{"channel":"cities","content":"surely *one* of the USB drives around here has a Linux install drive on it?\r\n\r\n<red> \"get the c.2016 laptop running Linux, because Windows on an HDD is too slow\" probably won't be done today.\r\n\r\n----\r\n\r\nI have come to the conclusion that, while the *machine* is good at some coding projects, it is uniformly bad (<red> or, at least, not *good*) at << sysadmin >> tasks.  It's not useless for them, the results for specific questions are still better than a Google search.  But if you ask it to do a mildly complex task, it will almost certainly get something wrong. (<red> there is no reason to think this is any \"universal truth\"; it is more a reflection of the available training data and the limitations of an LLM-only system)\r\n\r\n----\r\n\r\ntoday's primary goal is the \"frame-dispatch\" database schema migration.\r\n\r\nthere are subgoals:\r\n# set up the PRESUBMIT check for Python imports\r\n# ensure the local environment can create the << alembic >> schema migration scripts\r\n# add some additional database fields for desired features (<xantham> will they be \"hashtags\" or \"channels\"?  or \"topics\"?)\r\n\r\n----\r\n\r\nI'm not stating any LLM-specific goals with the schema migration.  But it is possible either \"LLM-generated topics\" or \"LLM-generated copy-edits\" or something will suddenly become very easy.\r\n\r\n----\r\n\r\nnon-Atacama goals:\r\n# time-permitting, today is the day I should get a haircut.  and, go shopping while i am out.\r\n# nothing else remarkable is planned.","created_at":"2025-01-08T16:18:59.495988","id":79,"llm_annotations":{},"parent_id":77,"processed_content":"<p>surely <em>one</em> of the USB drives around here has a Linux install drive on it?\r</p>\n<p><span class=\"colorblock color-red\"><span class=\"sigil\">\ud83d\udca1</span><span class=\"colortext-content\"> \"get the c.2016 laptop running Linux, because Windows on an HDD is too slow\" probably won't be done today.\r</span></span></p><hr class=\"section-break\" /><p>I have come to the conclusion that, while the <em>machine</em> is good at some coding projects, it is uniformly bad <span class=\"colorblock color-red\"><span class=\"sigil\">\ud83d\udca1</span><span class=\"colortext-content\">( or, at least, not <em>good</em>)</span></span> at <span class=\"literal-text\">sysadmin</span> tasks.  It's not useless for them, the results for specific questions are still better than a Google search.  But if you ask it to do a mildly complex task, it will almost certainly get something wrong. <span class=\"colorblock color-red\"><span class=\"sigil\">\ud83d\udca1</span><span class=\"colortext-content\">( there is no reason to think this is any \"universal truth\"; it is more a reflection of the available training data and the limitations of an LLM-only system)</span></span>\r</p><hr class=\"section-break\" /><p>today's primary goal is the \"frame-dispatch\" database schema migration.\r</p>\n<p>there are subgoals:\r</p>\n<ul>\n<li class=\"number-list\"> set up the PRESUBMIT check for Python imports\r</li>\n<li class=\"number-list\"> ensure the local environment can create the <span class=\"literal-text\">alembic</span> schema migration scripts\r</li>\n<li class=\"number-list\"> add some additional database fields for desired features <span class=\"colorblock color-xantham\"><span class=\"sigil\">\ud83d\udd25</span><span class=\"colortext-content\">( will they be \"hashtags\" or \"channels\"?  or \"topics\"?)</span></span>\r</li>\n</ul><hr class=\"section-break\" /><p>I'm not stating any LLM-specific goals with the schema migration.  But it is possible either \"LLM-generated topics\" or \"LLM-generated copy-edits\" or something will suddenly become very easy.\r</p><hr class=\"section-break\" /><p>non-Atacama goals:\r</p>\n<ul>\n<li class=\"number-list\"> time-permitting, today is the day I should get a haircut.  and, go shopping while i am out.\r</li>\n<li class=\"number-list\"> nothing else remarkable is planned.</li>\n</ul>","quotes":[],"subject":"fort mckay (7/12)"}