Add ArtiFinder discovered-artifact integration (website)

.github/workflows/pr-checks.yml

@@ -85,7 +85,10 @@ jobs:
           for data in "$DATA_DIR"/*_conf_authors.json; do
             [ -f "$data" ] && validate "$data" "$SCHEMA_DIR/conf_authors.schema.json"
           done
-          for data in "$DATA_DIR"/*_institution_rankings.json; do
+          # Only validate files that match the full InstitutionRanking schema.
+          # Per-conference *_institution_rankings.json variants intentionally
+          # omit fields such as top_authors and should not be validated here.
+          for data in "$DATA_DIR"/systems_institution_rankings.json "$DATA_DIR"/security_institution_rankings.json; do
             [ -f "$data" ] && validate "$data" "$SCHEMA_DIR/institution_rankings.schema.json"
           done
 

src/_data/navigation.yml

@@ -10,6 +10,8 @@ main:
     url: institution_rankings.html
   - title: Systems vs. Security
     url: overview.html
+  - title: ArtiFinder Discovery
+    url: artifinder.html
   - title: Geographic Statistics
     url: statistics/
   - title: Profiles

src/_includes/head/custom.html

@@ -50,6 +50,7 @@
 <link rel="stylesheet" href="{{ '/assets/css/reprodb-table.css' | relative_url }}">
 <link rel="stylesheet" href="{{ '/assets/css/reprodb-controls.css' | relative_url }}">
 <link rel="stylesheet" href="{{ '/assets/css/reprodb-search.css' | relative_url }}">
+<link rel="stylesheet" href="{{ '/assets/css/reprodb-artifinder.css' | relative_url }}">
 <script src="{{ '/assets/js/reprodb-utils.js' | relative_url }}"></script>
 
 <!-- Tabulator 6 — sortable / paginated / filterable tables -->

src/about.md

@@ -20,6 +20,7 @@ AE results and committee lists live in the upstream community sites. Contribute
 
 - [sysartifacts.github.io](https://github.com/sysartifacts/sysartifacts.github.io) — systems conferences (OSDI, SOSP, ATC, EuroSys, FAST, …)
 - [secartifacts.github.io](https://github.com/secartifacts/secartifacts.github.io) — security conferences (USENIX Security, CCS, NDSS, S&P, …)
+- [ArtiFinder-Data](https://github.com/DistriNet/ArtiFinder-Data) — automatically discovered artifact links (open to manual corrections via pull requests). These links are unverified, carry no badges, and never affect ReproDB scores; see the [ArtiFinder discovery page](/artifinder.html) and [methodology](/methodology.html#artifinder-discovered-artifacts).
 
