I just saw a comparison between the 787 and Concorde windows. I could be wrong but it almost seems like a hand could cover the majority of the window.
The Concorde flew above 15,000m. At this altitude a sudden reduction in cabin pressure would prove hazardous to crew and passengers with most falling unconscious within a few seconds. The low air-pressure would also render the oxygen supply system inefficient and most passengers would suffer from hypoxia.
Thus we have the reason for the smaller windows, should a breach occur then the size of the hole being small would reduce the rate if loss of air-pressure inside the cabin. This, combined with a reserve air-supply to augment the cabin-pressure as well as a rapid rate of descent manoeuvre to bring the aircraft to a safe altitude would reduce the the risk of hypoxia. So basically the small windows were designed to reduce the rate of air escape from the cabin should a hull breach occur.
The fuselage is a pressure tank, the window is a hole in the construction of the pressure tank. Adding windows also adds weight: the pressure vessel construction must be reinforced around the hole. The window glass is obviously airtight, but does not contribute in absorbing any of the stresses of the pressure differential.
The window size is a function of:
- The pressure differential. The higher the aeroplane flies, the larger the pressure differential between internal and external pressure.
- The relative size of the hole, relative to the fuselage diameter, fuselage length, and aerodynamic bending forces.
For construction engineers, the best size of a window is zero. The windows in Concorde must be smaller because the fuselage diameter is much less than that of a B787 while it must handle a higher pressure differential.
Airflow Rate through an Orifice will help give some idea of the time it takes for the cabin of the Concorde to evacuate after one window blows at 60,000 feet as follows:
Volume of the cabin: $50m × pi × 2m^2 = 620 m^3$, Diameter of orifice: $0.15 m$
Rate of outflow from calculator software, averaged for pressure difference from T0 to T fully evacuated: around $150 m^3/min$ yielding roughly 180 to 240 seconds to full evacuation, without pressure equalization (reserve air-supply), with debilitating unmasked hypoxia sooner.
The location of the orifice (open window), on the side of the fuselage, could possibly cause even lower evacuation time values due to the Bernoulli effect of the passing airstream.
This underscores the point of view that strengthening the windows was indeed of critical importance. Further examination of air replenishment capacity and emergency descent capability are in order, as newer versions of these supersonic transports may yet return to commercial use.