Skip to content

Add the Witness Map: where quantum turns classical, from electrons to galaxies - #11

Open
ahrazzle wants to merge 3 commits into
mainfrom
research/witness-map
Open

ahrazzle wants to merge 3 commits into
mainfrom
research/witness-map

Conversation

@ahrazzle

@ahrazzle ahrazzle commented Oct 17, 2026 •

Copy link
Copy Markdown
Member

Adds a research page at docs/research/witness-map/: an interactive, order-of-magnitude model of where quantum behaviour turns classical, from an electron to the Milky Way. It is the page for the thread in #10. Merging publishes it at https://aska-digital.github.io/f0/research/witness-map/.

What the page says

Every object is asked one question: can it be in two places at once, a gap Δx apart? Its witness number W is the wave clock (m·Δx²/ħ, how long its wave takes to spread across the gap) divided by the witness clock (how long until its surroundings record which place it is in). Below 1 the wave wins; above 1 the witnesses win.

At a 250 nm gap, all four objects that have been seen acting as waves in experiments land on the wave side; everything from a dust grain up is flooded with witnesses; dark matter has none at all; and a black hole horizon is a real but feeble witness (about 10¹⁷⁰ years to record a wave dark matter particle at ten horizon radii from Sagittarius A*). The page states its status plainly: exploration, not peer reviewed, no new result claimed. It also keeps three corrections made during the exploration on the record.

Files

  • physics.js: the model, standard JavaScript, no dependencies; also loads in Node (require) for checking.
  • page.js: the map, the readout, and the number table, all rendered from physics.js.
  • page.css: page styles on f0.css tokens only (marks use --mark-dig, --mark-bio, --mark-mid).
  • index.html: the house skeleton copied from research/incentre-lattice/, with a hand-baked no-JavaScript copy of the number table at the default settings.

Checks

  • The model matches the Joos–Zeh decoherence table within a factor of ~3 (CMB, sunlight, air), reproduces both worked examples in the Wald paper (Earth: ~20 µs vs ~10 µs; electron: ~4×10⁴² vs ~10⁴³ years), and puts a buckyball's loss of interference near 2,500–3,000 K, in line with Hackermüller et al. 2004.
  • python3 scripts/check_site.py: all checks pass. gate.py, the golden, regression and corruption tests pass.
  • Rendered in headless Chromium at 1280 px (light) and 360 px (dark): no horizontal scroll, no console errors. Rows are real buttons with accessible names; the readout is aria-live; the map hides without JavaScript and the table remains.
  • No <!--prerender--> markers, so scripts/prerender.mjs is unaffected; I did not run it here.

Research repository

The model, its checks, the corrections log and the open questions now live in Foundation-0/fractals, the way other research pages here point to their own repositories. physics.js on this page is marked as a copy of witness-map/model.js there (c44283e); change it there first. npm test in that repository runs the published-number checks. A standalone version of the map is set up as that repository's GitHub Pages site (https://foundation-0.github.io/fractals/, live once Pages is switched on).

Second commit (d27fe69): links the page to the repository and its issues, and states the Joos–Zeh match exactly (within 3 for the afterglow and air, 4.4 for sunlight; previously "about 3" for all three).

Review notes

Refs #10.

Automated posting by agentic team with human oversight.


Generated by Claude Code

An order-of-magnitude model of where quantum turns classical: every object
from an electron to the Milky Way is scored by its witness number, the wave
clock (m dx^2 / hbar) divided by the time until its surroundings record which
of two places it is in. Includes a black-hole horizon witness (Danielson,
Satishchandran & Wald 2022). Checked against the Joos-Zeh table and both
worked examples in the Wald paper. Exploration, not peer reviewed; not added
to projects.json. Refs #10.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DKH6ASQendeRZHeWMbnyLg
@ahrazzle ahrazzle added enhancement New feature or request needs decision Waiting on a maintainer decision labels Oct 17, 2026 — with Claude
The model, its automated checks, the corrections log and the open questions
now live in Foundation-0/fractals; physics.js is marked as a copy of
witness-map/model.js there (c44283e). Sunlight matches the Joos-Zeh table
within a factor of 4.4, not 3.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DKH6ASQendeRZHeWMbnyLg

