Landscape - Book Structure

This document describes the logical structure of Landscape: A Fact-Based Picture of Architecture and AI at Work, for Leaders, Architects and Citizens, a companion to The Transformed Architecture. The book is organised into four parts with 20 numbered chapters, followed by a glossary, three appendices, the references and a note about the author. It is modelled on Factfulness: it opens with a test, gives one notation drawn the same way every time, names nine instincts that make people misread the landscape of architecture and AI at work, corrects each with evidence from public datasets, and gives one rule of thumb per instinct. Its stance is possibilist, and every statement about the future is labelled observable, expected or speculation. The book stands on its own: every concept is defined in it before use, and the seven volumes of the series are cited as external works.

Table of contents

Part I: The Test and the Picture (Chapters 1 to 4)

The test the reader takes, the pictures that have shaped how people see technology, the new landscape in which AI acts with identities and under decisions, and the notation that draws it the same way every time

Part II: The Instincts (Chapters 5 to 13)

Nine instincts, one per chapter, each with the questions it explains, the instinct named, the evidence, the picture corrected, a rule of thumb and what an executive, an architect and a public official can do

Part III: Drawing Your Landscape (Chapters 14 to 18)

How to draw an organisation's own landscape from the registers it already has, read it over a year, place its suppliers and jurisdictions, see who decides and how we know, and compare two landscapes

Part IV: Seeing Clearly (Chapters 19 to 20)

The nine rules of thumb in practice, and the author's own view of the next years, labelled

Back matter

Glossary, Appendix A (the notation reference card and the landscape model), Appendix B (drawing the views by hand, and the rules on one page), Appendix C (data sources and methods), references and about the author

Chapter structure

Part I: The Test and the Picture

The part is introduced by part_i_the_test_and_the_picture.md.

Chapter	File	Title	Purpose
1	01_introduction.md	Introduction: A Test, a Picture and a Stance	Give the reader the test (Q1 to Qn, each traceable to a dataset in data/, with survey answers beside measured answers where a survey exists), introduce Nordvik Region and Hansa Mutual, name the possibilist stance and acknowledge Factfulness as the inspiration
2	02_the_old_pictures.md	The Old Pictures: Boxes, Binaries and the Cloud	Show the pictures people carry of technology at work, one box labelled "AI", automated against manual, the cloud as a place, and why each hides what matters
3	03_the_new_landscape.md	The New Landscape: Who Acts, With Whose Identity, Under Whose Decision	Describe what emerges when architecture becomes code and AI tools go to work: units that act, identities they act with, decisions someone holds, and define the modes, levels and other adopted vocabulary with attribution
4	04_the_notation.md	The Notation: One Picture, Drawn the Same Way Every Time	Teach Contour: the unit and the mark, the seven dimensions, the three frames, the six views, the legend and how to read a landscape in thirty seconds

Part II: The Instincts

The part is introduced by part_ii_the_instincts.md. Every chapter follows the Part II chapter form of CHAPTER_AUTHORING_GUIDE.md.

Chapter	File	Title	Purpose
5	05_the_box_instinct.md	The Box Instinct: AI Is Not One Thing	Replace the one box labelled "AI" with the six usage modes and the units they contain, against the data on which technologies enterprises report using
6	06_the_binary_instinct.md	The Binary Instinct: Neither Automated Nor Manual	Replace automated against manual with the five autonomy levels and the line between Level 2 Supervised and Level 3 Bounded
7	07_the_model_instinct.md	The Model Instinct: The Use Sets the Risk and the Value	Show that the use, not the model, sets risk and value: the same model in two uses is two units with different reversibility, reach and control
8	08_the_straight_line_instinct.md	The Straight-Line Instinct: Curves That Bend	Show adoption and capability series that bend, saturate or step, against the straight lines people draw through them
9	09_the_fear_instinct.md	The Fear Instinct: The Loud Harms and the Quiet Ones	Set the harms that make headlines beside the harms the incident and survey data show to be common, and draw where each sits on the landscape
10	10_the_size_instinct.md	The Size Instinct: Numbers Without Denominators	Put the denominators back: per enterprise, per person reached, per decision, per euro, and show how reach is drawn
11	11_the_destiny_instinct.md	The Destiny Instinct: Nothing Here Is Fixed, and Nothing Is Inevitable	Show how landscapes changed between surveys and within a year, that reversibility decays and can be restored, and that no mark is destined for the empty corner
12	12_the_single_perspective_instinct.md	The Single-Perspective Instinct: Security, Compliance and Productivity Share One Landscape	Show that the security, compliance and productivity views of an organisation are the same marks read differently, and what each perspective misses alone
13	13_the_blame_instinct.md	The Blame Instinct: When Nobody Designed Who Decides	Show that incidents blamed on a person or a model trace to marks without a seat, and that the seat and the policy point are designed, not found

Part III: Drawing Your Landscape

The part is introduced by part_iii_drawing_your_landscape.md.

