Perception and Creation of Diverse Alphabetic Styles

The Letter Spirit project explores the creative act of artistic letter-design. The aim is to mode] how the 26 lowercase letters of the roman alphabet can be rendered in many different but internally coherent styles. Viewed from a distance, the behavior of the program can be seen to result from the interaction of four emergent agents working together to form a coherent style sad to design s complete alphabet: the Imaginer (which plays with the concepts behind letterforms), the Drafter (which converts ideas for letterforms into graphical realisations), the Examiner (which combines bottom-up and top-down processing to perceive and categorise letterforms), and the Adjudlcator (which perceives and dynamically builds a representation of the evolving style). Creating a gridfont is an iterative process of guesswork and evaluation carried out by the four agents. This process is the "central feedback loop of creativity". Implementation of Letter Spirit is just beginning. This paper outlines our goals and plans for the project. 1. Fluid Concepts and Creativity The Letter Spirit project is an attempt to model central aspects of human high-level perception and creativity on a computer. It is based on the belief that creativity is an automatic outcome of the existence of sufficiently flexible and context-sensitive concepts m what we call fluid concepts. Accordingly, our goal is to implement a model of fluid concepts in a challenging domain. Not surprisingly, this is a very complex undertaking and requires several types of dynamic memory structures, as well as a sophisticated control structure involving an intimate mixture of bottom-up and top-down processing. The full realization of such a model will, we believe, shed light on human creativity. The specific focus of Letter Spirit is the creative act of artistic letter-design. The aim is to model how the 26 lowercase letters of the roman alphabet can be rendered in many different but internally coherent styles. Two important and orthogonal aspects of letterforms are basic to the project: the categorical sameness possessed by instances of a single letter in various styles (e.g., the letter ’a’ in Times and Optima) and the stylistic sameness possessed by instances of various letters in a single style (e.g., the letters ’a’ and ’k’ in Times). Figure 1 shows the relationship of these two ideas. The program will start with one or more seed letters representing a style and create the rest in such a way that all 26 share the same style, or spirit. To put the goals of Letter Spirit in perspective, it is enlightening to analyze what is lacking in many AI programs touted as "models of creativity". The problem is that they don’t Imow anything about ~hat they are doing. This phrase actually has two meanings, both of which apply to the programs under discussion. The first meaning is "the program has no knowledge about the domain in which it is working". The other is "the program has no internal representation of the actions it is taking, and no awareness of the products it is creating". These gaps are serious defects in anything that purports to be a model of creativity. .. . . . . . Fig. 1. Items in any column have letter in common. in any row have spirit in common. Items We insist that for a design program to be called "creative", it must meet the following requirements: ̄ The program must arguably make its own decisions, not simply carry out a set of design decisions all of which have already been made by a human; ̄ The program’s knowledge must be rich m that is, each concept must on its own be a nontrivial representation of some category, and among diverse concepts there must be explicit connections; ̄ The program’s concepts and their interrelations must not be static, but rather must be flexible and context-dependent; ̄ The program must be able to judge its own tentative output and be able to accept it, reject it totally, or come up with plausible ideas for improving it; ̄ The program must gradually converge on a satisfactory solution through a continual process in which suggestions and judgments are interleaved. We would argue that the deep (and controversial) question raised implicitly by the first point -"What would it mean for a program to make its own decisions?" -is answered by the last four points taken together. The Letter Spirit architecture is a rudimentary model 11 From: AAAI Technical Report SS-93-01. Compilation copyright © 1993, AAAI (www.aaai.org). All rights reserved. of these aspects of creativity. As is described below, several of its features -nondeterminism, parallelism, and statistical emergence --are key elements in allowing it to achieve these goals. 2. The Motivation of Letter Spirit