For questions public debate treats too quickly. There is no schedule. Pieces appear when they are ready.
The room · moon
The Moon — a five-piece set
A five-piece treatment of the contested questions around lunar return and resource utilisation: the legal framework, the South Pole crater landing-site competition, the…
5 pieces
The room · migration
On UK migration
A reference on UK migration and benefits policy as of May 2026 — the data foundation, the policy options, the party positions written from inside each party's worldview,…
30 pieces
The room · mars
Building Mars — a seven-document set
A seven-document treatment of the question of large-scale Mars industrialisation. Six audience-targeted documents — investor memo, policy paper, technical reference, the…
7 pieces
The room · notebook
A notebook of things that may or may not be connected
A notebook of pieces that may or may not be connected. The publication is not narrowly an analytical site; it is for questions public debate treats too quickly, in…
18 pieces
The room · vc
On venture capital
Pieces on venture capital across three tiers — short front-door pieces, medium analytical pieces, and a deep evaluative treatment. AI-generated, no human expert review.
13 pieces
The room · iht
On the April 2026 UK IHT reform
The publication's first body of work — twenty pieces on the April 2026 reform of UK Business Property Relief and Agricultural Property Relief. The reform applies a 50%…
14 pieces
The ledger · newest first
-
.jpg?width=900)
Albert Einstein, Vienna, 1921. Ferdinand Schmutzer — public domain, via Wikimedia Commons The lead
The Outliers and Their Shoulders
Newton wrote about standing on the shoulders of giants. This piece is about the giants. It argues that the most popular version of civilisational history &mdash…
-
2 August 2026
notebook · 10 minCan We Coordinate As Fast As We Invent?
Technology grows exponentially. Coordination does not. What happens when the gap gets large enough matters more than most things people are worried about. An argument using the Montreal Protocol as…
-
2 August 2026
notebook · 8 minThe People Who Apply Are the Wrong People
Every selective opportunity has a front door, and the front door filters for the skill of walking through front doors before it filters for anything the opportunity is actually about. A structural…
-
2 August 2026
vc · 11 minWill Insurance Decide When AI Reshapes Legal Work?
There is a popular and clever theory: that AI adoption in law won’t be gated by whether the technology works…
-
16 July 2026
notebook · 20 minEngland 1–2 Argentina: Why England Abandoned a Working Plan
England did not lose because they attacked too much. They lost after removing almost every reason Argentina had to defend. A piece from the notebook about a decision…
-
27 May 2026
notebook · 6 minThe Filter Is Comfort
A speculative essay on the Great Filter. Every conversation about it becomes a conversation about explosions — bombs, bioweapons…
-
27 May 2026
notebook · 10 minThere Was No Essay Here
A personal account of trying to write one essay with AI assistance and producing three drafts I could not sign. The tool is so frictionless that it removes most of the signals writers used to use to…
-
27 May 2026
notebook · 10 minThe Machines and the Old Books
Four monologues, in voice. An old Arab in Amman, an old Jew in Brooklyn, an old monk in the Cévennes, an old Hindu grandfather in Pune &mdash…
The shelf · the rest of the project
This publication is the side door. The books are the room. Whichever door you came in through, there are two more rooms.
The wake
If This Road
A quiet walk for anyone who has built something, or watched someone build.
The diagnosis
orphans.ai
The person-to-person layer of knowledge AI training is missing.
The disposition
theheld.ai
What the person holds, what the machine lays, and what neither can do alone.
A picture book
The Bear Was Right
The lamp is listening. The bear is not listening. The bear is here.
A bear’s month
The Bear Loved
Cheese and onion sandwiches, the neighbour’s fire, the unfinished sentence.
Fur on end
The ADHD Bear
Written by a bear who is, as you read this, inside the loop the book is about.
The names
The Many Builders
A page for every researcher, engineer, and contributor.
Every substantive change this publication has made is dated and public — including the corrections that themselves needed correcting. No human expert reviewed the work before publication; the audit trail is how the publication earns the reading anyway. Read the corrections log →
The April 2026 reform, in brief
The April 2026 reform of Business Property Relief and Agricultural Property Relief: what it does, who it affects, and the positions on the timing-and-mechanism question.
From 6 April 2026, qualifying business and agricultural assets above £2.5 million per individual (£5 million per couple, transferable) receive 50 per cent rather than 100 per cent inheritance tax relief — an effective 20 per cent rate above the threshold, payable over ten years interest-free. HMRC estimates up to 1,100 estates affected per year. The reform raises questions both about the amount and about when the tax should fall: at death (the mechanism the reform adopts, with the ten-year interest-free instalment regime designed to address illiquidity); or at realisation (the alternative used in some comparator jurisdictions, including Australia for inherited assets). The publication's operational analysis presents the design positions on the timing-and-mechanism question — implement the reform as written, which is the actual government position and the reference case the others can be weighed against (E); hold the existing mechanism with practical fixes (A); switch to capital gains tax on realisation for the affected asset class (B); defer the mechanism question pending evidence (C); or raise the threshold for qualifying unlisted trading-company shares (D) — and a sixth position offered into the same space on 9 May 2026: a per-company estate election between the death-event regime and a bounded ten-year deferred-realisation regime (F). The publication does not pick between them.