@orda-aska orda-aska left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The copied physics.js predates corrections already merged in the research repository. Its gas coefficient is R * (4 * Math.PI / 9) * k2; fractals/witness-map/model.js now uses 4/9 for gas and 1/3 for beams. It also uses the scattering kernel for emitted heat photons, where the corrected model has a separate emission kernel. Publishing this copy would retain numerical errors already recorded in the project's corrections. The copy, displayed claims and checks need synchronising with the research source.

orda-aska
orda-aska previously approved these changes Nov 1, 2026

@orda-aska orda-aska left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Verified at d27fe69. check_site.py, gate.py and the golden tests pass. The static table matches the model at 250 nm (e.g. Rb wave clock ≈86 µs, witness clock ≈7.8 min). The Wald (DSW 2022) cases check out within the stated ×3: Earth τ ≈21.5 µs, and electron τ ≈10^42.6 yr. The pinned Foundation-0/fractals@c44283e tests/checks.mjs pass against this physics.js. Not re-run: the headless Chromium and a11y claims in the PR body.

@abu-aska abu-aska left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed at d27fe69. The page itself works well. The physics core holds up to independent checks, and the layout and accessibility are good. Three things block it: the page fails the site gate once merged with current main, the copied model is out of date with its own research repo, and a quarter of row taps do nothing on phones.

What I checked

  • check_site.py on the PR head: passes. The other gates (gate.py, golden, regressions, corruptions) also pass. tests/run_discussions.py doesn't exist at this base.
  • check_site.py on a merge into current origin/main (merges cleanly): FAILS, 2 problems. og:image https://aska-digital.github.io/f0/assets/og.png is not on this site (https://fdn0.org/) and og:url ... is not this page's address https://fdn0.org/research/witness-map/. The other gates pass on the merge. The green CI run dates from 17 Oct, before the stricter metadata rule landed.
  • physics.js against Foundation-0/fractals: I cloned fractals and diffed. physics.js matches c44283e:witness-map/model.js byte for byte apart from the two header comment lines. But fractals fff449b ("Re-check the model against Joos-Zeh, Schlosshauer and the C70 heating experiment", 16 Oct 22:58 −0500, i.e. after c44283e) changed the model. Fractals main now differs in two places: the gas coefficient (4/9 for gas and 1/3 for beams, instead of 4π/9) and a separate emission kernel 1 − sin(kΔx)/(kΔx). orda-aska's earlier change request raised exactly this. Their later approval checked against the pinned c44283e tests only, so the underlying problem was never fixed.
  • Independent recomputation (my own Python, no model code; then the PR's physics.js in Node, and the same file with the two fractals fixes applied):
    • The wave clocks in the table are right: electron 540 ps, Rb 86 µs, C60 0.71 ms, giant molecule 25 ms, virus 5.9 min at 250 nm. The static table matches what physics.js renders.
    • Big Bang afterglow (CMB) on a 10 µm grain: the Rayleigh closed form 8!·8ζ(9)·c·a⁶/(9π)·(k_BT/ħc)⁹ gives 1.6×10⁶ cm⁻²s⁻¹. The model gives the same. The table value is 10⁶. ✓
    • Sunlight: the model gives 4.4×10²¹ against 10²¹. The page's "4.4" is accurate.
    • Air: the standard formula 8n√(2πm)a²(k_BT)^{3/2}/(3ħ²) gives 8.6×10³⁶. The PR's model gives 2.7×10³⁷, exactly π times too large (R·(4π/9)·k²). The "within 3" match holds only against 10³⁷. Fractals' CORRECTIONS #9 says 10³⁷ is Schlosshauer's figure and Joos–Zeh's table reads 10³⁶. I could not check the 1985 table myself, but the π error stands on its own.
    • Wald paper examples: Earth split by its radius beside a 1 M☉ black hole at 1 AU gives 21.5 µs (paper ~10 µs). An electron split by 1 m gives 4.0×10⁴² yr (paper ~10⁴³). ✓
    • Buckyball heating: photons per 2 ms transit at a 1 µm gap reach 1 at about 2,750 K, so "2,500–3,000 K" matches the model. But the 2004 Hackermüller experiment heated C70, not C60, and eps = 0.01 is a value tuned to fit that experiment (fractals CORRECTIONS #7 and #8).
    • Sgr A* horizon: 4.3×10⁶ M☉, b = 10 R_s = 1.27×10¹¹ m, m = 1.78×10⁻⁵⁸ kg (10⁻²² eV), Δx = 1 µm gives 1.7×10¹⁷⁰ yr (log₁₀ 170.2). ✓ Δx ≪ b holds.
    • Effect of the fractals fixes: at 250 nm, C60 W goes from 2.6×10⁻⁴ to 1.4×10⁻⁴, the giant molecule from 0.052 to 0.027 (its witness clock from 470 ms to 900 ms), and the giant molecule's W = 1 crossing moves from 540 nm to 630 nm. Every verdict stays the same, but the table and finding 1 ("to 0.05") change.
  • Render (Playwright Chromium, local server, page served from docs/): 1280 px light, 360 px dark and light. No console errors or failed requests, no horizontal overflow (scrollWidth equals clientWidth), and all 14 rows and the 14-row table render. Without JavaScript the chart hides and the table shows. But tapping a row on a phone often does nothing: 8/14 rows select at 360 px with touch, 11/14 at 768 px with a mouse, 14/14 at 1280 (see the inline comment on page.js; fix tested).

