Skip to content

Say what is checked, and what is not #20

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
DenisDrobyshev merged 1 commit into from
Sep 18, 2026
Merged

Say what is checked, and what is not #20

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
297 changes: 297 additions & 0 deletions checks/index.html
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,297 @@
<!DOCTYPE html>
<html lang="en" prefix="og: https://ogp.me/ns#">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>What is checked — DrobyshevDev</title>
<meta name="theme-color" content="#fdfcfa" media="(prefers-color-scheme: light)">
<meta name="theme-color" content="#131316" media="(prefers-color-scheme: dark)">
<meta name="description" content="Every number DrobyshevDev publishes, the file that holds it to the code, and the checks that run on a schedule — including the claims nothing checks yet.">
<link rel="canonical" href="https://drobyshevdev.github.io/checks/">
<link rel="alternate" hreflang="en" href="https://drobyshevdev.github.io/checks/">
<link rel="alternate" hreflang="ru" href="https://drobyshevdev.github.io/ru/checks/">
<link rel="alternate" hreflang="x-default" href="https://drobyshevdev.github.io/checks/">

<meta property="og:type" content="article">
<meta property="og:site_name" content="DrobyshevDev">
<meta property="og:locale" content="en_GB">
<meta property="og:title" content="What is checked — DrobyshevDev">
<meta property="og:description" content="Every number DrobyshevDev publishes, the file that holds it to the code, and the checks that run on a schedule — including the claims nothing checks yet.">
<meta property="og:url" content="https://drobyshevdev.github.io/checks/">
<meta property="og:image" content="https://drobyshevdev.github.io/assets/og.png">
<meta name="twitter:card" content="summary_large_image">

<link rel="icon" href="/assets/favicon.png" type="image/png">
<link rel="apple-touch-icon" href="/assets/apple-touch-icon.png">
<link rel="stylesheet" href="/assets/site.css">

<style>
/* .measure was written for the bright band, where the ink is near-black and
the borders are mixed against it. This page uses the default band, so the
header, the caption and the rules take the page's own colours instead --
otherwise the header row is #16160f on a dark background, which is to say
invisible. Scoped here rather than changed in site.css, because the bright
band still wants the original. */
.measure th { color: var(--ink-2); border-bottom-color: var(--line-2); }
.measure caption { color: var(--ink-3); }
.measure td { border-bottom-color: var(--line); }
/* On the front page the emphasised column is a measured number and the last
column is a footnote. Here the last column is the file that holds the claim
-- the point of the table -- so the emphasis moves there and the middle
column, which only says where the claim appears, goes quiet. */
.measure td:nth-child(2) { font-weight: 400; }
.measure td:last-child { font-weight: 600; opacity: 1; color: var(--ink); }
</style>

<script>
// Applied before first paint, so the page does not flash the wrong theme.
try {
var t = localStorage.getItem("theme");
if (t === "dark" || t === "light") document.documentElement.setAttribute("data-theme", t);
} catch (e) {}
</script>

</head>
<body>
<a class="skip-link" href="#main">Skip to content</a>

<header class="site-header">
<div class="wrap">
<a class="brand" href="/">
<img class="mark" src="/assets/mark.png" alt="" width="22" height="22">
DrobyshevDev
</a>

<nav class="nav" id="primary-nav" aria-label="Primary">
<a href="/#projects">Projects</a>
<a href="/#notes">Course</a>
<a href="/#method">How we work</a>
<a href="/checks/" aria-current="page">Checks</a>
<a class="ext" href="https://github.com/DrobyshevDev" rel="noopener">GitHub</a>
</nav>

<div class="nav-tools">
<a class="lang-switch" href="/ru/checks/">Русский</a>
<button class="icon-btn nav-toggle" type="button" aria-expanded="false" aria-controls="primary-nav" aria-label="Open menu">
<svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" aria-hidden="true"><path d="M4 7h16M4 12h16M4 17h16"/></svg>
</button>
</div>
</div>
</header>

<main id="main" class="band" style="padding-top:48px">
<div class="wrap">

<p class="eyebrow"><span class="dot"></span> Checks</p>
<h1 style="margin:14px 0 18px">What is checked, and what is not</h1>
<p class="lead" style="margin-bottom:32px">
Every project here publishes numbers. This page says, for each of them, which file holds it
to the code and what happens when the two disagree — and then lists the claims nothing
checks yet, because a page like this is worth reading only if the second list is real.
</p>

