Essays from the floor.

A reading that predicts is no longer a reading. Prediction is the engine's job — clean, dated, refutable, accountable to the ledger. A reading is what a person makes of the engine's output, and a person, on a closed line, with their own life in the room.

The chart says Saturn return. That is a fact. Whether it will steal your sleep, end a marriage, or finally finish the book that has been finishing you — that is the reading. The engine cannot perform the reading; the reader cannot perform the verification.

This essay defends the wall between them, and explains why every Lunar Luxury sitting carries a substrate-cert: we want the question of did it land separate from the question of was it true.

The closed orbit Ψ(n) = 2ⁿ mod 9 → {1, 2, 4, 8, 7, 5} is older than this house. It is older than us. We did not invent it; we noticed it, then refused to look away from it. This essay is the record of that refusal.

It is tempting to read the orbit as a poetic shorthand for the six daśā voices, the six tones, the six ṛtus. It is none of those things. It is an actual algebraic invariant on the additive group ℤ/9ℤ, with a fixed period of six, and it is what every TRIKĀLA endpoint signs against.

The reason matters. A metaphor cannot be refuted. The orbit can. We publish the proof in three places — Zenodo, the ledger, the substrate-cert header — so that anyone, ever, can break it.

If the framework cannot be refuted, the framework is not a framework. It is a feeling. We publish every miss with the same prominence as every hit — same fonts, same dates, same SHA. The ledger is not marketing; it is the only honest reading of our last six years.

We have been wrong. We will be wrong again. The point of the practice is to be wrong on the record, in public, with the timestamps that came before the event — not after.

The chart will not tell you whether to leave him. It will tell you that the seventh house is locked, that the lord is debilitated in a dusthāna, that Saturn is finishing a long aspect. It will not tell you whether you can survive that.

The engine is honest about its silences. So are we. This essay is a list of what we will not put a number on — death, the death of a child, the act of leaving, the act of staying — and why those silences are part of the discipline, not a failure of it.

Every Sannidhi sitting opens with an oath. The client writes one paragraph — the question, in their own hand, sealed and timestamped. Twelve months later, we open the envelope against the chart and the ledger. Both have to agree.

This is the most uncomfortable structure in the house. It is also the only one that has consistently produced sittings the client trusts a year on. The essay walks through three opened envelopes, anonymised, with the client's permission.

Three sittings, all closed in 2025, all anonymised at the client's request. A succession question. A foundational round. A marriage that needed to end without rupture. We publish the chart features that mattered, the decision that was made, and what happened twelve months later.

None of the three are presented as a victory lap. Two went the way the engine signed. One did not. Both kinds belong in the journal.

The 3-adic numbers are simpler than the real numbers and more honest about what they cannot see. This is an essay for the curious non-mathematician — no Galois, no completion, no apologies. Just a tower of remainders and what each floor sees of the chart.

If you have ever wondered why the substrate is built on ℤ/9ℤ rather than ℝ, the tower is the reason, and this essay is the friendly version.

The chart can show longevity indicators. It cannot show death. The classical texts are clearer about this than most contemporary readers, and we hold the classical line: we do not predict the date of death, the date of a child's death, or the date of a parent's death, ever. Not in writing, not in conversation, not in inference.

This is not a marketing posture. It is a refusal that the practice depends on, and this essay is the long form of why.