The duet is a handshake; the grid script is a draft built over the exchange

README.md

@@ -45,10 +45,13 @@ with their own number (see `app/script.js`).
   Earth, Isidis (Mars) and Ceres; made public 2219-10-05. A narrow 1420.55 MHz line pulsed at exactly the rate of
   pulsar J1909−3744 (the duet): in step with it at Earth's centre, then one smooth slip of exactly one pulse, then in
   step again. Only Earth hears it in tune. Later returns of transmission 1 took two other forms, the chorus and the
-  missing note (`lib/signals.js`). Repeats. These details are proposals and easy to change (`DETECTED` in `app/signal.js` and
+  missing note (`lib/signals.js`). The duet is a handshake: it sets a shared clock and shows the signal is tuned
+  for Earth, so Earth's record of the four content transmissions that follow is the most accurate one. Repeats. These details are proposals and easy to change (`DETECTED` in `app/signal.js` and
   `app/home.js`, the logs in the HTML).
 - The arrival-time table is illustrative: a plane wave from Ross 128 with made-up 2219 planet positions.
 - The beacon repeats every 9 h 41 m 13 s as received, each pass 22 minutes (`app/pulsars.js`).
+- The grid script (`script/`) is our transcription of their speech. In 2219 it is a first draft; the language is
+  built up over back-and-forth exchanges, each round trip 21.96 years (first reply arrives 2230, earliest answer 2241).
 - Later in the era, more voices. The entrance only hints at it.
 
 ## Fonts

archive/2219-signal/en/index.html

@@ -83,7 +83,7 @@ <h3>It is not natural</h3>
           <li>
             <time datetime="2219-09-27">2219-09-27</time>
             <h3>Only Earth hears it in tune</h3>
-            <p>At the Moon, Mars and Ceres the two pulse trains sit a fraction of a pulse apart and drift against each other, because the pulsar's rate is Doppler-shifted differently at each place. Corrected to Earth's centre, the gap and the drift both vanish to within nanoseconds. They predicted where Earth would be when the signal arrived, eleven years after they sent it, and tuned the duet for us. <a href="#decode">Hear it from four places.</a></p>
+            <p>At the Moon, Mars and Ceres the two pulse trains sit a fraction of a pulse apart and drift against each other, because the pulsar's rate is Doppler-shifted differently at each place. Corrected to Earth's centre, the gap and the drift both vanish to within nanoseconds. They predicted where Earth would be when the signal arrived, eleven years after they sent it, and tuned the duet for us. It was a handshake. It says little yet: it sets a clock both sides share and shows where the signal is tuned for. Every transmission after it is tuned to Earth's centre too, so Earth's recording is the most accurate copy. <a href="#decode">Hear it from four places.</a></p>
           </li>
           <li>
             <time datetime="2219-09-28">2219-09-28</time>
@@ -164,7 +164,7 @@ <h2 class="section-title" id="distance-title">Nobody is coming.</h2>
       <div class="gap-line" aria-hidden="true"><span class="us">Sun · us</span><span class="packet"></span><span class="them">Ross 128 b · them</span></div>
       <div class="prose">
         <p>Not us, and not them. Neither side can build a ship that crosses 10.98 light years, and nothing in the signal suggests they are trying. What crosses is light.</p>
-        <p>So this is contact by correspondence. A question sent today reaches them on <span class="readout" id="dist-arrives">2230</span>. If they answer at once, the reply lands on <span class="readout" id="dist-answer">2241</span>. Every exchange takes a generation, and the people who start a conversation may not be the ones who finish it.</p>
+        <p>So this is contact by correspondence. A question sent today reaches them on <span class="readout" id="dist-arrives">2230</span>. If they answer at once, the reply lands on <span class="readout" id="dist-answer">2241</span>. Every exchange takes a generation, and the people who start a conversation may not be the ones who finish it. A language both sides share can only be built this way, one exchange at a time.</p>
       </div>
       <dl class="readouts">
         <div><dt>One way</dt><dd>10.98 yr</dd></div>