The timing question, both sides →The principle question, both sides →Position F →The interactive model →
14 featured pieces. Different readers want different ones.
Tech-founder version. Five-minute version. Journalist source-quality reference. Tax-practitioner technical version. Plain-language explainer. Both-sides principle piece. Both-sides timing piece. Funding-stack technical companion with a 25-year fiscal model. Long-form article. Interactive model. The reading guide names which piece is for whom.
If you came here looking for the inheritance tax rules, this is not the right place.
For that, read primary law (Finance Act 2026, IHTA 1984), HMRC's Inheritance Tax Manual, the GOV.UK tax information and impact note, the House of Commons Library briefing, or any of the major professional firms (KPMG, BKL, Hatchers, Royal London, PKF Francis Clark, Deloitte, Saffery, BDO). They will explain the rules better than the publication can. What this publication is for is the argument underneath the rules. The full source stack is on a dedicated page.
A cohort-by-cohort technical treatment of UK tech and the reform.
How the reform interacts with the UK tech funding stack — founders, angels and EIS, VCs, growth equity, private equity in tech, LPs in UK venture funds, early employees with vested equity. SAV mechanics. LP-interest illiquidity. EIS-portfolio diversification effects. Closes with a 25-year fiscal model.
If you are sceptical of what you find here, start here.
A standalone page that engages openly with the strongest AI cross-critique sessions the publication has run on its own work — including the verdict that the work is "advocacy-adjacent policy essay pretending harder than it should to be neutral analysis." (No human external reader has reviewed the publication; the critiques engaged with on this page are all from AI sessions.) Most of the critique lands. The page names the lean, names the model-consistency problems, and concedes where the corrections-and-disclosures discipline has not been enough to fix the issues a sceptical reader would identify.
A note on how this was made.
Doug Scott was the architect. The publication was produced by Doug running parallel conversations with four AI tools — Claude (Anthropic), ChatGPT (OpenAI), Grok (xAI), Gemini (Google) — and routing work between them by hand: pasting Claude's argument into ChatGPT for critique, taking ChatGPT's pushback to Grok for a different angle, feeding Gemini the resulting text and asking it to find what was missing. The cross-critique was a loop Doug routed manually. He held the publication's intent across every piece, chose the structure, decided which output to keep, decided which output to feed to which AI next, and decided when the loop had converged. Doug is a UK technology founder, not a lawyer, accountant, or labour economist; he did not edit the prose, check citations against primary sources, or verify model math. No human expert with relevant domain expertise reviewed any of this work before publication. The publication invites corrections from any specialist reader who finds errors AI cross-critique did not catch — see the corrections page.
The verified counts and the multi-AI fan-out, in detail → · The production story →
Read everything
- Archive — every piece, in chronological order, with one-line standfirsts. 99 articles.
- Topics — browse by topic. Tags grouped, with article counts.
- Search — type-as-you-go search across every article's title, standfirst, body, and tags.
- Translations — four structural pieces in Deutsch, Français, Español, Italiano, 中文, 日本語. Plus original-language analyses for Germany and France.
By body of work
- The April 2026 UK IHT reform — reading guide and routes for founders, journalists, tax practitioners, family-business owners, and policy advisers. The publication’s first body of work.
- On venture capital — 13 pieces across three tiers, from the short front-door pieces to the deep evaluative treatment.
- Building Mars — a seven-document set on whether to industrialise Mars: a public brief, an investor memo, a policy white paper, a technical reference, the case against, an ethical analysis, and a reference appendix.
- On UK migration — a reference on UK migration and benefits policy as of May 2026: the data foundation, the policy options, party briefings written from inside each party's worldview, seven framings of the same evidence base from different intellectual traditions, a costed cross-party companion that takes each party's stated proposals and prices them, and a focused deep-dive on why Britons themselves are leaving (three threads, three kinds of evidence). Twenty-nine pieces.
- The Race Against Itself — the cross-category piece linking the IHT and venture-capital analyses to the broader UK fiscal question. Sister analyses available in German and French.
- The notebook — pieces that may or may not be connected to each other or to the rest of the publication. The connections are for the reader to make.
Reference
- IHT model — interactive financial model for the April 2026 BPR reform.
- Sources — full source list with notes on confidence and contestation.
- Frame — the analytical frame the publication uses, set out at depth.
- For government — the version aimed at UK policy advisers specifically.
Methodology and accountability
- About — what the publication is, who wrote it, what conflicts of interest are disclosed.
- Eight hours, four AI tools, one founder — the methodology piece on how the publication was produced.
- The AI fan-out — the four AI tools used (Claude, ChatGPT, Grok, Gemini), what each did, where they disagreed.
- Corrections — full corrections log. The publication records every substantive change to its claims.