The five ‘global’ Space Syntax-type measures, Choice, Mean Metric Depth, Mean Visual Depth, Mean Angular Depth and Integration (HH) are all calculated using a unique ‘visibility flooding’ algorithm. The algorithm, based on a standard ‘jump flood fill’ algorithm (Rong & Tan, 2006), propagates an information cascade from a single origin point in the subject area to all other points. At each step in the flood process, the direct visibility, the metric depth, and the angular bearing of any newly-flooded point from its flooding point is assessed. The outcome is stored at the flooded point, and passed on to the next level of the cascade, informing the spread of the flood across all points. Fields of relative point depth can be rapidly composed using the resulting sequential visibility links.
Above: Cycles of visibility flooding in Hans Hollein’s Monchengladbach Museum
One complete flood of information across all points in the subject area is counted as a single ‘global cycle’. At the conclusion of each cycle, the depth values recorded at each subject point are concatenated into mean average values. A new origin point is then stochastically selected and the flood cycle restarted. As the cycles iterate, values are increasingly refined for all global measures at all locations within the subject area.