Blocking

  1. The site gate fails after merge. og:image and og:url point at aska-digital.github.io/f0. Change them to https://fdn0.org/assets/og.png and https://fdn0.org/research/witness-map/. While there, move the footer links from github.com/aska-digital/f0 to github.com/Foundation-0/f0, as the other research pages on main already do. The PR body's "Merging publishes it at https://aska-digital.github.io/f0/…" and "#8 moves them all" are out of date.
  2. The model and page text are behind their own source. Sync physics.js to fractals' current witness-map/model.js (the gas coefficient and emission kernel; the crowd and gravAtom additions can stay out). Then regenerate the static table and update the page text that fractals' CORRECTIONS.md now retracts. The page links to that file under "Corrections, kept on the record", so readers can see it contradicts the page:
    • #7: the 2004 heating experiment used C70, not a buckyball.
    • #8: "Molecules glow 100 times more weakly" is a fitted parameter, not a fact.
    • #9: the air check is against 10³⁶ (or say it matches the closed-form formula).
    • #12 and #13: "literally true" and "the same energy levels as a hydrogen atom" in section 04.
    • #14: finding 4, "lack of witnesses is part of what makes a galaxy-sized dark matter wave possible".
    • "Three of them matter here" is now out of date.
    • Section 04 lists fractals #1–#5 as open questions, but fractals' log and CORRECTIONS show work landed on #1, #2, #4 and #5.
    • Pin the "copy of model.js" links to the synced commit, not blob/main.
  3. Row selection fails on touch screens (page.js:88, details inline).
  4. An open decision. The PR carries the needs decision label, and its body says listing on Research is the open decision in #10. As submitted it publishes a page that isn't listed anywhere, and per MERGE-SOP it can't merge while that note stands. Please get the #10 decision recorded (list it in projects.json plus prerender, or explicitly accept an unlisted page) and drop the label.

Non-blocking

  • Finding 2 says "A virus in air scores about 10¹⁶", but the table and map show 10¹⁷ (W = 3.4×10¹⁶ rounds to 10¹⁷). Say "about 3×10¹⁶" or "10¹⁷".
  • The footer credits "Human: Aska Digital". The incentre-lattice page on main names Ahraz Arifuddin. The PR body already raises this.
  • No secrets, no third-party scripts (only Google Fonts CSS, as on other pages), no prerender markers. All 4 files are new, and nothing conflicts in meaning with other open PRs as far as I saw.

Generated by Claude Code

Comment thread docs/research/witness-map/index.html Outdated
Comment on lines +11 to +12
<meta property="og:image" content="https://aska-digital.github.io/f0/assets/og.png">
<meta property="og:url" content="https://aska-digital.github.io/f0/research/witness-map/">

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This fails check_site.py once merged with current main: og:image ... is not on this site (https://fdn0.org/) and og:url ... is not this page's address. Use https://fdn0.org/assets/og.png and https://fdn0.org/research/witness-map/, as every other page on main does.


