Script&Shift: A Layered Interface Paradigm for Integrating Content Development and Rhetorical Strategy with LLM Writing Assistants

TL;DR

Script&Shift introduces a layered, zoomable interface to integrate content development and rhetorical strategy with LLM writing assistants, addressing the disruption caused by traditional linear writing tools. The approach combines three layer types (Writing, Meta, Document) with a Prompt Composer, Writer's Friends, and a Workspace Manager to support non-linear editing, cross-layer prompting, and iterative exploration. Through three example workflows and two evaluation studies, the paper demonstrates improved non-linear knowledge transformation, reduced cognitive friction, and richer creative exploration compared to inline-LMM and chat-based interfaces. The work offers a generalizable architecture for layer-based interaction and shows potential for broader adoption in education, journalism, and policy writing, while acknowledging learning curves and bias risks that warrant future work.

Abstract

Good writing is a dynamic process of knowledge transformation, where writers refine and evolve ideas through planning, translating, and reviewing. Generative AI-powered writing tools can enhance this process but may also disrupt the natural flow of writing, such as when using LLMs for complex tasks like restructuring content across different sections or creating smooth transitions. We introduce Script&Shift, a layered interface paradigm designed to minimize these disruptions by aligning writing intents with LLM capabilities to support diverse content development and rhetorical strategies. By bridging envisioning, semantic, and articulatory distances, Script&Shift's interactions allow writers to leverage LLMs for various content development tasks (scripting) and experiment with diverse organization strategies while tailoring their writing for different audiences (shifting). This approach preserves creative control while encouraging divergent and iterative writing. Our evaluation shows that Script&Shift enables writers to creatively and efficiently incorporate LLMs while preserving a natural flow of composition.

Figures (11)

• Figure 1: Workspace view of Script&Shift: (a) Zoomable, scrollable writing workspace, (b) workspace level operations, (c) a text layer with layer toolbar on top, and (d) the compiled document viewer.
• Figure 2: Key Interactions supported by Script&Shift.
• Figure 3: Content transformation through user prompts. The following 10 features allow writers to issue descriptive instructions for invoking specialized LLM assistance. The example column consists of real prompts issued by participants in the usability assessment.
• Figure 4: (A) system architecture and (B) example of how it is leveraged for Tone Tara. (A1&B1) The meta layer and writing layers provide the content, prompts and other details for prompt composer. (A2&B2) The prompt composer uses LLM Organizer and base prompts to structure the context and send the details for generation constraints using Layer Templatizer. (A3&B3) LLM outputs the structured responses. (A4&B4) The Workspace manager projects the response onto writing layers and document layers using the layer handler and LLM broadcaster.
• Figure 5: Free Writing Example with Script&Shift (A) Writer calls Template for organizing their writing into milestones for argumentative writing. (B) They select the "Ground" layer for further development. (C) They invoke Idea Ivy to brainstorm what to write next and then use Structure Sam to organize that into subheadings and paragraphs.