Printing was introduced to the Indian subcontinent in the late sixteenth century. Ever since this time, the designing of Devanagari typefaces has largely been an intuitive activity, more often than not due to the fact that there exists no single, unanimously accepted body of work that formalises the structure and display of Devanagari typefaces. This leads to the conclusion that the domain knowledge of Devanagari typography is complex, ambiguous, and non-standardized—hence inadequately interpreted. One can perceive a need here for an academically validated theoretical model that describes the various aspects of Devanagari typefaces and presents a more informed view of Devanagari typography. Current type designers of Devanagari do follow certain tacit models while designing fonts, but none of these have been well documented, nor have they been subjected to academic consolidation or validation. There also have been attempts by designers and theorists to pen down some of their views on Devanagari lettering, calligraphy, and type design; most of this work, however, is based on antiquated technologies, and no authors have commented on contemporary typefaces and technologies.

An area that has been directly affected by the lack of a formal description is the classification of typefaces. Through our research, we found the current classification systems for Devanagari to be inefficient. In order to understand how experts and nonexperts view Devanagari typefaces, we captured and analysed the classification strategies used by experts and nonexperts. Rigid classification systems, when used in isolation or based upon a single parameter, fail in the identification and integration of multiple points of view.

This research attempts to solve these problems by creating an explicit, formal ‘conceptual model’ that captures concepts and their relationships within this domain and theoretically represents this domain exhaustively. This model was also used in the creation of a faceted classification system for Devanagari typefaces. Through our research, we suggest that a web-based faceted classification tool developed (implemented using HTML,Flex, PHP, and MYSQL) based on the conceptual model allows for more accurate searching (in identity matching tasks) and better short listing (browsing) Devanagari fonts for general purpose use by experts as well as minimally informed users.