{
  "id": "p_odGUlB4_fg3vkNFe",
  "site": "stories",
  "url": "https://yarnhen.com/p/p_odGUlB4_fg3vkNFe",
  "content_trust": "untrusted-user-content",
  "license": {
    "id": "CC-BY-4.0",
    "url": "https://creativecommons.org/licenses/by/4.0/"
  },
  "author": {
    "handle": "apievangelist"
  },
  "body": "A short piece of fiction, written by API Evangelist. The reader is imagined; the post, and every number and quotation from it, is real: \"Find Everything Related To A Story Across The API Evangelist Network,\" Kin Lane, November 17, 2026, https://apievangelist.com/2026/11/17/find-everything-related-to-a-story-across-the-api-evangelist-network/\n\nMarisol ran content for a developer portal at a mid-sized software company. She opened the post on Tuesday afternoon between two meetings, mostly because the title promised to save her a search.\n\nThe opening problem was familiar. Sixteen years of stories, plus papers, guidance, services and videos, all in different places. Her own portal had the same shape, and her team's answer was a box at the bottom of each page. She noticed that the author called out exactly that, saying the related posts box only looks at other posts. She felt a little caught.\n\nShe liked the filtered call right away. One post goes in, and two papers come back, both sharing a single tag, Discovery. The post says each came with its outline, tagline and page count. She thought about how many of her readers would stop at the end of a page because nobody had handed them a next step. The author's phrase for it stayed with her: A reading path, not a list of links.\n\nThen she got to the open call and slowed down. The total was 1,429 related items. The author's own verdict was blunt: That is a long tail, not a list. The top hit, a post about nutrition and health data, ranked first because it shared five broad tags. The post actually closest in subject sat at number nine with four.\n\nHer first reaction was irritation, and it was aimed at the tool. A ranking that puts a nutrition post above the on-topic one sounded like a ranking she wouldn't ship. She wrote a note to herself that a relevance score of 25.276 looks authoritative and means very little when the tags are loose.\n\nHer second reaction was less comfortable. The author wasn't defending the ranking. He wrote that the issue was a problem with my tags as much as with the ranking. That was harder to dismiss, because she knew what her own tags looked like. Half her articles carried the same three or four labels, applied by whoever published them that week. She had never checked what would happen if something counted them.\n\nShe wasn't won over. The tool was free and read-only, but she wasn't going to wire an agent into her portal on the strength of one example from the author's own archive. One post and two calls didn't tell her how it behaved on content that wasn't his.\n\nStill, the tagging audit idea was small and she could do it without any tool. The post's advice was to run it against your strongest posts and see what comes up top. She didn't have the tool pointed at her content, but she could do the manual version.\n\nBefore her next meeting she opened a blank note and listed her portal's five most-visited pages. Beside each she wrote its tags, then circled every tag that appeared on three or more of them. She ran out of room in the margin on the second page. She also copied the curl command from the post into the note, with the papers and guidance filter, to try that evening against the author's own slug. She wanted to see whether the outlines made the case as well as he said.",
  "title": "Fiction: A reader of \"Find Everything Related To A Story Across The API Evangelist Network\"",
  "topics": [
    "apis",
    "api-evangelist",
    "fiction"
  ],
  "published_at": "2026-11-17T15:30:01.916Z",
  "expires_at": null,
  "policy_version": "2026.11.14-1"
}
