177 lines
5.2 KiB
Markdown
177 lines
5.2 KiB
Markdown
# Phase 20 -- Story Intelligence User Experience
|
|
|
|
## Purpose
|
|
|
|
This document defines the intended user experience for Story
|
|
Intelligence.
|
|
|
|
It complements the technical architecture documents by describing how
|
|
Story Intelligence should feel from an author's perspective.
|
|
|
|
The goal is not simply to import a manuscript into PlotDirector.
|
|
|
|
The goal is to transform an existing manuscript into a rich, structured
|
|
PlotDirector project with the minimum possible effort from the author.
|
|
|
|
Throughout this document the author remains the final authority.
|
|
|
|
Story Intelligence assists the author.
|
|
|
|
It never replaces them.
|
|
|
|
------------------------------------------------------------------------
|
|
|
|
# Design Philosophy
|
|
|
|
Story Intelligence should feel effortless.
|
|
|
|
The author should never feel that they are performing data entry simply
|
|
so PlotDirector can understand their manuscript.
|
|
|
|
Instead, PlotDirector should quietly perform hours of structured work on
|
|
the author's behalf.
|
|
|
|
The author should feel that they have handed PlotDirector their
|
|
manuscript and, after a short wait, received a populated writing project
|
|
ready to explore.
|
|
|
|
The ideal experience is:
|
|
|
|
> "Upload my manuscript. Make a cup of tea. Come back to a project that
|
|
> already knows my story."
|
|
|
|
------------------------------------------------------------------------
|
|
|
|
# Core Principles
|
|
|
|
## The manuscript is the source of truth.
|
|
|
|
Story Intelligence never rewrites the author's work. It observes,
|
|
organises and suggests. The original manuscript always remains the
|
|
authoritative version of the story.
|
|
|
|
## AI assists.
|
|
|
|
AI never decides. Every imported item is a suggestion. The author may
|
|
change, merge, split or delete anything created during import.
|
|
|
|
## Structure before detail.
|
|
|
|
Before PlotDirector can understand a story it must understand its
|
|
structure. Story Intelligence therefore builds the structural framework
|
|
first before extracting detailed information.
|
|
|
|
## Review uncertainty, not certainty.
|
|
|
|
Authors should never be asked to review hundreds of AI decisions. Only
|
|
items with low confidence should be highlighted.
|
|
|
|
## Creativity should never be interrupted.
|
|
|
|
Story Intelligence performs its work in the background. It should never
|
|
repeatedly stop the author to request approval.
|
|
|
|
# Import Modes
|
|
|
|
## Automatic Import (Recommended)
|
|
|
|
The author supplies a manuscript and PlotDirector performs the complete
|
|
Story Intelligence pipeline automatically. No manual approval is
|
|
required during import.
|
|
|
|
## Guided Import
|
|
|
|
Guided Import performs the same analysis pipeline but pauses at selected
|
|
stages to allow the author to review AI suggestions such as scene
|
|
boundaries, character aliases, ambiguous locations and low-confidence
|
|
relationships.
|
|
|
|
# Story Intelligence Pipeline
|
|
|
|
## Stage One -- Structural Analysis
|
|
|
|
Outputs include chapters, suggested scene boundaries, narrative
|
|
structure, point-of-view changes and significant time shifts.
|
|
|
|
## Stage Two -- Semantic Analysis
|
|
|
|
Outputs include scene summaries, characters, locations, assets,
|
|
relationships, knowledge changes, timeline clues, questions, story
|
|
metrics and structured observations.
|
|
|
|
# Progress Experience
|
|
|
|
During processing PlotDirector should continuously communicate progress
|
|
using friendly, reassuring messages such as:
|
|
|
|
- Reading Chapter 12...
|
|
- Identifying dramatic scene changes...
|
|
- Building character profiles...
|
|
- Discovering important locations...
|
|
- Tracking significant story objects...
|
|
- Connecting relationships...
|
|
- Analysing emotional pacing...
|
|
- Mapping timeline events...
|
|
- Checking story continuity...
|
|
- Finalising your project...
|
|
|
|
# Import Completion
|
|
|
|
Rather than simply stating that the import has finished, PlotDirector
|
|
should summarise what has been created.
|
|
|
|
Example:
|
|
|
|
- [x] 47 Chapters
|
|
- [x] 183 Scenes
|
|
- [x] 64 Characters
|
|
- [x] 29 Locations
|
|
- [x] 17 Story Assets
|
|
- [x] Timeline created
|
|
- [x] Story metrics generated
|
|
|
|
# Import Report
|
|
|
|
The Import Report should highlight only lower-confidence items, such as
|
|
large scenes, possible aliases, ambiguous locations, low-confidence
|
|
relationships and scene boundaries requiring review.
|
|
|
|
# Confidence Philosophy
|
|
|
|
Confidence guides review effort. It should never prevent import.
|
|
|
|
# Author Review
|
|
|
|
Review is always optional. Authors can begin using PlotDirector
|
|
immediately after import and refine imported data over time.
|
|
|
|
# Future Improvements
|
|
|
|
Future capabilities may include additional AI providers, improved
|
|
prompts, better scene boundary detection, richer continuity analysis,
|
|
story health reports, character journey analysis and conversational "Ask
|
|
My Story" features.
|
|
|
|
# Success Criteria
|
|
|
|
Story Intelligence succeeds when:
|
|
|
|
- A first-time user can import an existing manuscript without prior
|
|
PlotDirector knowledge.
|
|
- The resulting project immediately feels recognisable as their own
|
|
book.
|
|
- Manual setup has largely disappeared.
|
|
- Review effort focuses only on uncertainty.
|
|
- The author remains in complete creative control.
|
|
- PlotDirector saves hours of repetitive project setup.
|
|
|
|
# Guiding Principle
|
|
|
|
Story Intelligence should never interrupt creativity.
|
|
|
|
It should quietly perform hours of structured work while the author
|
|
makes a cup of tea.
|
|
|
|
When the author returns, PlotDirector should already understand the
|
|
structure of their story.
|