Chapter	File	Title	Purpose
14	14_the_one_page_landscape.md	The One-Page Landscape: Drawing It From What You Already Have	Populate the landscape model from the registers an organisation already keeps (the AI system register, mandates, the delegation register) and render V1
15	15_the_landscape_over_a_year.md	The Landscape Over a Year: Reading Change	Read V2: trails, new and retired marks, drift between modes and levels, and the decay of reversibility within a year
16	16_supply_sovereignty_and_jurisdiction.md	Supply, Sovereignty and Jurisdiction: Where the Landscape Runs	Read V3: the organisation's own boundary, suppliers by role, EU, EEA and non-EU jurisdictions, and model chips, without brands
17	17_decision_rights_and_evidence.md	Decision Rights and Evidence: Who Decides, and How We Know	Read V4 and V5: the seat and its tier per unit, the policy point, and whether the proof of control is automatic, attested or absent
18	18_a_year_at_nordvik_and_hansa.md	A Year at Nordvik Region and Hansa Mutual: Two Landscapes Compared	Follow both organisations through 2027 in V2 and compare them in V6, as illustration, never as data

Part IV: Seeing Clearly

The part is introduced by part_iv_seeing_clearly.md.

Chapter	File	Title	Purpose
19	19_the_rules_of_thumb.md	The Rules of Thumb in Practice	Collect the nine rules of thumb, map each test question to the chapter that explains it, and show the rules applied to a board paper, an architecture review and a public procurement
20	20_the_next_years_and_a_conclusion.md	The Next Years, and a Conclusion	Give the author's own projections for the landscape, labelled as such, and conclude: what the data show, what is possible, and what the reader can do

Back matter

The back matter is introduced by part_appendices.md.

Item	File	Title	Description
Glossary	21_glossary.md	Glossary	Defines the vocabulary of the book: the notation's names and scales, the adopted terms with their attribution, the method's terms and the European regulatory terms
Appendix A	22_appendix_a_notation_reference.md	Appendix A: The Notation Reference Card and the Landscape Model	The legend and the seven dimensions on one card, the six views, and the landscape model schema with an annotated example
Appendix B	23_appendix_b_drawing_by_hand_and_the_rules.md	Appendix B: Drawing the Views by Hand, and the Rules on One Page	One hand-drawing page per view, and the nine rules of thumb on one page
Appendix C	24_appendix_c_data_sources_and_methods.md	Appendix C: Data Sources and Methods	Every dataset the book uses, with publisher, licence, query, retrieval date and caveats; how the test questions were built; how estimates were made
References	25_references.md	References and Sources	Canonical bibliography; the seven volumes are Sources 1 to 7
About the Author	26_about_the_author.md	About the Author	Background and perspective

The figure kinds

Every figure is rendered from a committed source (DIAGRAM_STYLE_GUIDE.md):

Kind	File name	Rendered from	Caption
Notation view (view_*)	docs/images/view_<org>_<view>_<date>.png	landscape/models/<org>.yaml by scripts/render_landscape.py	"Illustrative"
Data chart (chart_*)	docs/images/chart_NN_topic.png	charts/chart_NN_topic.vl.json and data/<dataset>/tidy.csv by scripts/check_charts.py	"Source: ... (retrieved 2026)"
Process diagram (diagram_*)	docs/images/diagram_NN_topic.png	docs/images/diagram_NN_topic.mmd by scripts/check_mermaid_diagrams.py	"Illustrative" or a source line

How the parts build

Part I gives the reader three things to carry: the test, which shows them their own picture; the vocabulary of the new landscape, defined in the book before use; and the notation, which draws any organisation the same way every time. Part II uses all three: each instinct chapter returns to the test questions it explains, shows the evidence in a data chart, and shows the picture corrected in a notation view, ending in a rule of thumb. Part III turns from seeing to drawing: how a reader populates the landscape model from the registers an organisation already keeps, and how each of the six views is read, followed through a year at the two illustrative organisations. Part IV collects the rules and applies them, and closes with the author's own projections, labelled, and a conclusion. The back matter gives the reference card, the hand-drawing pages, the rules on one page and the methods behind every number.

Companion files

File	Purpose
BOOK_BLUEPRINT.md	Premise, audience, the brief of every chapter and its status
NOTATION.md	The specification of Contour
FOUNDATIONS.md	The conclusions of the seven volumes the book adopts (F1 to F84), the diagram survey (N1 to N13), the open questions (Q1 to Q9) and the tables of contents of the volumes for cross-references
VISUAL_IDENTITY.md	The visual identity of the book and its figures
EXAMPLE_ORGANISATIONS.md	Canonical profiles of Nordvik Region and Hansa Mutual
CHAPTER_AUTHORING_GUIDE.md, STYLE_GUIDE.md, AUTHOR_PERSONA.md, DIAGRAM_STYLE_GUIDE.md	Rules for structure, language, voice and figures

Canonical ordering

scripts/toc_definition.py holds the table of contents and regenerates docs/book_index.json, the mkdocs.yml navigation and the chapter list in BOOK_REQUIREMENTS.md. docs/book_index.json and mkdocs.yml must agree; tests/test_book_index.py enforces that they do.