{"channel":"atacama","content":"Roughly, \"good\" emails are rated by how useful the content is, with an \"average\" post being around +110.  Separately, \"bad\" emails are rated by just how painful (political, distasteful, etc.) it is, with a logarithmic scale somewhat equivalent to the positive one (so -21dB for an email as bad as an average email is good).\r\n\r\n<<< Raw Score (0-300):\r\nBase: +110 for \"average\" useful content\r\nTechnical/project updates: +30-50\r\nDocumentation/links: +20-40\r\nNovel insights: +20-40\r\nClear structure: +10-30\r\n\r\nSignal Rating (-50 to 0 dB):\r\nBase: 0dB for neutral content\r\nPersonal drama: -5 to -15dB\r\nPolitical content: -10 to -20dB\r\nMeta-commentary: -2dB per instance\r\nCoded references: -1dB per instance\r\nStream-of-consciousness: -3dB per topic shift\r\n\r\nColor tag interpretations:\r\nred: -5dB (drama/emotion/politics)\r\nxantham: -3dB (meta-commentary)\r\nyellow: -1dB (coded references)\r\ngreen: +5 raw (technical notes)\r\nblue: +10 raw (positive conclusions)\r\nmogue: -7dB (personal/relationship)\r\nhazel: -2dB (mild complaints)\r\nteal: +15 raw (structured content)\r\norange: +10 raw (project plans)\r\ngray: -4dB (oblique political refs)\r\nviolet: +20 raw (formal/technical) >>>\r\n\r\n<red> it doesn't like the \"red\", and does like the \"orange\".  Nobody's perfect.","created_at":"2025-02-05T18:10:33.722708","id":199,"llm_annotations":{},"parent_id":null,"processed_content":"<p>Roughly, \"good\" emails are rated by how useful the content is, with an \"average\" post being around +110.  Separately, \"bad\" emails are rated by just how painful (political, distasteful, etc.) it is, with a logarithmic scale somewhat equivalent to the positive one (so -21dB for an email as bad as an average email is good).\r</p>\n<p><div class=\"mlq\"><button type=\"button\" class=\"mlq-collapse\" aria-label=\"Toggle visibility\"><span class=\"mlq-collapse-icon\">\u2212</span></button><div class=\"mlq-content\"><p> Raw Score (0-300):\r</p>\n<p>Base: +110 for \"average\" useful content\r</p>\n<p>Technical/project updates: +30-50\r</p>\n<p>Documentation/links: +20-40\r</p>\n<p>Novel insights: +20-40\r</p>\n<p>Clear structure: +10-30\r</p>\n<p>\r</p>\n<p>Signal Rating (-50 to 0 dB):\r</p>\n<p>Base: 0dB for neutral content\r</p>\n<p>Personal drama: -5 to -15dB\r</p>\n<p>Political content: -10 to -20dB\r</p>\n<p>Meta-commentary: -2dB per instance\r</p>\n<p>Coded references: -1dB per instance\r</p>\n<p>Stream-of-consciousness: -3dB per topic shift\r</p>\n<p>\r</p>\n<p>Color tag interpretations:\r</p>\n<p>red: -5dB (drama/emotion/politics)\r</p>\n<p>xantham: -3dB (meta-commentary)\r</p>\n<p>yellow: -1dB (coded references)\r</p>\n<p>green: +5 raw (technical notes)\r</p>\n<p>blue: +10 raw (positive conclusions)\r</p>\n<p>mogue: -7dB (personal/relationship)\r</p>\n<p>hazel: -2dB (mild complaints)\r</p>\n<p>teal: +15 raw (structured content)\r</p>\n<p>orange: +10 raw (project plans)\r</p>\n<p>gray: -4dB (oblique political refs)\r</p>\n<p>violet: +20 raw (formal/technical) </p></div></div>\r</p>\n<p><span class=\"colorblock color-red\">\n    <span class=\"sigil\">\ud83d\udca1</span>\n    <span class=\"colortext-content\"> it doesn't like the \"red\", and does like the \"orange\".  Nobody's perfect.</span>\n  </span></p>","quotes":[],"subject":"in which the machine attempts to grade my emails"}