Computational Developing: Rendering Semantic Pressure as a Perceptual Breakdown Surface
Source of Publication
Proceedings of the 31st International Symposium on Electronic/Emerging Art (ISEA2026)
Conference Year
2026
Abstract
Beneath the demand for coherence, linguistic variation, historicity, and narrative context begin to resist alignment, and it is in this forced proximity that semantics are pressed into alignment, producing a field of semantic pressure 1—a perceptual breakdown surface. This study establishes a reproducible artistic method that positions corpus data as developable material, stripping it of any function of referring to an absolute or external truth. Drawing on the conceptual lines of process cinema and cameraless photography, the method re-conceives “developing” as a computational operation. Through a data science framework, it registers the internal pressure of the corpus. To demonstrate the method’s operability, four sixty-second video variants are rendered from the same pressure field2, each shaped by a distinct temporal substrate. Duration and luminance modulation follow directly from local sparsity and neighbourhood contrast, making hesitation, drag, and boundary intensification perceptible as consequences of semantic deformation. Deterministic embeddings, operator definitions, parameter sweeps, and model–implementation alignment ensure that the perceptual behaviours originate from the corpus’s internal dynamics rather than implementation artifacts. The corpus, drawn from publicly accessible government art archives in Taiwan, exemplifies the tension between artistic narration and institutionalized formatting, making it an apt site for examining how pressure accumulates and becomes perceptible under computational conditions.
Document Type
Conference Proceeding
Publication Date
11-6-2026
Recommended Citation
Tai, Ming Shan, "Computational Developing: Rendering Semantic Pressure as a Perceptual Breakdown Surface" (2026). ISEA 2026 Proceedings. 18.
https://zuscholars.zu.ac.ae/isea2026/18