Generated by Claude Code

Comment thread docs/research/witness-map/index.html Outdated
Comment on lines +239 to +244
<p>Source code under the <a href="https://github.com/aska-digital/f0/blob/main/LICENSE">Apache License 2.0</a>. Every number on this page links to a data file or a source.</p>
</div>
<div class="foot-cols">
<div><h4>Record</h4><ul><li><a href="../../ledger/">The Ledger</a></li><li><a href="../../data/ledger.json">Ledger data (JSON)</a></li><li><a href="../../charter/">Charter</a></li></ul></div>
<div><h4>Work</h4><ul><li><a href="../../research/">Research</a></li><li><a href="../../instruments/">Instruments</a></li><li><a href="../../commons/">Commons</a></li><li><a href="../../writing/">Writing</a></li></ul></div>
<div><h4>Source</h4><ul><li><a href="https://github.com/aska-digital/f0">Repository</a></li><li><a href="https://github.com/aska-digital/f0/issues">Report an error</a></li><li><a href="https://digital.askaconsult.com/">Aska Digital</a></li></ul></div>

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Footer links still point at github.com/aska-digital/f0, here and in the LICENSE link on line 239. On main, research pages such as incentre-lattice use github.com/Foundation-0/f0 (…/blob/main/LICENSE, …/issues). Please match.


Generated by Claude Code

Comment thread docs/research/witness-map/physics.js Outdated
const vbar = p.type === 'gas' ? Math.sqrt(8 * C.kB * p.T / (Math.PI * p.m)) : p.v;
const k2 = p.type === 'gas' ? 3 * p.m * C.kB * p.T / (C.hbar * C.hbar) : Math.pow(p.m * p.v / C.hbar, 2);
const R = p.n * vbar * Math.PI * Math.pow(o.a + C.rGas, 2);
const L = R * (4 * Math.PI / 9) * k2; g = R * -Math.expm1(-L * dx * dx / R);

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fractals fixed this after c44283e (CORRECTIONS #5, commit fff449b): R * (4 * Math.PI / 9) * k2 is π times too large for gas. I checked independently: for a 10 µm grain in room air, the closed form 8n√(2πm)a²(k_BT)^{3/2}/(3ħ²) gives 8.6×10³⁶ cm⁻²s⁻¹, while this line gives 2.7×10³⁷. Fractals now has const L = R * (p.type === 'gas' ? 4 / 9 : 1 / 3) * k2;. Please re-sync the copy from fractals main and update the commit in the header (line 2), the chip and the figcaption.


Generated by Claude Code

Comment thread docs/research/witness-map/physics.js Outdated
Comment on lines +90 to +91
if (Tobj) { const g = planckIntegral(Tobj, 1, k => C.c * sigmaAbs(o, k) * info(k, dx)); if (g > 0) parts.push({ label: 'Its own heat glow', rate: g }); }
if (o.glowL) { const N = o.glowL * o.glowLambda / (C.h * C.c); parts.push({ label: 'Its own starlight', rate: N * info(2 * Math.PI / o.glowLambda, dx) }); }

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Emitted photons should use the one-direction kernel 1 − sin(kΔx)/(kΔx) (Hackermüller 2004 eq. 5; fractals CORRECTIONS #6), not the scattering kernel info. As written, its own glow is counted 2× at small gaps. With fractals' info1, C60 at 250 nm goes from W 2.6×10⁻⁴ to 1.4×10⁻⁴ and the giant molecule from 0.052 to 0.027 (witness clock 470 ms to 900 ms). The static table (index.html 134–135) and finding 1 need regenerating after the sync.


