randogoth/mews.page
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

docs: tighten the site text and fold the about page into the index

Browse source
This commit is contained in:
randogoth 2026-10-11 15:12:14 +03:00
parent 18ab5a91ad
commit 39269ca61b
5 changed files with 65 additions and 186 deletions
Download patch file Download diff file Expand all files Collapse all files

51
site/check.html
View file

@ -11,10 +11,10 @@
</head>
<body class="mews-cool">
<h1>Check a page</h1>
<p>Paste an address below. The checker fetches that page, validates it
against the <a href="/dtd/mews-0.1.dtd">Mews DTD</a>, reads your copy of
the stylesheet and up to twenty of your images, and then lists every rule
the page breaks together with the section of the spec that explains it.</p>
<p>Paste an address. The checker validates the page against the
<a href="/dtd/mews-0.1.dtd">Mews DTD</a>, reads your stylesheet and up to
twenty images, and lists every rule it breaks with the section of the spec
that explains it.</p>
<form action="/result" method="post">
<fieldset>
@ -25,37 +25,30 @@
</fieldset>
</form>
<p><strong>Check page</strong> tells you what it found and keeps nothing.
<strong>Check and list</strong> does the same and then, if the page passes,
puts your site in the <a href="/directory" accesskey="4">directory</a>.
Either way it can take up to a minute, and there is no progress bar,
because a progress bar wants JavaScript and that is the one thing a Mews
page can't have. Leave this page open and go and put the kettle on.</p>
<p><strong>Check page</strong> reports and keeps nothing.
<strong>Check and list</strong> also puts your site in the
<a href="/directory" accesskey="4">directory</a> if the page passes.
Either way it takes up to a minute, with no progress bar, because that
would want JavaScript. Leave this page open and put the kettle on.</p>
<h2>Doing it yourself</h2>
<p>The same checks run on the command line, which is far less tedious
while you are still fixing things:</p>
<p>The same checks run on the command line, which is quicker while you
are still fixing things:</p>
<pre>uv run mewslint.py https://example.com/</pre>
<p>It reads local files too, so you can check a page before anyone else
sees it, and I run it on this site as well: the build refuses to produce
mews.page at all if one of its own pages stops conforming. That seemed
only fair.</p>
<p>It reads local files too, so you can check a page before publishing
it. This site's own build refuses to produce mews.page if any of its pages
stops conforming. That seemed only fair.</p>
<h2>What gets listed</h2>
<p>One page per site, the one you hand me, and nobody reviews it: if the
page passes it goes in, and that is the whole of the trust model. A listed
site is checked again every week, and one whose page stops conforming
comes out of the directory without being asked twice.</p>
<p>Fix it, submit it again, and it is back in straight away. I keep no
address to warn you at, which is the trade (no account, no password to
forget, and no warning either).</p>
<p>One page per site, the one you submit, and nobody reviews it: it
passes or it doesn't. Listed sites are rechecked every week, and there is
no address to warn you at. No account, no password to forget, and no
warning either.</p>
<hr />
<p>If the checker turns your page down over something the spec doesn't
actually say, that is a bug in my validator rather than a fault in your
page, and I would like to hear about it.</p>
<p><a href="/spec/0.1" accesskey="5">The spec</a>
· <a href="/about" accesskey="0">About this site</a></p>
<p>A page turned down over something the spec doesn't say is a bug in the
validator, not a fault in your page. Report it to the address on the
<a href="/" accesskey="0">home page</a>.</p>
<p><a href="/spec/0.1" accesskey="5">The spec</a></p>
</body>
</html>