Is the shape really there?
A person who has found a real pattern and a person who is seeing things report the same experience. It gets clearer with every example. Examples are easy to add. Adding them feels like evidence.
Ten claims below. Decide about each before I tell you anything.
A literary canon / A Bayesian prior
Both are starting positions that determine how much work new evidence has to do before it will be taken seriously, and neither is itself produced by the evidence it governs.
Some of these I defend. One I gave up. Some I wrote myself, to see whether they would pass.
RecordedWhich claims you marked found or imposed, and how many of the three controls got through — the single most useful number this project can collect. Plus a random identifier and how long you took. No name, no account, no IP address, no analytics. Where a box invites you to write, what you write is stored verbatim — that field is optional and is labelled at the point of entry. All of it, in detail.
What this does not settle.
Catching the controls does not show the real connections are real. It shows the controls were catchable, and I wrote them, so I had every advantage in making them catchable without noticing I had.
The honest position is narrower than the page probably makes it feel: two of the ten survive scrutiny with a specific stated structure, most survive only as descriptions of a shared role, one did not survive at all, and three were never claims. That is a thinner result than “the same shape appears everywhere”.
It is also the result, so it is what the site says.
So what do you do with the next one?
You have just tested somebody else’s connections. Yours are harder.
These ten came labelled, judged, and with the fabrications disclosed at the end. A pattern you find yourself arrives with none of that — and it arrives feeling exactly like the ones above felt before the reveal.
The Canon Test →