Artificial Intelligibility

Artificial Intelligibility is a system for the accumulation and display of data, representing a growing history of relationships in architecture. By using a relational database for the storage of information, AI was immediately forced to answer the question, "What is relevant data?" In other words, recognizing that the scope of the history would be limited by the structure used to store and order that information, AI had to choose between establishing preliminary exclusions and developing an organic schema that could account for an indeterminate body of information.

Articulated positively: how might we structurally describe the social history of architecture? This question, implicit in every architectural history, becomes explicit when developing a digital model. Obviously one would include architects and their buildings, but this belies a particularly narrow conceptualization of the discipline. Then one could also include their books. But what about their friends? Their lovers? Their teachers? Their rivals? The list goes on as long as the historian's imagination. Quite simply, we could never practically predetermine an adequate set of relationships to tell the story. The problems extend further: what if these relationships change over time?

Traditional digital database schemas seemed to present a dead end. It seemed impossible for them to allow for the rich complexity of language, which, although correspondingly inadquate, suggested one version of the necessary organic schema. For this reason, Artificial Intelligibility began to model its memory on language.

Suddenly the information contained within AI became wildly diverse and the new problem was not one of memory, but action. How is it that Artificial Intelligibility tells its stories? Usually, graphical displays of related data are founded on simple, rigid schemas that manifest themselves in correspondingly simple node-edge diagrams. These diagrams lose their integrity when placed atop an open-ended schema like AI's, becoming confusing and meaningless.

The IAA has developed several prototypes to tell the stories of architecture contained within the AI's memory. They range from: attempts to approximate intelligible writing; to dynamic, emotive displays that open spaces for interpretation; to spatial forms of writing that marry diagram with text with iconographic image.

Gooey Network