<?xml version="1.0" encoding="UTF-8"?>
<!--
  THREE ENTRIES, AND THEY ARE ALL THE JOURNAL. The product is not one of them
  and neither is the Method.

  `/` IS THE JOURNAL SINCE 2026-09-07, not the register. The root rule lives on
  the host and this file only reports it; the canonicals in journal.html and
  index.html are the other two thirds of the same decision and the three move
  together.

  THE REGISTER AND THE METHOD ARE NOT LISTED, and the absence is the decision.
  Both answer 200 at /index.html and /method.html, neither carries a noindex,
  and robots.txt disallows nothing: they are not hidden and this file does not
  pretend they are. What it does is stop handing anybody the map to two
  sections that publish before they are meant to be read — a sitemap is an
  invitation, and the invitation is what was withdrawn. A crawler that finds
  them another way is welcome to them, which is the same terms every other
  page on this site has always been served on.
  PUTTING THEM BACK is two blocks, and they go back the day the bar reopens,
  in the commit that reopens it.

  THE PRODUCT IS STILL NOT ONE URL. A fragment never reaches the server, so
  the register's twenty-four pages were always one address to anything that
  crawls. That was known item 1 while it was listed and it is known item 1
  now; delisting neither fixes it nor makes it worse.
  A fragment never reaches the server, so the twenty-four pages of the register
  are one URL to anything that crawls: everything under #/ on index.html is the
  first entry below and nothing else. Listing a hash route here would be this
  file claiming an address the host has never been asked for. That is known
  item 1 in manual/known-items.md, this file does not fix it, and the fix is
  the router and the head rather than more lines in here.

  ONLY A PATH THAT ANSWERS 200 MAY BE LISTED. The host serves a whitelist, not
  a tree, and an entry it does not name is a 404 with a sitemap pointing at it.
  This file has to be on that list itself. See WHITELIST.md.

  A CASE IS A FOLDER, since 2026-09-07: journal/<case>/<case>-<slug>.html, and
  the two filed cases moved into theirs on that day. The old flat addresses
  never answered 200 on this host, so nothing that was published has moved and
  there is nothing here to redirect.

  lastmod is by hand, like the edition, and it moves with the document. The
  filed case carries the day it was published and not the day the file was last
  touched: a filed case is a historical citation, and restamping one would
  claim a reading it never made. 026-002 is stamped 2026-09-07, the day it
  publishes, which is the day this file may ship and not before; 026-003 is
  stamped 2026-09-17 on the same terms, and 026-001 and 026-002 keep the dates
  they were published under although the Journal moved above them.
  THE ROOT IS NOT A FILED CASE and its stamp does move: it is the Journal, it
  gained a new lead and edition Nº 003 on 2026-09-17, and a document that
  changed today saying it last changed ten days ago is the one lie a lastmod
  can tell.

  NO priority AND NO changefreq. Google reads neither, and a number this site
  cannot check is the one kind of figure it does not print.
-->
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">

  <!-- `/` IS THE JOURNAL, and journal.html names it as its canonical. The file
       also answers at /journal.html and that address is deliberately NOT listed
       below: it is the same document at a second URL, and a sitemap listing
       both hands a crawler a duplicate to resolve when the canonical has
       already resolved it. -->
  <url>
    <loc>https://mint.st/</loc>
    <lastmod>2026-09-17</lastmod>
  </url>

  <url>
    <loc>https://mint.st/journal/026-001/026-001-tokenized-stocks-depth.html</loc>
    <lastmod>2026-08-19</lastmod>
  </url>

  <url>
    <loc>https://mint.st/journal/026-002/026-002-memecoins-dirty-work.html</loc>
    <lastmod>2026-09-07</lastmod>
  </url>

  <!-- 026-003, filed and published 2026-09-17. It is a case and a case is a
       folder, so it is listed on the `journal/*/*.html` rule and on nothing
       else: the hero, the door plate and the two figures in the same folder
       are images and images are not invited, they are fetched by the page
       that prints them. Probe this URL by hand on the host the first time it
       ships — no local check catches a whitelist fault, and a sitemap
       pointing at a 404 is worse than a case nobody was told about. -->
  <url>
    <loc>https://mint.st/journal/026-003/026-003-velvet-rope-not-subject-to-sec-review.html</loc>
    <lastmod>2026-09-17</lastmod>
  </url>

</urlset>
