| … | … | @@ -489,10 +489,14 @@ boundaries were **spatial**. That exhausts the options in chemistry; it does not |
| 489 | 489 | digital substrate, where individuality is routinely established by neither containment nor |
| 490 | 490 | connectivity — a writer set, a keypair, a signature. So Part C sweeps the *kind* of boundary. |
| 491 | 491 | |
| 492 | | All six use chemistry identical to `ribozyme.c` (φ = hairpin fold, parasites 3× faster, F0=0.85, |
| 493 | | F1=0.999, 4800 units, 200 gens, start φ≈5.99 which is also the neutral expectation of 6.0), with |
| 494 | | **no compartments, no space, no group selection, global Wright-Fisher** — so they are directly |
| 495 | | comparable to a known failure: ribozyme well-mixed goes 5.99 → 5.50. |
| 492 | **Substrates differ across Part C — this paragraph was wrong until a later cold read caught it.** |
| 493 | `coupling.c`, `partition.c` and `signal.c` do use chemistry identical to `ribozyme.c` (φ = hairpin |
| 494 | fold, parasites 3× faster, F0=0.85, F1=0.999, 4800 units, 200 gens, start φ≈5.99 = the neutral |
| 495 | expectation 6.0) with **no compartments, no space, no group selection, global Wright-Fisher** — so |
| 496 | those three are directly comparable to a known failure, ribozyme well-mixed at 5.99 → 5.50. The rest |
| 497 | are **not**: `linkage.c` shares none of that chemistry (8 gene types, distinct-type counting); |
| 498 | `byproduct.c` is explicitly a 600-site **lattice**; `chemoton.c` and `rafcomp.c` are built **on** |
| 499 | compartments, which is their subject. Only the first three are cross-comparable by φ. |
| 496 | 500 | |
| 497 | 501 | | kind of boundary | file | result | |
| 498 | 502 | |---|---|---| |
| … | … | @@ -500,7 +504,7 @@ comparable to a known failure: ribozyme well-mixed goes 5.99 → 5.50. |
| 500 | 504 | | inherited wiring (**persistence**) | `coupling.c` | **falsified by its own control** | |
| 501 | 505 | | **topology** (breakable graph) | `partition.c` | partitioning and gating **anti-correlated** | |
| 502 | 506 | | **verification** (costly signal) | `signal.c` | **first rise above neutral** — but invadable | |
| 503 | | | **concatenation** (linkage) | `linkage.c` | **EMERGES** — the crossing | |
| 507 | | **concatenation** (linkage) | `linkage.c` | **EMERGES** — a crossing on rung 2→3 | |
| 504 | 508 | | **precipitation** (byproduct wall) | `byproduct.c` | free and non-discretionary, and **the wrong half** | |
| 505 | 509 | | **growth-coupled division** | `chemoton.c` | **compartment-level selection emerges** — no decree | |
| 506 | 510 | |
| … | … | @@ -544,7 +548,12 @@ partitioning real, and then: |
| 544 | 548 | | 0.40 | 5.76 | 5.76 | **0.3%** | **+0.00** | |
| 545 | 549 | |
| 546 | 550 | **Partitioning and gating are anti-correlated.** Exactly as the graph fragments, gating's advantage |
| 547 | | collapses to zero. Breaking the network to exclude parasites also cuts catalysts off from each |
| 551 | collapses to zero. ⚠️ **Unverified causally, and flagged by a later cold read as the likeliest |
| 552 | surviving claim to break.** Maintenance cost moves *both* variables in this sweep — partitioning was |
| 553 | never varied independently of it. The gain-vs-control comparison does cancel cost effects (the |
| 554 | control carries the same cost), which puts this on better footing than Test B's T4; but T4 and |
| 555 | chemoton's C3 were both exactly this shape and both fell when checked at matched conditions. This |
| 556 | one has not had that check. Breaking the network to exclude parasites also cuts catalysts off from each |
| 548 | 557 | other; selective edge-deletion cannot keep the in-group connected while severing the out-group |
| 549 | 558 | without already knowing which is which — the thing that was supposed to emerge. And k never evolves |
| 550 | 559 | to 1, settling at ≈0.32–0.36 at every cost: selection feels the trade-off directly. |
| … | … | @@ -589,7 +598,7 @@ so defection is not an available move. |
| 589 | 598 | Every unit starts at **length 1**. Fusion/fission are mutations. Payoff needs 4 distinct of 8 types |
| 590 | 599 | in one unit. **Longer units replicate strictly slower.** |
| 591 | 600 | |
| 592 | | | length cost | A mean len | A frac≥4 | B (pool-draw control) | B frac≥4 | |
| 601 | | length cost (5 of 7 swept shown) | A mean len | A frac≥4 | B (pool-draw control) | B frac≥4 | |
| 593 | 602 | |---|---|---|---|---| |
| 594 | 603 | | 0.00 | 7.19 | 0.952 | 4.52 | 0.477 | |
| 595 | 604 | | 0.02 | 6.77 | 0.942 | **2.17** | 0.116 | |