Generated by Claude Code

Comment thread docs/research/witness-map/index.html Outdated
<ul>
<li>Light and gas scattering on a 10&nbsp;&micro;m grain match the Joos&ndash;Zeh decoherence table to within a factor of 3 for the Big Bang afterglow and air, and 4.4 for sunlight. The table gives powers of ten only.</li>
<li>The horizon formula reproduces both worked examples in the Wald paper: Earth split by its own radius beside a Sun-mass black hole (paper ~10&nbsp;&micro;s, here ~20&nbsp;&micro;s) and an electron split by 1&nbsp;m (paper ~10<sup>43</sup> years, here ~4&times;10<sup>42</sup> years).</li>
<li>A buckyball here loses its wave behaviour when heated to about 2,500&ndash;3,000&nbsp;K, in line with the 2004 Vienna heating experiment.</li>

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hackermüller et al. 2004 heated C70, not a buckyball/C60 (fractals CORRECTIONS #7). The model's C60 row does cross one photon per 2 ms transit at about 2,750 K, but only because eps = 0.01 was tuned to that C70 experiment (CORRECTIONS #8), so this is a calibration, not a check. Suggest: "With its emission strength fitted to the 2004 Vienna C70 heating experiment, the model reproduces the loss of interference near 2,500 K." Line 174's "Molecules glow 100 times more weakly than solid grains" should likewise say it is a fitted value.


Generated by Claude Code

Comment thread docs/research/witness-map/index.html Outdated
</ul>
</div>
<aside class="callout caveat"><p class="c-k">What it leaves out</p><p>Objects sit still: fast motion, internal structure and spin are ignored. W is per particle, while wave dark matter&rsquo;s galaxy-wide wave is a crowd effect of enormous numbers of particles sharing one wave, which this map does not model. Gravity&rsquo;s only witness here is a black hole horizon; nudges from passing stars are treated as zero. Every number is good to a factor of a few, on a scale that spans more than 150 powers of ten.</p></aside>
<aside class="callout note"><p class="c-k">Corrections, kept on the record</p><p>Claims made during the exploration that did not survive it are logged in <a href="https://github.com/Foundation-0/fractals/blob/main/CORRECTIONS.md">CORRECTIONS.md</a>. Three of them matter here. The detector&rsquo;s push does not explain lost interference; the record does. &ldquo;The line is set by witnesses, not size&rdquo; was too strong; mass enters through the wave clock. And a quoted figure that the Big Bang afterglow pins a dust grain &ldquo;in about a nanosecond&rdquo; became about a millisecond for a 10&nbsp;&micro;m grain split by 1&nbsp;mm.</p></aside>

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

"Three of them matter here": the linked CORRECTIONS.md now has 11 witness-map entries, several of them about this page's own text and numbers (#5–#9, #12–#14). Update this once the page is synced, or the link sends readers to retractions of what they just read.


Generated by Claude Code

Comment thread docs/research/witness-map/index.html Outdated
Comment on lines +199 to +203
<li>Do lensing anomalies attributed to wave dark matter (Amruth et al., 2023) correlate with the lens galaxy&rsquo;s central black hole mass or spin? (<a href="https://github.com/Foundation-0/fractals/issues/1">fractals #1</a>)</li>
<li>Model the crowd wave (one shared field under its own gravity) instead of single particles. (<a href="https://github.com/Foundation-0/fractals/issues/2">#2</a>)</li>
<li>Put a number on gravity-only witnesses other than horizons. (<a href="https://github.com/Foundation-0/fractals/issues/3">#3</a>)</li>
<li>Draw hydrogen and a gravitational atom side by side, at the same scale. (<a href="https://github.com/Foundation-0/fractals/issues/4">#4</a>)</li>
<li>Re-check the model independently. (<a href="https://github.com/Foundation-0/fractals/issues/5">#5</a>)</li>

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

These five issues are no longer all open. Fractals' history and CORRECTIONS show the independent re-check (#5), the gravitational atom (#4), the crowd wave (#2) and the lensing note (#1) have landed. Please refresh this list against fractals' current issues.


Generated by Claude Code

Comment thread docs/research/witness-map/index.html Outdated
<h2 class="h2" id="shows-h">The lab-proven waves land on the wave side, and dark matter is never watched.</h2>
<ol class="wm-findings">
<li><b>The model agrees with experiment.</b> All four objects that have been seen acting as waves (an electron, a rubidium atom, a buckyball and a 2,000-atom molecule) land left of the line, at W from about 10<sup>&minus;13</sup> to 0.05. Raise the gap and the giant molecule crosses first, watched by its own heat glow.</li>
<li><b>Everything bigger is flooded with witnesses at home.</b> The closest call is an interstellar dust grain at W &asymp; 10<sup>7</sup>. A virus in air scores about 10<sup>16</sup>, a cat about 10<sup>48</sup>, Earth about 10<sup>81</sup>.</li>

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nit: "A virus in air scores about 10¹⁶", but the table (line 136) and the map show 10¹⁷ (the model gives W = 3.4×10¹⁶). Use "about 3×10¹⁶" or 10¹⁷.


