Skip to content
Thesis Info
- LABS ID
- 00623
- Thesis Title
- Towards a Model for Artificial Aesthetics: Contributions to the Study of Creative Practices in Procedural and Computational Systems
- Author
- Miguel Carvalhais
- E-mail
- miguel AT carvalhais.org
- 2nd Author
- 3rd Author
- Degree
- PhD
- Year
- 2010
- Number of Pages
- 488
- University
- University of Porto
- Thesis Supervisor
- Heitor Alvelos
- Supervisor e-mail
- halvelos AT gmail.com
- Other Supervisor(s)
- Penousal Machado
- Language(s) of Thesis
- English
- Department / Discipline
- Art and Design
- Copyright Ownership
- Miguel Carvalhais
- Languages Familiar to Author
- English, Portuguese
- URL where full thesis can be found
- www.carvalhais.org/txt/Carvalhais2010.pdf
- Keywords
- Algorithmic, Art, Design, Digital, Ergodic, Generative, Media, Procedural
- Abstract: 200-500 words
- This work proposes the development of an analytical model and associated terminology for computational aesthetic artifacts. Faced with the growing presence and widespread usage of computational media, we start by studying how they quantitatively transform previous media through remediation, and the qualitative transformations induced by their procedural and computational properties. To grasp the creative potential and uniqueness of computational media we develop a framework for their practice, critique and analysis.
We resort to Espen Aarseth’s typology for cybertexts as a starting point for the work, studying its adequacy for the analysis of ergodic visual and audiovisual pieces, adapting and expanding it with three variables and associated possible values. The model is tested in a set of samples that represent diverse approaches to procedural creation and contemporary clusters of activity. A control analysis is developed to assert the usability and usefulness of the model, its capability for objective classification and the rigor of the analysis.
We demonstrate the partial adequacy of Aarseth’s model for the study of artifacts beyond text-based systems and expand it to better suit the objects in study. We conclude that the model we present produces a good description of the pieces, clustering them logically, reflecting stylistic and procedural affinities between systems that, if studied from their physical or sensorial properties or from their surface structures alone would probably not be found to be similar. The similitudes revealed by the model are structural and procedural, attesting the importance of computational characteristics in the aesthetic appreciation of the pieces. We verify our initial conjecture about the importance of procedurality not only in the development and implementation stages of the works but also as conceptual grounding and aesthetic focus in artistic creation and appreciation, as an aesthetic pleasure in itself.