<nav class="chips" aria-label="Sections" style="margin-bottom:40px">
<a class="chip chip-accent" href="#pinned">Numbers pinned to code</a>
<a class="chip chip-accent" href="#scheduled">What runs on a schedule</a>
<a class="chip" href="#unchecked">What nothing checks yet</a>
</nav>

<!-- ==================== Pinned ==================== -->
<section id="pinned" class="legal">
<h2>Numbers pinned to code</h2>
<p>
A number in prose rots quietly; a failing test does not. Each row below is a claim that
appears in a README, a citation file or on this site, and the file that makes it
answerable to the package that ships.
</p>

<div class="measure" style="margin-top:24px">
<table>
<caption>
Every claim here fails a build when it stops being true. The right-hand column is the
file to read if you want to know exactly what is being asserted.
</caption>
<thead>
<tr><th>Claim</th><th>Where it is stated</th><th>What holds it</th></tr>
</thead>
<tbody>
<tr>
<td>decisionrl ships 32 algorithms, 24 environments, 9 of them applied</td>
<td>README prose and badges, <code>CITATION.cff</code>, <code>.zenodo.json</code>, <code>pyproject.toml</code>, this site</td>
<td><code>tests/test_documented_counts.py</code></td>
</tr>
<tr>
<td>The applied tasks where a learned policy beats the classical baseline still do</td>
<td>decisionrl README results table</td>
<td><code>.github/workflows/verify.yml</code>, nightly, 3 seeds</td>
</tr>
<tr>
<td>decisionrl line coverage does not fall below 80%</td>
<td>this site, Codecov badge</td>
<td><code>fail_under</code> in <code>pyproject.toml</code></td>
</tr>
<tr>
<td>glia needs no dependency outside the standard library</td>
<td>glia README badge, this site</td>
<td><code>dependencies = []</code>, offline test suite</td>
</tr>
<tr>
<td>glia line coverage does not fall below 90%</td>
<td>glia CI</td>
<td><code>--cov-fail-under=90</code></td>
</tr>
<tr>
<td>The English and Russian mlango READMEs carry the same badges and commands</td>
<td>both READMEs, and the PyPI landing page</td>
<td><code>tests/test_pipelines.py::TestReadme</code></td>
</tr>
<tr>
<td>The praxis golden set is the size the README says it is</td>
<td>praxis README, both languages</td>
<td><code>tests/test_eval.py</code></td>
</tr>
<tr>
<td>praxis depends on this organisation's own packages, not on strangers' names</td>
<td><code>pyproject.toml</code> extras</td>
<td><code>tests/test_packaging.py</code></td>
</tr>
<tr>
<td>This site makes no third-party request of any kind</td>
<td>the legal page</td>
<td><code>scripts/check_site.py</code>, on every change</td>
</tr>
<tr>
<td>The figures on the front page match the repositories they describe</td>
<td>the front page, both languages</td>
<td><code>scripts/check_figures.py</code>, weekly</td>
</tr>
</tbody>
</table>
</div>
</section>

<!-- ==================== Scheduled ==================== -->
<section id="scheduled" class="legal" style="margin-top:56px">
<h2>What runs on a schedule</h2>
<p>
A check that runs only on a commit stops telling the truth the moment the world moves
without one — a link rots, a model is withdrawn, a dependency is patched. These run
whether or not anyone pushed.
</p>

<div class="measure" style="margin-top:24px">
<table>
<caption>
Scheduled checks write into one standing issue each and close it themselves when the
problem resolves, so an open issue means an open problem rather than a backlog of
notifications.
</caption>
<thead>
<tr><th>Check</th><th>How often</th><th>When it fails</th></tr>
</thead>
<tbody>
<tr>
<td>Re-verify decisionrl's published results</td>
<td>nightly</td>
<td>the build goes red — the claim in the README is no longer reproducible</td>
</tr>
<tr>
<td>OpenSSF Scorecard: signed releases, pinned actions, patch latency</td>
<td>weekly, per repository</td>
<td>the grade drops, publicly, and findings land in code scanning</td>
</tr>
<tr>
<td>CodeQL static analysis</td>
<td>on every change and weekly</td>
<td>an alert opens in the repository's security tab</td>
</tr>
<tr>
<td>Outbound links on this site and in the research notes</td>
<td>weekly</td>
<td>one standing issue lists what broke</td>
</tr>
<tr>
<td>The figures on this site against their sources</td>
<td>weekly</td>
<td>one standing issue names the page, what it says, and what the source says</td>
</tr>
<tr>
<td>Organisation-wide maintenance triage</td>
<td>weekly</td>
<td>one issue lists what needs a decision, ordered by that rather than by repository</td>
</tr>
</tbody>
</table>
</div>