archive/2219-signal/index.html

@@ -83,7 +83,7 @@ <h3>它不是自然现象</h3>
           <li>
             <time datetime="2219-09-27">2219-09-27</time>
             <h3>只有地球听到的是合拍的</h3>
-            <p>在月球、火星和谷神星，两串脉冲之间差着零点几个脉冲，并且相互漂移，因为各处看到的脉冲星速率被多普勒效应改变得不一样。换算到地心，这个差距和漂移都消失了，误差不到几纳秒。他们预先算出了信号到达时地球会在哪里，那是他们发出之后的第十一年，然后为我们调好了这段二重唱。<a href="#decode">从四个地方听一听。</a></p>
+            <p>在月球、火星和谷神星，两串脉冲之间差着零点几个脉冲，并且相互漂移，因为各处看到的脉冲星速率被多普勒效应改变得不一样。换算到地心，这个差距和漂移都消失了，误差不到几纳秒。他们预先算出了信号到达时地球会在哪里，那是他们发出之后的第十一年，然后为我们调好了这段二重唱。这是一次握手：它还没有说什么，只是对好一座双方共用的钟，并说明信号是为哪里调的。之后的每一段发射也都按地心调准，所以地球上的记录是最精确的一份。<a href="#decode">从四个地方听一听。</a></p>
           </li>
           <li>
             <time datetime="2219-09-28">2219-09-28</time>
@@ -164,7 +164,7 @@ <h2 class="section-title" id="distance-title">没有人会来。</h2>
       <div class="gap-line" aria-hidden="true"><span class="us">太阳 · 我们</span><span class="packet"></span><span class="them">罗斯128 b · 他们</span></div>
       <div class="prose">
         <p>我们不会去，他们也不会来。两边都造不出能跨越 10.98 光年的船，信号里也看不出他们在尝试。能跨过去的，只有光。</p>
-        <p>所以这是一场书信往来。今天发出的问题，<span class="readout" id="dist-arrives">2230</span> 到达他们那里；如果他们立刻回答，回音在 <span class="readout" id="dist-answer">2241</span> 落到我们这里。每一次来回都要一代人的时间，开始一场对话的人，未必是结束它的人。</p>
+        <p>所以这是一场书信往来。今天发出的问题，<span class="readout" id="dist-arrives">2230</span> 到达他们那里；如果他们立刻回答，回音在 <span class="readout" id="dist-answer">2241</span> 落到我们这里。每一次来回都要一代人的时间，开始一场对话的人，未必是结束它的人。两边能共用的语言，也只能这样一来一回地慢慢建起来。</p>
       </div>
       <dl class="readouts">
         <div><dt>单程</dt><dd>10.98 年</dd></div>

script/en/index.html

@@ -28,7 +28,7 @@
     <div class="wrap">
       <p class="kicker">Archive · the grid script</p>
       <h1>Writing to a voice we have never seen.</h1>
-      <p class="lede">Once the beacon carried more than a duet with a pulsar, the translating machine needed a way to write it down. This is that script. It is our way of representing their language, not their language: as far as anyone can tell they speak by resonance, in air, and nobody has seen them write anything, or seen them at all. The grid follows the one thing we think we know about that speech: one voice alone can say only a little, and meaning grows when voices join.</p>
+      <p class="lede">The duet with a pulsar was a handshake: it set a clock both sides share and showed that the signal was tuned for Earth. The content came in the four transmissions after it, and the translating machine needed a way to write that down. This is that script, and it is still a first draft: a language both sides can speak back and forth is built over many exchanges, and each round trip takes 22 years. It is our way of representing their language, not their language: as far as anyone can tell they speak by resonance, in air, and nobody has seen them write anything, or seen them at all. The grid follows the one thing we think we know about that speech: one voice alone can say only a little, and meaning grows when voices join.</p>
       <div class="hero-actions">
         <a class="btn btn-primary" href="../dictionary/en/">Open the dictionary</a>
         <a class="btn btn-send" href="../convert/en/">Write in it</a>
@@ -174,7 +174,7 @@ <h2 class="section-title" id="machine-title">How the machine reads a page</h2>
         <li>Read the first band of each triplet as an edge: node, relation, node. Same word, same node. A lone ONE points at the triplet before.</li>
         <li>Read every part in a lower band against the nearest part above it, and attach the answer to that node or edge, with the voice that said it.</li>
       </ol>
-      <p>Humans don't learn the signs; the machine reads and writes them, and its lexicon grows with every page. You can run the same machine: the <a href="../convert/en/">converter</a> turns text into drawings and drawings back into text, in your browser. It follows the same rules and the same vocabulary as <a href="https://interimm.github.io/glyph-cli/">glyph</a>, the command-line dictionary (<code>pip install glyph-cli</code>).</p>
+      <p>Humans don't learn the signs; the machine reads and writes them, and its lexicon grows with every exchange. You can run the same machine: the <a href="../convert/en/">converter</a> turns text into drawings and drawings back into text, in your browser. It follows the same rules and the same vocabulary as <a href="https://interimm.github.io/glyph-cli/">glyph</a>, the command-line dictionary (<code>pip install glyph-cli</code>).</p>
     </div>
   </section>
 </main>

