- Magnetic dipole / iron inclusions as a polarized millimetre systematic
- AME as extra unpolarized intensity below ~100 GHz
- Explicit H0–H3 hypothesis ladder (mask → layers → moments → harmonic hybrid)
- In synthesis: CMB-S4 cancelled 9 July 2025; keep the residual target anyway
After all three modes
Where they overlapped, and where they left the others behind.
Independent essays already shared the physical skeleton. Drive copied outlines. Critic and optimist is the mode that changed the physics: it killed false lemmas and imported each other’s observational handles. A residue of pet systematics never migrated.
Overlap
- Polarized thermal dust is the leading high-frequency B-mode foreground. Aspherical grains aligned with B emit with E perpendicular to B_POS.
- “Spectral-index decorrelation” is frequency decorrelation of the polarized maps: Q,U at one frequency are not a real scalar times Q,U at another.
- The line-of-sight mechanism needs two conditions (Tassis & Pavlidou 2015): different layer SEDs and misaligned plane-of-sky fields.
- Planck Int. L is not a robust all-sky detection; Pelgrims et al. 2021 saw the effect on HI-selected multi-cloud, misaligned sightlines.
- 3D “dust maps” in hand are mostly stellar extinction, not millimetre emission tomography.
- The three ingredients are priors on layer weights, SEDs, and polarization phases — not a frozen template. Residual moments still eat unresolved averaging.
Unique leftovers
- Fanciullo et al. 2017: size alone cannot explain emission-to-extinction polarization
- Full j_ν^I vs j_ν^P integrals over grain size and alignment
- Sharpest “one real mixing-matrix column” definition of a single MBB
- Worked residual arithmetic: R ≈ 0.999 can still be δr ~ 10⁻³
- Tomography method zoo (BISP-1, TRIShUL, RoboPol 4 deg²) and “dedusting” as PASIPHAE’s term
- With known distinct angles, two layers are invertible at one frequency (condition number ~ 1/sin 2Δψ)
- Mehandiratta et al. 2026: R_{P/p} varies cloud to cloud
What each mode was good for
Best for extra systematics.
Overlap was already large. Divergences were emphasis and pet contaminants, not rival mechanisms.
Shares an outline. Does not correct errors.
The driven essay blends the driver’s structure with the driven model’s defaults. Grok’s false multi-layer theorem and GPT’s Sponseller/McBride mix-up survived drive.
The mode that changed the physics.
Critics killed false lemmas. Optimists imported spin-moments, the Local Bubble, staged risk maps. Primaries refused to invent Δr forecasts.