<p style="margin-top:24px">
Test suites themselves run on Linux, macOS and Windows, across every interpreter each
package claims to support, because "works on my machine" is the failure mode all of this
exists to rule out.
</p>
</section>

<!-- ==================== Unchecked ==================== -->
<section id="unchecked" class="legal" style="margin-top:56px">
<h2>What nothing checks yet</h2>
<p>
This is the part of the page that makes the rest worth reading. These claims are published
and are not held by anything; they are listed here rather than quietly counted among the
checked ones.
</p>

<dl class="facts" style="max-width:44rem;margin-top:20px">
<div>
<dt>praxis retrieval metrics</dt>
<dd>
recall@5 0.92 and MRR 0.94 on the full corpus are stated in prose. They depend on
which models are installed, so holding them needs a scheduled re-run like
decisionrl's, not a unit test. The golden set's size and a recall floor of 0.7 are
pinned; the headline figures are not.
</dd>
</div>
<div>
<dt>The praxis corpus against the official publication</dt>
<dd>
The six codes shipped in the repository are transcriptions from Wikisource, marked in
the corpus as pending a check against pravo.gov.ru. The parser for that source is how
an edition gets confirmed, and it has not been run against every article.
</dd>
</div>
<div>
<dt>Anything about adoption</dt>
<dd>
No download counts, no star counts, no user numbers appear anywhere on this site,
because none of them would survive the standard the rest of this page sets.
</dd>
</div>
</dl>

<p style="margin-top:24px">
Each of these is a gap to close where it lives, not something to reword here. When one
closes, it moves to the first table.
</p>
</section>

<p class="lm-fine" style="margin-top:48px;color:var(--text-faint);font-size:.85rem">
This page describes the checking regime rather than its current status; for the state of any
individual run, the Actions tab of the repository in question is the source.
History of this page —
<a href="https://github.com/DrobyshevDev/DrobyshevDev.github.io/commits/main/checks/index.html" rel="noopener">in the repository</a>.
</p>

</div>
</main>

<footer class="site-footer">
<div class="wrap">
<div class="footer-bottom" style="border-top:0;padding-top:0">
<p>© <span data-year>2026</span> DrobyshevDev. Source released under MIT and Apache-2.0.</p>
<p><a href="/">← Home</a></p>
</div>
</div>
</footer>

<script src="/assets/site.js" defer></script>
</body>
</html>
2 changes: 2 additions & 0 deletions index.html
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,6 +132,7 @@
<a href="#evidence">Evidence</a>
<a href="#notes">Notes</a>
<a href="#method">Method</a>
<a href="/checks/">Checks</a>
<a href="https://github.com/DrobyshevDev" rel="noopener">GitHub</a>
</nav>

Expand Down Expand Up @@ -649,6 +650,7 @@ <h4>Organisation</h4>
<ul role="list">
<li><a href="https://github.com/DrobyshevDev" rel="noopener">GitHub</a></li>
<li><a href="#method">Method</a></li>
<li><a href="/checks/">What is checked</a></li>
<li><a href="#licensing">Licensing</a></li>
<li><a href="mailto:denidrob138@gmail.com?subject=DrobyshevDev%20—%20enquiry">Contact</a></li>
<li><a href="/ru/" hreflang="ru" lang="ru">Русская версия</a></li>
Expand Down
1 change: 1 addition & 0 deletions legal/index.html
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,6 +45,7 @@
<a href="/ru/#projects">Проекты</a>
<a href="/ru/#notes">Курс</a>
<a href="/ru/#method">Как мы работаем</a>
<a href="/ru/checks/">Проверки</a>
<a class="ext" href="https://github.com/DrobyshevDev" rel="noopener">GitHub</a>
</nav>

Expand Down
Loading
Loading