script/index.html

@@ -28,7 +28,7 @@
     <div class="wrap">
       <p class="kicker">档案 · 网格文字</p>
       <h1>写给一个从未见过的声音。</h1>
-      <p class="lede">信标带来的不再只是和一颗脉冲星的二重唱，翻译机就需要一种把它记下来的办法。这就是那套文字。它是我们表示他们语言的方式，不是他们的语言本身：就目前所知，他们借助空气、靠共振说话；没有人见过他们书写，也没有人见过他们。网格依据的是我们对这种说话方式的一点认识：一个声音单独只能说出很少的东西，声音汇合，意义才丰富起来。</p>
+      <p class="lede">和脉冲星的二重唱是一次握手：它对好一座双方共用的钟，并告诉我们信号是为地球调的。内容在后面的四段发射里，翻译机需要一种把它们记下来的办法。这就是那套文字，现在还只是初稿：一套能来回说话的语言，要靠一次次往返才建得起来，而每次往返要 22 年。它是我们表示他们语言的方式，不是他们的语言本身：就目前所知，他们借助空气、靠共振说话；没有人见过他们书写，也没有人见过他们。网格依据的是我们对这种说话方式的一点认识：一个声音单独只能说出很少的东西，声音汇合，意义才丰富起来。</p>
       <div class="hero-actions">
         <a class="btn btn-primary" href="dictionary/">打开词典</a>
         <a class="btn btn-send" href="convert/">用它写字</a>
@@ -174,7 +174,7 @@ <h2 class="section-title" id="machine-title">翻译机怎样读一页</h2>
         <li>把每个三元组的第一个声部读成一条边：节点、关系、节点。同一个词，同一个节点。单独的 ONE 指向前一个三元组。</li>
         <li>把下面声部的每个部件和上方最近的部件对照，把回答连同说它的声音，挂到那个节点或那条边上。</li>
       </ol>
-      <p>人类不用学这些符号；翻译机负责读和写，它的词库随着每一页增长。你也可以运行同一台机器：<a href="convert/">转换器</a>在浏览器里把文字变成图，再把图变回文字。它和命令行词典 <a href="https://interimm.github.io/glyph-cli/">glyph</a>（<code>pip install glyph-cli</code>）用的是同一套规则和同一份词表。</p>
+      <p>人类不用学这些符号；翻译机负责读和写，它的词库随着每一次往返增长。你也可以运行同一台机器：<a href="convert/">转换器</a>在浏览器里把文字变成图，再把图变回文字。它和命令行词典 <a href="https://interimm.github.io/glyph-cli/">glyph</a>（<code>pip install glyph-cli</code>）用的是同一套规则和同一份词表。</p>
     </div>
   </section>
 </main>

