Designing Visitor Flow: What Istanbul's Rami Library Play Pavilions Teach Students
How do designers guide visitor flow through a large exhibition or cultural space? Our answer, drawn from a project reported by designboom: give people destinations worth reaching, then design the journeys between them. Milimetrik architects transformed the courtyard and interiors of Istanbul's historic Rami Library into a 91,000 sqm environment for play and shared family experiences, fitted with chromatic pavilions and modular play structures, and welcoming approximately 825,000 visitors in eleven days. We read that as a studio problem in destination density, and the four stages below turn it into original exercises for analyzing, designing, connecting, and testing destinations in any large hall.
Stage 1: Measure destination density before drawing anything
Before drawing a single wall, this stage treats a large hall as a supply of reasons to move. The reported facts are spare: Milimetrik Architects turned the courtyard and interiors of Istanbul's historic Rami Library into a 91,000 sqm environment for play and shared family experiences, fitted with chromatic pavilions and modular play structures, and it welcomed approximately 825,000 visitors in eleven days. Divided out, that is an average of roughly 75,000 arrivals a day, though the source gives no hourly breakdown, so we treat it as a planning average rather than a measured curve. What the numbers imply is our own reading, not a source claim: at that scale, movement has to be organized by destinations, and the first measurable quantity is how many distinct destinations a floor actually offers per 1,000 square metres.
As our suggestion for this stage, pick a large civic space you know well — a station, a museum, a library hall — and inventory every distinct destination it offers before any design work. Fix the counting rule first: a destination is a spot that justifies a trip on its own, an exhibit, a café, a play unit, a service desk, while corridors, stairs and toilets count when people cross the building solely for them. Compute destinations per 1,000 square metres, then set that figure against the Rami Library's reported scale of one 91,000 sqm play environment and write a short comparison of what your number suggests about legibility at walking pace. To evaluate the result, swap counts with a partner and re-check the plan: if your densest zones match where people actually linger, the baseline is doing its job.
Stage 2: Design one self-explaining destination unit
Stage 2 turns from counting destinations to building one. The reported project gives the scale of the problem: Milimetrik Architects fitted chromatic pavilions and modular play structures into the courtyard and interiors of Istanbul's historic Rami Library, a 91,000 sqm environment for play and shared family experiences that welcomed approximately 825,000 visitors in eleven days. The source does not describe how any individual pavilion communicates its purpose; that gap is where our method starts. We treat each pavilion as a destination that must earn a trip on its own, so the design method is object-first: fix one shared activity, one small group size, and one freestanding form, and let geometry rather than captions carry the meaning. Everything else in this stage follows from that single-unit discipline.
As our suggestion for this stage, design a single freestanding pavilion for one shared activity, sized for a small group, under three constraints of our own. First, no label or legend anywhere: the form, the floor treatment, and the built-in furniture must imply the activity by themselves. Second, insertion without alteration: the reported project places new pavilions and play structures inside the courtyard and interiors of an existing historic library, so your unit must keep a visible clearance gap to walls, floor finishes, and services, touching the host fabric nowhere. Third, one activity only. To evaluate, hand the drawing to a classmate who must name the activity, the group size, and where users stand from the drawing alone; revise until all three answers come without asking you.
Stage 3: Draw three different journeys between the units
Stage 3 moves from single objects to the connections between them. The reported project spans a 91,000 sqm environment for play and shared family experiences, with chromatic pavilions and modular play structures set through the courtyard and interiors of Istanbul's historic Rami Library, and it drew approximately 825,000 visitors in eleven days — an average of roughly 75,000 arrivals a day by simple arithmetic. At that scale, the source reports what was placed but not how people were meant to travel between the placements, and that gap defines our method: treat the routes between destinations as a design layer of their own. Our premise, offered as our reading rather than a source claim, is that three route types with different priorities — shortest, browsing, sheltered — expose what a large floor actually does to a journey.
As our suggestion for this stage, take a shared plan of a large hall, place the four Stage 2 pavilions, and draw three alternative routes linking them. The first minimizes walking distance outright. The second is a browsing route that passes secondary attractions and lets visitors drift. The third is a sheltered route chosen for daylight and sightlines rather than speed. Two constraints of our own apply: every route must serve all four destinations, and no two routes may share more than a single crossing point, so each stays a genuinely different journey. To evaluate, score each route on length, what it lets a visitor see ahead, and how much of the hall it exposes, then require the three results to differ on at least two axes; where two routes score alike, redraw until the alternatives are real.
Stage 4: Audit the ensemble with a silent plan test
Stage 4 closes the sequence by auditing the whole ensemble rather than any single piece. The reported project sets the stakes: Milimetrik Architects fitted chromatic pavilions and modular play structures into the courtyard and interiors of Istanbul's historic Rami Library, a 91,000 sqm environment for play and shared family experiences that welcomed approximately 825,000 visitors in eleven days. The source reports what was built and how many came, but not whether the whole read as one coherent, walkable place, and that gap defines our evaluation method. Our premise, offered as our reading rather than a source claim, is that destinations and routes only work when a first-time visitor can assemble them unaided, so the final test must be silent: the drawing alone has to carry a stranger through every destination.
As our suggestion for this stage, exchange anonymized plans with a classmate and attempt to route a first-time visitor through every destination using only what the drawing communicates, with no written labels permitted anywhere. Two constraints of our own apply: wherever the reader hesitates, the designer must answer with a drawn spatial cue — a level change, a floor material shift, or a framed view — rather than added text, and every fix must keep the Stage 2 clearance and Stage 3 route rules intact. To evaluate, mark each hesitation point on the plan, apply the drawn cue, and hand the revised sheet to a second reader; the ensemble passes when the new reader routes through all four destinations without stopping to ask you anything.
Sources
- chromatic pavilions and modular play structures activate historic istanbul rami library — www.designboom.com,