 ### Suggest New Analyses
 

src/artifinder.md

@@ -0,0 +1,97 @@
+---
+title: "ArtiFinder — Automatically Discovered Artifacts"
+permalink: /artifinder.html
+---
+
+[ArtiFinder](https://github.com/DistriNet/ArtiFinder) scrapes conference papers directly and identifies links to their artifacts, independently of any artifact-evaluation (AE) process. This page summarises the ArtiFinder links that ReproDB ingests from the [ArtiFinder-Data](https://github.com/DistriNet/ArtiFinder-Data) repository and how they have grown over time.
+
+ArtiFinder and its underlying longitudinal study are described by Vansteenhuyse et al., *[Not All Those Who Share Are Lost: Analyzing 25 Years of Cybersecurity Artifact Sharing Practices Through Automated Discovery](https://github.com/DistriNet/ArtiFinder)* (USENIX Security 2026).
+
+{: .artifinder-note}
+**These links are _not_ manually verified.** ArtiFinder-discovered artifacts carry **no badges** and are **excluded from every ReproDB score** (artifact rate, reproducibility rate, combined score, institution and author rankings). They are shown for discovery purposes only, marked with an <img src="{{ '/assets/images/artifinder-logo.svg' | relative_url }}" alt="" style="height:1em;vertical-align:-0.15em"> **Artifinder** sign wherever they appear in search and profiles. The one exception is repository statistics: when ArtiFinder finds a GitHub repository for a paper that _did_ go through AE, that repository may be counted in the repository stats.
+
+{% if site.data.artifinder_summary %}
+
+## High-Level Summary
+
+<div class="rdb-cards">
+  <div class="rdb-card">
+    <div class="rdb-card-value">{{ site.data.artifinder_summary.total_discovered }}</div>
+    <div class="rdb-card-label">Discovered Artifacts</div>
+  </div>
+  <div class="rdb-card">
+    <div class="rdb-card-value">{{ site.data.artifinder_summary.total_matched_ae }}</div>
+    <div class="rdb-card-label">Matched to AE Papers</div>
+  </div>
+  <div class="rdb-card">
+    <div class="rdb-card-value">{{ site.data.artifinder_summary.github_count }}</div>
+    <div class="rdb-card-label">GitHub Links</div>
+  </div>
+  <div class="rdb-card">
+    <div class="rdb-card-value">{{ site.data.artifinder_summary.discovery_pct }}%</div>
+    <div class="rdb-card-label">Discovery Rate</div>
+  </div>
+  <div class="rdb-card">
+    <div class="rdb-card-value">{{ site.data.artifinder_summary.year_range }}</div>
+    <div class="rdb-card-label">Years Included</div>
+  </div>
+</div>
+
+<p class="rdb-scatter-caption">Across {{ site.data.artifinder_summary.total_papers }} scanned papers in {{ site.data.artifinder_summary.conferences | join: ", " }}, ArtiFinder discovered {{ site.data.artifinder_summary.total_discovered }} artifact links. ArtiFinder data last updated {{ site.data.artifinder_summary.data_updated }} (only editions from {{ site.data.artifinder_summary.year_range | split: "\u2013" | first }} onward are included).</p>
+
+---
+
+## Discovered Artifacts Over Time
+
+Number of ArtiFinder-discovered artifact links per year, split by whether the paper also went through artifact evaluation (AE) in ReproDB. Papers that never went through AE make up the bulk of ArtiFinder's coverage.
+
+<div class="rdb-chart-wide rdb-chart-wrap--lg">
+  <div id="afDiscoveredChart" style="width:100%;height:100%"></div>
+</div>
+
+## Discovery Rate Over Time
+
+Share of scanned papers for which ArtiFinder found an artifact link, per year. This is a coverage measure of ArtiFinder itself, not an artifact-evaluation rate.
+
+<div class="rdb-chart-wide rdb-chart-wrap--lg">
+  <div id="afRateChart" style="width:100%;height:100%"></div>
+</div>
+
+## By Conference
+
+Discovered artifact links per conference, and how many were matched to an AE paper tracked by ReproDB.
+
+<div class="rdb-chart-wide rdb-chart-wrap--lg">
+  <div id="afConfChart" style="width:100%;height:100%"></div>
+</div>
+
+## Conference Timeline Heatmap
+
+Discovered artifact links by conference and year. Darker cells indicate more discovered artifacts that year.
+
+<div class="rdb-chart-wide">
+  <div id="afHeatmap" style="height:360px"></div>
+</div>
+
+{% else %}
+
+*ArtiFinder statistics are being generated. Please check back soon.*
+
+{% endif %}
+
+<!-- Data blob: Liquid injects site data, JS reads it -->
+<script id="artifinder-data" type="application/json">
+{
+  "years": [{% for y in site.data.artifinder_by_year %}{{ y.year }}{% unless forloop.last %},{% endunless %}{% endfor %}],
+  "discovered": [{% for y in site.data.artifinder_by_year %}{{ y.discovered }}{% unless forloop.last %},{% endunless %}{% endfor %}],
+  "matched": [{% for y in site.data.artifinder_by_year %}{{ y.matched_ae }}{% unless forloop.last %},{% endunless %}{% endfor %}],
+  "totalPapers": [{% for y in site.data.artifinder_by_year %}{{ y.total_papers }}{% unless forloop.last %},{% endunless %}{% endfor %}],
+  "conferences": [
+    {% for c in site.data.artifinder_by_conference %}
+    { "name": "{{ c.name }}", "category": "{{ c.category }}", "total_papers": {{ c.total_papers }}, "discovered": {{ c.discovered }}, "matched_ae": {{ c.matched_ae }},
+      "years_data": [{% for yd in c.years %}{ "year": {{ yd.year }}, "discovered": {{ yd.discovered }} }{% unless forloop.last %},{% endunless %}{% endfor %}] }{% unless forloop.last %},{% endunless %}
+    {% endfor %}
+  ]
+}
+</script>
+<script src="{{ '/assets/js/reprodb-artifinder.js' | relative_url }}"></script>

