A Methodology for The Image of the City

In The Image of the City Lynch defined a concept of imageability or legibility of the physical form by the pattern or structure of the interrelations among five elements. The well-known elements are: paths, edges, districts, nodes, and landmarks. When considered in pairs (for example, paths — nodes,...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Environment and planning. B, Planning & design. Planning & design., 1999-01, Vol.26 (1), p.133-144
1. Verfasser: Banai, R
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In The Image of the City Lynch defined a concept of imageability or legibility of the physical form by the pattern or structure of the interrelations among five elements. The well-known elements are: paths, edges, districts, nodes, and landmarks. When considered in pairs (for example, paths — nodes, paths — districts) Lynch noted the elements could strengthen or weaken, ‘reinforce’, or ‘destroy’ each other. The seminal study method, however, deemphasized the interrelations of the elements, critical to the observation of how a ‘complex visual whole’ is organized. To surmount this limitation a method is presented which aids in the measurement of the interrelations of Lynch's elements. Formally, the pairwise comparisons of the (five) elements are performed by means of Saaty's square, a reciprocal matrix with unit rank whose eigenvector solution gives the priority or the relative importance, or dominance, of the elements on a ratio scale. The inputs to such a matrix are provided by means of a questionnaire which solicits each field surveyor's perception of the relative dominance of the elements within each pair by using a nine-point scale. The proposed protocol is applied to an analysis of the image of a university campus as seen in the field by a group of surveyors. The new approach retains the essentially qualitative conceptual property of Lynch's near-classic methodology while allowing for a statistical quantitative account of observer variation and group consensus of the image(s) of structure.
ISSN:0265-8135
1472-3417
DOI:10.1068/b260133