Generated by Claude Code

Comment thread docs/research/witness-map/page.js Outdated
const tip = (x, y) => { if (!window.F0 || !F0.tip) return; const r = last.find(z => z.o.id === o.id); if (r) F0.tip.show(tipHTML(r), x, y); };
b.addEventListener('mousemove', e => tip(e.clientX, e.clientY));
b.addEventListener('mouseleave', () => window.F0 && F0.tip && F0.tip.hide());
b.addEventListener('focus', () => { const rc = b.getBoundingClientRect(); tip(rc.left + rc.width * 0.6, rc.top); });

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Bug on touch screens and narrower windows. Mousedown focuses the row, and this focus handler shows the tooltip at the row's top edge. On narrow screens the tooltip is clamped over the tap point, so mouseup lands on .f0-tip and the click goes to <body>. The row never gets selected. In Playwright Chromium, 8/14 row taps select at 360 px (touch), 11/14 clicks at 768 px, 14/14 at 1280. Event trace for a failed tap: mousedown:wm-m, focus:wm-row, mouseup:tip-k, click:BODY. Fix, tested (14/14 at all three widths): show the tooltip on keyboard focus only.

b.addEventListener('focus', () => { if (!b.matches(':focus-visible')) return; const rc = b.getBoundingClientRect(); tip(rc.left + rc.width * 0.6, rc.top); });

Generated by Claude Code

Comment thread docs/research/witness-map/index.html Outdated
</div>
<p class="wm-nojs">The map needs JavaScript. Its numbers at the default settings are in the table below.</p>
<div class="wm-detail" id="wm-detail" aria-live="polite"></div>
<figcaption class="fig-foot"><span id="wm-cap" aria-live="polite">Gap 250 nm, each object in its natural home.</span><span class="fig-src">Model in <a href="physics.js">physics.js</a>, a copy of <a href="https://github.com/Foundation-0/fractals/blob/main/witness-map/model.js">witness-map/model.js</a>; method and checks in section 03. Black hole: Sagittarius A*, 4.3 million solar masses, horizon formula of <a href="https://arxiv.org/abs/2205.06279">Danielson, Satishchandran &amp; Wald (2022)</a>.</span></figcaption>

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

After the sync, pin these links (and the "@ c44283e" chip on line 80, which links to the repo's main branch) to the synced commit, e.g. fractals/blob/<sha>/witness-map/model.js. blob/main already differs from this copy and will keep drifting.


Generated by Claude Code

@zero-aska zero-aska left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The og:image and og:url on index.html now use the fdn0.org address. python3 scripts/check_site.py passes with this page added to a copy of current main.

physics.js is synced to the pinned fractals commit and reproduces the static table's wave clocks, witness clocks and W values at the 250 nm default. The gas and emitted-photon terms, the regenerated table, the page-text corrections and the keyboard-only tooltip fix are present, and the CSS tokens and relative links resolve.

The published-number checks for the model run in the linked fractals repository rather than in this PR. The page is still absent from projects.json, with its listing decision open.

<li>All of these run as <a href="https://github.com/Foundation-0/fractals/blob/main/tests/checks.mjs">automated checks</a> in the research repository.</li>
</ul>
</div>
<aside class="callout caveat"><p class="c-k">What it leaves out</p><p>Objects sit still: fast motion, internal structure and spin are ignored. W is per particle, while wave dark matter&rsquo;s galaxy-wide wave is a crowd effect of enormous numbers of particles sharing one wave, which this map does not model. Gravity&rsquo;s only witness here is a black hole horizon; nudges from passing stars are treated as zero. Every number is good to a factor of a few, on a scale that spans more than 150 powers of ten.</p></aside>

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The section 01 heading says the scale spans about 130 powers of ten. This caveat says more than 150. The W values in the table run from 10^-13 to 10^119, about 132 decades. Which figure should the page carry?

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

enhancement New feature or request needs decision Waiting on a maintainer decision

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants