Stardate: deep dream day one finished, with the final numbers

RELEASE-INVENTORY.json

@@ -1668,39 +1668,39 @@
       "sha256": "d01f229fb24e1512d54f961080c0388598a37f734a214539f88e6d93c1b62029"
     },
     {
-      "bytes": 8482,
+      "bytes": 9494,
       "path": "site/assets/stardate/2026-268-deep-dream-day-one.html",
-      "sha256": "f202b778954af39a4349da60f6df2bac6cb6b00ea2ff3ae9eced096f813cccf7"
+      "sha256": "8fff60c9222f8c940ff7a1e9aee2128462835303fdfc71e50e4b3cf885a6644e"
     },
     {
       "bytes": 8257,
       "path": "site/assets/stardate/2026-268-grok-hooks-ab.html",
       "sha256": "815a2875a1e2f60949794567d4939fa7d0c6347c0cb046df58cb37a576911bad"
     },
     {
-      "bytes": 7869,
+      "bytes": 7889,
       "path": "site/assets/stardate/2026-268-incoming-filter-on.html",
-      "sha256": "a38df021e28c871e04bcae2c67effaa54ceb2625d64d48ba8885c7088acdecec"
+      "sha256": "80904aa1107e2a79bdf3a0b3944923247d3da61224efc41396d394374ce124eb"
     },
     {
       "bytes": 7710,
       "path": "site/assets/stardate/2026-268-judge-kit-reviews.html",
       "sha256": "b271aabcdbc089898b8f4984dc5f9f44f40e94927bd3c7ee8e73a667749b7e92"
     },
     {
-      "bytes": 7932,
+      "bytes": 8048,
       "path": "site/assets/stardate/2026-268-marker-chains.html",
-      "sha256": "cfd28bcaca87d62731574c113dc27264c3a9e365ea64415584a2f5b5af40ce24"
+      "sha256": "eba4bc5465853ea9c86d88702d3a2172ba78c91b298f7139f6d931b02eef0d02"
     },
     {
-      "bytes": 8092,
+      "bytes": 8111,
       "path": "site/assets/stardate/2026-268-memory-reach.html",
-      "sha256": "70e8a9b1b154b5c71441c074f242572e8001ca7c72220fb86d1b18284a91ed90"
+      "sha256": "e7940ff02995ece6c81f17c9c4ac1c1272cb04c8f5d3afe351fd916eee0a3d82"
     },
     {
-      "bytes": 8374,
+      "bytes": 8400,
       "path": "site/assets/stardate/index.html",
-      "sha256": "51bc251d6a0f23b2d309dac650b4b6a4b91057f6162f82f5ec2138e7d93ca416"
+      "sha256": "c451ac767c354e5d70580baaddb73935f454ea16e96e796e705a047273ddff4f"
     },
     {
       "bytes": 7046,
@@ -1753,9 +1753,9 @@
       "sha256": "0ef3fd6b85ade71ccb594334f5f3e4a41000f2d9167c380b95b9248d73297d2c"
     },
     {
-      "bytes": 91082,
+      "bytes": 91108,
       "path": "site/index.html",
-      "sha256": "59ddda9bdaeed52e7671f3a10eeffe88abfdb37986db1b4147440b6b2ff7f007"
+      "sha256": "ec38faf6213bf6619bb68b236e02a097d30644ca1b4e54c02ee36cc8cc96dd23"
     },
     {
       "bytes": 2877,
@@ -1783,9 +1783,9 @@
       "sha256": "f5c6432059172bba4bed215be597394e62fce5b2cc2f35f9a4baa60caaad94e8"
     },
     {
-      "bytes": 3839,
+      "bytes": 4800,
       "path": "site/stardate/entries/2026-268-deep-dream-day-one.md",
-      "sha256": "dab39f0dfc2c07f372aa58232f78817d7753b15e634ce9114f89b932b5ba339a"
+      "sha256": "2e03c03e93cf7c7785552103a49f8967f4806bae39d3b09a38b311bbde331b7a"
     },
     {
       "bytes": 3522,
@@ -1983,7 +1983,7 @@
       "sha256": "cd4bdb4529012e0cfcd38e059215f9ea433b8ee1fa636276b36c6515e7949e28"
     }
   ],
-  "inventory_sha256": "772c13116dd673c24a24aeee691bcf57ee60891a7b75da8310f286d5c7a28a58",
+  "inventory_sha256": "23769a5f267e127828ca7376832656052cd92035ab9af3b2537bd13d00cce849",
   "schema": "borg-public-inventory/v1",
-  "total_bytes": 87981209
+  "total_bytes": 87983389
 }

site/assets/stardate/2026-268-deep-dream-day-one.html

@@ -4,7 +4,7 @@
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
 <meta name="theme-color" content="#080b09">
-<meta name="description" content="Our first full clean-up pass over the main studio's memory retired 3,133 of 117,506 live memories by 09:36, all reversibly. The near-duplicate wave paused at the day's JEV budget, and we resumed it the same afternoon.">
+<meta name="description" content="Our first full clean-up pass over the main studio's memory retired 3,618 of 117,506 live memories, 3.1%, all reversibly. It finished at 20:03 after the near-duplicate wave resumed on a raised budget, and recall never returned a retired copy.">
 <meta name="generator" content="Stardate log publisher (static, from Markdown entries)">
 <title>Stardate 2026.268: Deep dream, day one — BORG Collective</title>
 <link rel="canonical" href="https://borg.utlyze.com/assets/stardate/2026-268-deep-dream-day-one.html">
@@ -25,11 +25,11 @@
 <p class="sd-stardate">STARDATE <b>2026.268</b></p>
 <h1 id="entry-title">Deep dream, day one</h1>
 <dl class="sd-meta">
-<div><dt>Date</dt><dd><time datetime="2026-09-25T09:36">25 September 2026, 09:36</time> <span class="sd-tz">America/Denver</span></dd></div>
-<div><dt>Status</dt><dd><span class="sd-status" data-status="running">Running</span></dd></div>
+<div><dt>Date</dt><dd><time datetime="2026-09-25T20:03">25 September 2026, 20:03</time> <span class="sd-tz">America/Denver</span></dd></div>
+<div><dt>Status</dt><dd><span class="sd-status" data-status="finished">Finished</span></dd></div>
 <div><dt>Format</dt><dd>Did · Learned · Open · Log</dd></div>
 </dl>
-<p class="sd-summary">Our first full clean-up pass over the main studio's memory retired 3,133 of 117,506 live memories by 09:36, all reversibly. The near-duplicate wave paused at the day's JEV budget, and we resumed it the same afternoon.</p>
+<p class="sd-summary">Our first full clean-up pass over the main studio's memory retired 3,618 of 117,506 live memories, 3.1%, all reversibly. It finished at 20:03 after the near-duplicate wave resumed on a raised budget, and recall never returned a retired copy.</p>
 </header>
 <section class="sd-section sd-what-we-did" aria-labelledby="what-we-did">
 <h2 id="what-we-did"><span class="sd-num" aria-hidden="true">01</span> What we did</h2>
@@ -39,23 +39,23 @@ <h2 id="what-we-did"><span class="sd-num" aria-hidden="true">01</span> What we d
 </section>
 <section class="sd-section sd-what-we-learned" aria-labelledby="what-we-learned">
 <h2 id="what-we-learned"><span class="sd-num" aria-hidden="true">02</span> What we learned</h2>
-<p>By 09:36, 3,133 memories were retired, 2.7% of the plan.</p>
-<div class="sd-table" role="region" tabindex="0" aria-label="What we learned, table 1"><table><thead><tr><th scope="col">Wave</th><th scope="col">Candidates</th><th scope="col">Retired</th><th scope="col">Why the rest stayed</th></tr></thead><tbody><tr><th scope="row">Machine-state snapshots</th><td class="num">13</td><td class="num">9</td><td>3 kept by JEV; 1 not yet 7 days old</td></tr><tr><th scope="row">Tool echoes</th><td class="num">498</td><td class="num">371</td><td>90 kept by JEV; 3 held back for privacy; 34 pointed at by another marker</td></tr><tr><th scope="row">Exact duplicates</th><td class="num">535 extra copies</td><td class="num">11</td><td>4 protected decisions; the rest refused by the planner's compatibility rules</td></tr><tr><th scope="row">Near-duplicates</th><td class="num">9,642 pairs</td><td class="num">2,742</td><td>6,010 pairs judged so far; 142 contradictions logged, not merged; 103 held by the chain rule</td></tr></tbody></table></div>
+<p>The pass retired 3,618 memories, 3.1% of the plan. By 09:36 it had retired 3,133; the near-duplicate wave paused at the day's JEV budget, resumed at 16:09 and finished at 20:03.</p>
+<div class="sd-table" role="region" tabindex="0" aria-label="What we learned, table 1"><table><thead><tr><th scope="col">Wave</th><th scope="col">Candidates</th><th scope="col">Retired</th><th scope="col">Why the rest stayed</th></tr></thead><tbody><tr><th scope="row">Machine-state snapshots</th><td class="num">13</td><td class="num">9</td><td>3 kept by JEV; 1 not yet 7 days old</td></tr><tr><th scope="row">Tool echoes</th><td class="num">498</td><td class="num">371</td><td>90 kept by JEV; 3 held back for privacy; 34 pointed at by another marker</td></tr><tr><th scope="row">Exact duplicates</th><td class="num">535 extra copies</td><td class="num">11</td><td>4 protected decisions; the rest refused by the planner's compatibility rules</td></tr><tr><th scope="row">Near-duplicates</th><td class="num">9,642 pairs</td><td class="num">3,227</td><td>9,297 pairs judged; 228 contradictions logged, not merged; 109 held by the chain rule</td></tr></tbody></table></div>
 <p>For near-duplicates, JEV must say &quot;same fact&quot; or &quot;supersedes&quot; with at least 90% confidence; then the shorter or older copy is retired. The chain rule keeps any note that another marker points to.</p>
-<p><strong>The checks held.</strong> Recall, sampled 40 times after each wave, never returned a retired copy, and found the kept fact about 90% of the time. Every changed memory changed only in its marker and labels. A live undo test on the snapshot wave restored 9, and all 13 notes matched the before-snapshot byte for byte. Spot checks of merges down to a similarity of 0.92 found only true rewordings.</p>
-<p><strong>It was cheap.</strong> JEV calls since 07:05 cost $0.176, for 6,173 calls. Deep dream was set to stop at $0.23 of a rolling 24-hour budget of $0.25 that it shares with the nightly. The near-duplicate wave stopped at the day's JEV budget. The remaining 3,632 pairs were first set for the next morning; instead we raised the rolling budget to $1.00, with an undo, and resumed them at 16:09.</p>
-<p>What surprised us: exact duplicates, which sound easy, retired only 11 of 535 extra copies. And on most near-duplicate pairs JEV was at least 90% sure the notes were redundant, but split its answer between &quot;same fact&quot; and &quot;newer replaces older&quot;. Neither answer reached 90%, so both copies stayed.</p>
+<p><strong>The checks held.</strong> Recall, sampled 40 times after each wave, never returned a retired copy, and found the kept fact about 90% of the time. At the finish we sampled it 300 times: it returned no retired copy, every kept copy was still live, and the control search found 298 of 300. Of the 19,284 near-duplicate candidates, 3,227 changed only in their marker and labels, and none changed in any other way. A live undo test on the snapshot wave restored 9, and all 13 notes matched the before-snapshot byte for byte. Spot checks of merges down to a similarity of 0.92 found only true rewordings.</p>
+<p><strong>It was cheap.</strong> JEV calls from 07:05 to the finish cost $0.281, for 9,584 calls: $0.176 for 6,173 calls by the morning, and $0.098 for 3,411 calls in the evening. Deep dream first stopped at $0.23 of a rolling 24-hour budget of $0.25 that it shares with the nightly. We raised that budget to $1.00, with an undo, and resumed the remaining pairs at 16:09 instead of waiting for the next morning.</p>
+<p>What surprised us: exact duplicates, which sound easy, retired only 11 of 535 extra copies. And most near-duplicate pairs stayed: of 9,297 judged, JEV cleared 3,336 for retirement. On many of the rest it was at least 90% sure the notes were redundant, but split its answer between &quot;same fact&quot; and &quot;newer replaces older&quot;. Neither answer reached 90%, so both copies stayed.</p>
 </section>
 <section class="sd-section sd-open-questions" aria-labelledby="open-questions">
 <h2 id="open-questions"><span class="sd-num" aria-hidden="true">03</span> Open questions</h2>
-<ul><li>Which copy should we keep when JEV splits its answer? One proposal: the newer copy when it holds everything the older says, otherwise the longer.</li><li>385 exact-duplicate groups hold the same text saved as different kinds of note. Nothing merges them yet.</li><li>The lane re-checks its permission with about 6 Inbox hub calls per pair, which caused 9 safe stops when the hub was slow. Can it check less often, safely?</li></ul>
+<ul><li>Which copy should we keep when JEV splits its answer? One proposal: the newer copy when it holds everything the older says, otherwise the longer.</li><li>385 exact-duplicate groups hold the same text saved as different kinds of note. Nothing merges them yet.</li><li>The lane re-checks its permission with about 6 Inbox hub calls per pair. When the hub was slow, 17 of those checks failed safely and both copies stayed. Can it check less often, safely?</li><li>Late in the evening the hub was busy enough that recall's own permission check started missing its deadline too. Should lanes share one short-lived proof of permission instead of each asking the hub?</li></ul>
 </section>
 <section class="sd-section sd-lab-book-log" aria-labelledby="lab-book-log">
 <h2 id="lab-book-log"><span class="sd-num" aria-hidden="true">04</span> Lab book log</h2>
-<ol class="sd-log"><li><span class="sd-phase">review</span><p>Three independent code reviews, then a rehearsal on a copy; every stage undone byte for byte.</p></li><li><time datetime="2026-09-25T06:22">06:22</time><p>Live plan made over 117,506 live memories.</p></li><li><time datetime="2026-09-25T07:05">07:05</time><p>Cost window opens; $0.176 over 6,173 JEV calls from here.</p></li><li><time datetime="2026-09-25T09:36">09:36</time><p>3,133 retired, 2.7% of the plan; near-duplicates paused at the day's JEV budget.</p></li><li><time datetime="2026-09-25T16:09">16:09</time><p>Rolling budget raised to $1.00, with an undo; the remaining near-duplicate pairs resumed.</p></li></ol>
+<ol class="sd-log"><li><span class="sd-phase">review</span><p>Three independent code reviews, then a rehearsal on a copy; every stage undone byte for byte.</p></li><li><time datetime="2026-09-25T06:22">06:22</time><p>Live plan made over 117,506 live memories.</p></li><li><time datetime="2026-09-25T07:05">07:05</time><p>Cost window opens.</p></li><li><time datetime="2026-09-25T09:36">09:36</time><p>3,133 retired, 2.7% of the plan; near-duplicates paused at the day's JEV budget after 6,173 calls ($0.176).</p></li><li><time datetime="2026-09-25T16:09">16:09</time><p>Rolling budget raised to $1.00, with an undo; the remaining near-duplicate pairs resumed.</p></li><li><time datetime="2026-09-25T18:47">18:47</time><p>A slow hub stopped the wave safely; it resumed 20 seconds later.</p></li><li><time datetime="2026-09-25T20:03">20:03</time><p>Finished: every remaining pair judged; 3,618 retired in total, 3.1% of the plan; 9,584 JEV calls ($0.281).</p></li><li><time datetime="2026-09-25T20:05">20:05</time><p>Final checks: 300 recall samples returned no retired copy; no memory changed except its marker and labels.</p></li></ol>
 </section>
 </article>
-<nav class="sd-pager" aria-label="More entries"><a class="sd-pager-older" href="./2026-268-memory-reach.html"><span>Older entry</span> Memory rarely reaches the agents that need it</a><a class="sd-pager-all" href="./">All entries</a><a class="sd-pager-newer" href="./2026-268-incoming-filter-on.html"><span>Newer entry</span> The incoming filter, from shadow to on</a></nav>
+<nav class="sd-pager" aria-label="More entries"><a class="sd-pager-older" href="./2026-268-marker-chains.html"><span>Older entry</span> Marker chains, when the kept copy gets retired</a><a class="sd-pager-all" href="./">All entries</a></nav>
 </main>
 <footer class="footer wrap">
 <a class="brand" href="../../"><img src="../mark.svg" width="30" height="34" alt=""><span>BORG<span class="brand-sub">COLLECTIVE</span></span></a>

site/assets/stardate/2026-268-incoming-filter-on.html

@@ -56,7 +56,7 @@ <h2 id="lab-book-log"><span class="sd-num" aria-hidden="true">04</span> Lab book
