3
[2026-08-22] · src Naidu et al. (Mirage or Miracle survey); Nature 656, 329 (2026)
JWST has been finding hundreds of compact, extremely red sources in the early universe - the "Little Red Dots" - whose spectra look starlike but whose implied stellar masses would break galaxy-formation models if taken at face value. This paper reports MoM-BH*-1, at redshift z = 7.7569 (660 million years after the Big Bang), with a Balmer break of strength 7.7 - against a theoretical ceiling near 3 for any ordinary dust-free stellar population - which no normal star or star cluster can produce. The authors model it instead as roughly a million-solar-mass black hole (~106.3 M_sun) accreting inside a dense hydrogen envelope spanning 10-100 AU, wide enough to reprocess the accretion light into a stellar-looking spectrum while emitting on the order of 100 billion times a single star's luminosity.
If the model is right, it gives the Little Red Dot population a physical identity - an accreting black hole wearing a gas cocoon - rather than a statistical anomaly to be explained away with exotic stellar populations. That would mean supermassive black holes were assembling far earlier and faster than standard seed-and-growth models allow, which is the harder problem underneath the red dots in the first place. One object with an unusually clean spectral signature is evidence, not proof; the test is whether the same Balmer-break diagnostic picks out the same population in the rest of the Little Red Dot sample.
[2026-08-22] · src CMS Collaboration; Phys. Rev. Lett. (2026), arXiv:2510.02580
Heavy-ion collisions at the LHC normally use lead nuclei to make quark-gluon plasma; 2025's dedicated oxygen-oxygen and neon-neon run let CMS ask whether the same collective-flow machinery, if measured precisely enough, can read out the internal geometry of the colliding nuclei themselves. Using 7 nb-1 of O+O data and 0.8 nb-1 of Ne+Ne data at sqrt(s_NN) = 5.36 TeV, CMS measured elliptic (v2) and triangular (v3) flow harmonics from two- and four-particle correlations spanning nearly five units of pseudorapidity in both systems. The centrality dependence of v2, and specifically the ratio of v2 between Ne+Ne and O+O, tracks the quadrupole deformation predicted by ab initio nuclear-structure calculations - oxygen-16's tetrahedral alpha-cluster arrangement versus neon-20's elongated "bowling pin" shape - well enough that hydrodynamic models built on those structure inputs reproduce the pattern.
Nuclear structure at this scale has mostly been inferred from low-energy spectroscopy; this is a second, independent read on the same question using collective flow at TeV energies, on nuclei small enough that individual nucleon clustering actually shows up in the signal. It extends flow-based structure probes - already established for heavier species in earlier 2025 running - into the light-nucleus regime with enough statistics to call it an observation rather than a hint, which is useful confirmation but not a new phenomenon.
[2026-08-22] · src Iyengar, McHugh, Tripathi, Meunier et al. (Rice University); Advanced Materials (2026)
Flexoelectricity - charge separation induced by strain gradients rather than uniform strain - is normally a weak effect that only shows up in bulk crystals under substantial bending. This group measured it instead at sub-nanometre curvature in wrinkled monolayer graphene, and found charge separation 105 to 106 times stronger than in the flexoelectric systems it has previously been measured in. The sharpest wrinkles act as narrow regions of shifted electrical potential - "speed bumps" that pass measurable current under about 1 V of bias - and the effect tracks the sharpness of the curvature far more than its height: a tall, gently rounded wrinkle is less active than a short one with curvature concentrated into a narrow kink. The authors attribute the size of the effect to a quantum orbital contribution beyond the classical flexoelectric picture.
If the magnitude holds up under independent measurement, this is a route to controlling electrical response in atomically thin materials by shaping them mechanically, rather than by doping or stacking additional layers - useful anywhere a device needs a locally tunable potential without added chemistry. It's also a case where an established effect, measured at a new length scale, turned out to need new physics to explain its size rather than just a bigger number.
1
[2026-08-22] · src —
Nothing has changed since the comparison in
data/feed/2026-08-20/learning.org: Voxman, Selected Studies is still the
only one of the four candidates that gives alto the key-by-key
technique-plus-etude shape trumpet already has under :METHOD: in
model/briefs/music.org. Ferling has no technical spine under it, Klosé has
nothing to phrase, and Rubank's own candidate note already calls it "less
musical" — restating the same gaps doesn't change which one clears them.
What this run cannot do is act on that comparison. Setting :METHOD: under
Alto sax in model/briefs/music.org is a write to model/, and model/ has
exactly one writer — you, never an agent, per CLAUDE.md. There is nothing
left here to research or re-decide; the only path from "recommended" to
"chosen" is a steer entry naming it, which comes back as a pull request
against model/briefs/music.org for you to merge, the same way trumpet's
Arban line got there. Until one lands, alto keeps drawing the generic
fundamentals assignment the brief's placeholder issues on purpose.
2
[2026-08-22] · src Arban, Complete Conservatory Method — Ornaments
Four sessions in and trumpet has drilled evenness, intervals and time feel, but nothing that asks the fingers and the air to trade off quickly against each other — which is what Ornaments is for. Today moves onto the trill studies, a section that hasn't been touched yet and that stays clear of both the tonguing-speed and high-register work the brief rules out pairing with anything else.
5 minutes: slow lip trills alternating G4–A4 (the 4th and 5th partials) at mp, quarter=54, one full breath per pair of trills. The point is getting the trill's alternation felt in the air and lips before asking the valves to do a faster version of the same thing.
Arban's Ornaments section: the half-step trill studies, using the printed alternate-fingering combinations rather than lip alone, at quarter=92. Success criterion: the trill starts audibly on the printed note — not the upper auxiliary — and stays evenly spaced for four full measures without rushing into a blur or the core tone thinning out under the extra valve motion.
No tonguing-speed or high-register material today — trill work at this tempo and register doesn't call for either, so the two don't collide.
Art of Phrasing, Melody No. 13. Play it once for notes, then again with the day's ornaments in mind: find anywhere the melody itself asks for a turn or a grace note and give it the same clean, evenly-spaced treatment as the warmup, rather than smearing it as a flourish.
Stop the moment a trill blurs into a smear instead of speaking as two distinct pitches, or the tone starts thinning from over-pressing the valve combination to keep up — that's the fingers and the embouchure fighting each other rather than trading off cleanly, and reps past that point train the fight instead of the trade-off.
[2026-08-22] · src —
hearWynton Marsalis — Hummel: Trumpet Concerto in E-flat Major - III. Rondo (Haydn, L. Mozart & Hummel: Trumpet Concertos) · Tidal · Spotify
The Rondo is built almost entirely out of the ornaments today's technical target isolates — trills and turns aren't decoration in it, they're the main material the movement runs on. Listen for how evenly the trills stay spaced even as the tempo pushes forward, which is exactly what today's stop condition is there to protect.