is planning, at the finest grain, distributing the single most valuable and most contested public asset the city has. Streets typically make up a quarter to a third of a city's land, more public space than all its parks combined, and unlike parks they touch every plot and every citizen daily. Reallocating that asset is therefore among the highest-leverage moves an urban government can make, and it can begin tomorrow with paint. Master the cross-section and you hold the lever that every other module in this course ultimately pulls, from density and land use down to the placement of a single bench in the shade.