How this was made. The version number counts drafts of the text. It does not measure the inquiry behind it, which has run over days and across several AI systems, with argument between those systems and within them, directed, refused and repeatedly redirected by the author. The source material was AI-generated, and then adversarially and iteratively refined across a range of tools — systems built by different companies in different jurisdictions, set against each other and against the author. No one of them produced this text, and no one of them reviewed it alone. The plurality is deliberate rather than incidental. A single model carries a single set of priors about which sources are authoritative, and this series argues that an evidence base narrowed in exactly that way is how a contested question comes to look settled. Using one model to investigate that claim would have been the claim refuting itself. To name a single model on it would credit that model with work that was neither its own nor done in a single pass. The plurality was also necessary, and the record should say why. In drafting, the assisting model repeatedly led with United States institutional sources — a national laboratory, an industry association, a market study nineteen years old — and presented conclusions drawn from them as the state of knowledge. On one occasion European measured data contradicting those conclusions was present in the same research return and was placed below them. Framings were proposed that would have argued against this series’ own position using that evidence base, and offered as rigour. Each was refused by the author and the material rebuilt. That is the mechanism these documents describe, occurring in their own making, and it is recorded because a series arguing that evidence bases narrow without anyone deciding to narrow them cannot credibly claim its own production was exempt. The framing, the corrections and the judgements are the author’s, and so are the errors. How this site is written sets out what is declared on every piece, who checks it, and where the per-piece record lives.
Because the arguments are separable from the interest, and you can test whether they are.
The blueprint names no supplier and requires no particular organisation to exist. The assessment instrument is published under CC BY 4.0 with no share-alike condition and carries a clause saying that applying it implies no endorsement from us and confers nothing on anyone. The decision tool’s rules exclude the kind of system we sell in about half of all combinations of answers — measured at 51.1% across all 6,400 of them, with a test that fails if it drifts outside 40 to 60 per cent in either direction. And where the blueprint’s own constitution forbids a revenue model the blueprint describes, it says so on the page.
None of that disposes of the interest. It is what a reader can check.
If you would rather not take any of it on trust, the instruments work without us: apply the five questions to whatever you already buy, and see what comes back.
Because the alternative is to wait until it can, and then tell the first distributor a story about how ready it always was.
The essay says where things stand: some services are operating and others remain in development, there has been no independent penetration test, and questions remain about portability, governance, capacity and the shape of a sustainable operating model. It is not presented as finished or independently assured.
A distributor who found all that after signing would have a worse deal than one who read it first. The proposal is for somebody willing to be early in exchange for knowing exactly how early.
Then a distributor holding their own cell keeps operating, and that is the point of the design rather than a reassurance about our finances.
The members hold their own records. Where the arrangement is built out fully they hold their own machines. The substitution time — how long it would take to stand a cell up on something else — is meant to be measured and published rather than asserted, precisely so that this question has a number attached instead of an answer.
If that number is never produced, treat the arrangement as untested on exactly this point.
It is the opposite of the evasion available. The margin is not settled — two internal documents hold different splits against a price list reset in July — and a range published now would be quoted back as a number.
The sixth essay in this series publishes no figures at all, for a reason it states: a placeholder gets quoted back as an estimate, and that had already happened once. A prospective distributor will be given one number in a conversation, not a range in an essay.
What the essay does publish is the shape of the constraint that most affects the arithmetic, which is that compute capacity rather than compute cost is what binds, and that the ceiling moved when a research grant ended at the end of August 2026.
Not capital, not a minimum volume, and not exclusivity in perpetuity.
A cell — one kind of community, in one place you already work and where organisations already know you. Judgement about what would actually help there. Patience with a first drill that should fail something. And answers to two questions that cannot be settled from here: whether a drilled and published right to leave changes a buyer’s decision, and what they would pay for it.
Both are recorded as open in our own register. A distributor is the only party positioned to answer them.
Say so, and we would rather hear it now than build it and find out. The strongest version of that objection has a better claim on space here than another paragraph of agreement, and where a reply corrects something, the correction is made and the fact of it recorded on the page rather than absorbed silently.
If you would rather test it than argue about it, the six essays are instruments and they work on us. The five questions are as askable of this arrangement as of a hyperscaler. If it cannot answer them, that is a finding, and it belongs in public.
What this publication does not claim, and what is outstanding against it in the register.
A question this rests on is parked: What secession premium will members bear?
We publish no pricing and no affordability claim, and we do not assert that members value the secession guarantee.
A question this rests on is parked: Does NZ domicile actually resist foreign compulsion?
We do not claim that a New Zealand entity is beyond foreign compulsion. We claim only that its exposure differs from that of a foreign-parented operator, and that the difference is untested.
A question this rests on is parked: Does accredited contribution in exchange for entitlement create an employment relationship?
We do not claim that accredited contribution in exchange for an entitlement of monetary value is outside New Zealand employment law, and we do not claim that it is inside it. The mechanism is described as proposed, not as available, until counsel has answered.
A question this rests on is open: Does secession capability actually convert a buyer?
A question this rests on is open: Can we author a standard we also hold?
Alongside: the publication · sources and provenance