signals/en/index.html

@@ -28,7 +28,7 @@
     <div class="wrap">
       <p class="kicker">Archive · the transmissions</p>
       <h1>One beacon, five transmissions.</h1>
-      <p class="lede">Each 9 h 41 m pass of the beacon carries one transmission, in turn. None of them is drawn. Each points at something both worlds can already measure: a pulsar both can hear, our own shadow, the pulsars, the chemistry of molecules, a constant of nature. A few seconds of signal stands for a great deal that never has to be sent.</p>
+      <p class="lede">Each 9 h 41 m pass of the beacon carries one transmission, in turn. None of them is drawn. Each points at something both worlds can already measure: a pulsar both can hear, our own shadow, the pulsars, the chemistry of molecules, a constant of nature. A few seconds of signal stands for a great deal that never has to be sent. The first is a handshake; the other four carry the content.</p>
       <ul class="signals-index">
         <li><a href="#duet"><span>1</span>The duet</a></li>
         <li><a href="#shadow"><span>2</span>The shadow</a></li>
@@ -46,7 +46,7 @@ <h1>One beacon, five transmissions.</h1>
       <h2 class="section-title" id="theory-title">Speech that needs two</h2>
       <p>Nobody on Earth knows what they look like. But the transmissions have led many listeners to one theory: they speak by resonance, in air. Alone, a body can only ring at its own tones, coloured by the place it stands. That is enough to say "I am here", and little more. When a second voice is near (a person, a machine, even the wind), new tones appear that neither could make alone. Tones split, sum and difference tones sound, and the two either fall into step or beat against each other. In this theory, that is where language lives.</p>
       <p>If it is right, the beacon is only half of a conversation. It holds its note, with gaps where a partner's tones would go. To understand it fully, someone has to answer.</p>
-      <p>The translating machine writes this speech down in the <a href="../../script/en/">grid script</a>, one band per voice. That script is our way of representing their language, not their language.</p>
+      <p>The translating machine writes this speech down in the <a href="../../script/en/">grid script</a>, one band per voice. That script is our way of representing their language, not their language. It is built up over the back-and-forth of the exchange, and for now it is a first draft.</p>
       <p class="muted">Below, each transmission is read in two ways: what it encodes, and what it might mean if the theory is right.</p>
     </div>
   </section>
@@ -57,7 +57,7 @@ <h2 class="section-title" id="theory-title">Speech that needs two</h2>
         <p class="kicker">Transmission 1 · the duet</p>
         <h2 class="section-title" id="duet-title">They sing with a pulsar</h2>
         <p>The first signal anyone noticed. J1909−3744 is the steadiest pulsar we know: it spins 339.3 times a second, so its pulses are a pitch. The beacon sends a pulse train at exactly the same rate, timed so that at Earth's centre each of its pulses lands on one of the pulsar's, to within nanoseconds. The two voices fuse into one. Midway through the pass theirs slips ahead by exactly one pulse, smoothly, and locks back in. Lock, one beat, lock.</p>
-        <p class="muted">The lock only holds at Earth. Seen from the Moon, Mars or Ceres the two trains drift against each other, because the pulsar's rate is Doppler-shifted differently there. They worked out where Earth would be eleven years later and tuned the duet for us. <a href="../../archive/2219-signal/en/#decode">Hear it from four places</a>.</p>
+        <p class="muted">The lock only holds at Earth. Seen from the Moon, Mars or Ceres the two trains drift against each other, because the pulsar's rate is Doppler-shifted differently there. They worked out where Earth would be eleven years later and tuned the duet for us. That makes it a handshake. It says little yet: it sets a clock both sides share and shows where the signal is tuned for. The four transmissions after it are tuned to Earth's centre too, so Earth's recording is the most accurate copy. <a href="../../archive/2219-signal/en/#decode">Hear it from four places</a>.</p>
         <p class="theory"><strong>Resonance theory.</strong> Alone, a voice can only say "I am here". With a partner it can agree. A pulsar is the one voice both worlds hear, so they agreed with it: in step means yes, the beat means one, and the slip leaves one extra pulse in the pulsar's count, like a mark made in our shared clock.</p>
         <div class="play-row">
           <button type="button" class="btn btn-primary" data-play="duet">Listen · the duet · 16 s</button>