Abstract
We revisit Life in the Garden more than a quarter of a century later, aiming to leverage advances in modular storytelling techniques and large language models (LLMs) to address limitations of the original: (1) narrative incoherence and (2) the lack of a strong sense of authorship. We built an interactive storytelling system and co-performed the story in a live session to test participant engagement. The experiment produced both a screenplay and a visual comic in near real time. Here, we share practical insights from the process and lessons from participant feedback. Those seeking to create coherent and author-driven interactive stories may find this particularly useful.
Get full access to this article
View all access options for this article.