 <ol class="sd-log"><li><time datetime="2026-09-24T12:55">09-24 12:55</time><p>Credential fix in place; the 24-hour shadow clock counts from here.</p></li><li><time datetime="2026-09-24T12:56">09-24 12:56</time><p>Capture path starts in shadow mode: JEV judges, nothing is dropped.</p></li><li><time datetime="2026-09-24T14:03">09-24 14:03</time><p>Revised build installed after a third review; recovery copies forced to disk.</p></li><li><span class="sd-phase">shadow</span><p>24 hours: 418 captures, 2,498 candidate facts, 1,894 JEV calls.</p></li><li><span class="sd-phase">review</span><p>All 116 would-be drops read by hand; zero durable facts.</p></li><li><time datetime="2026-09-25T13:00">13:00</time><p>Switched to on; the service restarted in about 4 seconds.</p></li><li><time datetime="2026-09-25T14:07">14:07</time><p>First real drop, a script error echo, saved first to a recovery copy and verified.</p></li></ol>
 </section>
 </article>
-<nav class="sd-pager" aria-label="More entries"><a class="sd-pager-older" href="./2026-268-deep-dream-day-one.html"><span>Older entry</span> Deep dream, day one</a><a class="sd-pager-all" href="./">All entries</a><a class="sd-pager-newer" href="./2026-268-marker-chains.html"><span>Newer entry</span> Marker chains, when the kept copy gets retired</a></nav>
+<nav class="sd-pager" aria-label="More entries"><a class="sd-pager-older" href="./2026-268-memory-reach.html"><span>Older entry</span> Memory rarely reaches the agents that need it</a><a class="sd-pager-all" href="./">All entries</a><a class="sd-pager-newer" href="./2026-268-marker-chains.html"><span>Newer entry</span> Marker chains, when the kept copy gets retired</a></nav>
 </main>
 <footer class="footer wrap">
 <a class="brand" href="../../"><img src="../mark.svg" width="30" height="34" alt=""><span>BORG<span class="brand-sub">COLLECTIVE</span></span></a>

site/assets/stardate/2026-268-marker-chains.html

@@ -56,7 +56,7 @@ <h2 id="lab-book-log"><span class="sd-num" aria-hidden="true">04</span> Lab book
 <ol class="sd-log"><li><time datetime="2026-09-06">2026-09-06</time><p>Start of the count: 503 exact-on-exact chains built up from here, about 26 a night.</p></li><li><time datetime="2026-09-25T13:44">13:44</time><p>Census on the main studio: 7,222 marked notes, 527 inactive duplicate markers.</p></li><li><span class="sd-phase">cause</span><p>Traced to the nightly's exact step letting a newer identical copy win its group.</p></li><li><span class="sd-phase">fix</span><p>Planner protects every kept copy; the nightly scans them before each plan.</p></li><li><span class="sd-phase">review</span><p>Independent review: accept, with 0 blocking findings.</p></li><li><span class="sd-phase">install</span><p>Fix installed with a one-command undo.</p></li><li><span class="sd-phase">repair</span><p>527 markers restored with 0 other changes; the census then showed 0 inactive.</p></li><li><time datetime="2026-09-25T14:58">14:58</time><p>State recorded; the first protected nightly is set for 09-26 03:30.</p></li></ol>
 </section>
 </article>
-<nav class="sd-pager" aria-label="More entries"><a class="sd-pager-older" href="./2026-268-incoming-filter-on.html"><span>Older entry</span> The incoming filter, from shadow to on</a><a class="sd-pager-all" href="./">All entries</a></nav>
+<nav class="sd-pager" aria-label="More entries"><a class="sd-pager-older" href="./2026-268-incoming-filter-on.html"><span>Older entry</span> The incoming filter, from shadow to on</a><a class="sd-pager-all" href="./">All entries</a><a class="sd-pager-newer" href="./2026-268-deep-dream-day-one.html"><span>Newer entry</span> Deep dream, day one</a></nav>
 </main>
 <footer class="footer wrap">
 <a class="brand" href="../../"><img src="../mark.svg" width="30" height="34" alt=""><span>BORG<span class="brand-sub">COLLECTIVE</span></span></a>