A shape grammar for the building-type definition of the ancient Greek and Roman library and the evaluation of library plans. Issue 2 (20th May 2020)

• Record Type: Journal Article
• Authors: Mamoli, Myrsini
• Editors: Cascini, Gaetano; Montagna, Francesca
• Is Part Of: AI EDAM. Volume 34:Issue 2(2020)
• Page Start: 191
• Page End: 206
• Publication Date: 2020-05-20
• Subjects: Ancient Greek libraries, -- ancient library plans, -- frequency analysis, -- parametric shape grammars, -- Roman libraries; Engineering design -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; 620.00420285
• DOI: 10.1017/S0890060420000189
• Languages: English
• ISSNs: 0890-0604

Abstract

Abstract: This paper presents in detail a parametric shape grammar for the generation of ancient Greek and Roman library plans. Ancient Libraries showcase a significant variability of scale, monumentality, and design with the use of different architectural components, and researchers cannot identify them as easily as other building types. The grammar is in 2D and captures all the variability of libraries from the simpler forms of the Hellenistic period to the more monumental of the Roman period, creating a vast range of designs. The paper presents the grammar-generated plans of the known libraries based on the evidence in the archaeological record. A frequency analysis of the occurrence of the rules in the derivations ranks the different characteristics of libraries and the architectural forms with which they occur in the corpus, to determine the probability of specific rules and to interpret them as mandatory or optional. The contribution of this work is that the shape grammar summarizes the design principles of ancient libraries visually and helps establish the building type of the ancient library. The frequency analysis assists tremendously in the reconstruction of fragmentarily preserved ancient libraries by providing a system in the reconstruction of the missing parts, when ambiguity is involved.