src/assets/css/reprodb-artifinder.css

@@ -0,0 +1,105 @@
+/**
+ * reprodb-artifinder.css — styling for the ArtiFinder provenance marker
+ * (shown in search results and profile tables) and the ArtiFinder statistics
+ * page. Mirrors the ".avail-warn" tooltip pattern so the two markers feel
+ * consistent.
+ */
+
+/* ── Provenance marker (search + profile) ─────────────────────────── */
+.artifinder-tag {
+  position: relative;
+  cursor: help;
+  display: inline-flex;
+  align-items: center;
+  gap: 3px;
+  font-size: 0.8em;
+  color: #4a5aa8;
+  background: #eef1fb;
+  padding: 1px 6px;
+  border-radius: 3px;
+  border: 1px solid #c9d2f0;
+  vertical-align: middle;
+}
+.artifinder-tag .artifinder-logo {
+  width: 0.95em;
+  height: 0.95em;
+  flex: 0 0 auto;
+}
+.artifinder-tag .avail-tip {
+  display: none;
+  position: absolute;
+  bottom: 130%;
+  left: 50%;
+  transform: translateX(-50%);
+  background: #333;
+  color: #fff;
+  font-size: 0.9em;
+  font-weight: normal;
+  line-height: 1.3;
+  padding: 5px 9px;
+  border-radius: 4px;
+  width: max-content;
+  max-width: 240px;
+  white-space: normal;
+  text-align: left;
+  z-index: 100;
+  pointer-events: none;
+}
+.artifinder-tag:hover .avail-tip {
+  display: block;
+}
+
+/* An ArtiFinder-only artifact link (no AE artifact_urls alongside it). */
+.rdb-result-links .artifinder-link {
+  color: #4a5aa8;
+}
+
+/* ArtiFinder-discovered paper rows in profile tables (author + institution). */
+.af-title,
+a.af-title {
+  color: #4a5aa8;
+  font-style: italic;
+}
+
+/* ── Statistics page ──────────────────────────────────────────────── */
+.artifinder-note {
+  font-size: 0.92em;
+  color: var(--ov-muted, #666);
+  border-left: 3px solid #c9d2f0;
+  padding: 0.4em 0.9em;
+  margin: 1em 0;
+  background: rgba(200, 210, 240, 0.12);
+  border-radius: 0 4px 4px 0;
+}
+
+/* ── Dark mode ────────────────────────────────────────────────────── */
+html[data-theme="dark"] .artifinder-tag {
+  color: #aeb8ea;
+  background: #262a3d;
+  border-color: #3d445f;
+}
+html[data-theme="dark"] .rdb-result-links .artifinder-link {
+  color: #aeb8ea;
+}
+html[data-theme="dark"] .af-title,
+html[data-theme="dark"] a.af-title {
+  color: #aeb8ea;
+}
+html[data-theme="dark"] .artifinder-note {
+  border-left-color: #3d445f;
+  background: rgba(60, 68, 95, 0.25);
+}
+@media (prefers-color-scheme: dark) {
+  html:not([data-theme="light"]) .artifinder-tag {
+    color: #aeb8ea;
+    background: #262a3d;
+    border-color: #3d445f;
+  }
+  html:not([data-theme="light"]) .rdb-result-links .artifinder-link {
+    color: #aeb8ea;
+  }
+  html:not([data-theme="light"]) .artifinder-note {
+    border-left-color: #3d445f;
+    background: rgba(60, 68, 95, 0.25);
+  }
+}

src/assets/js/reprodb-artifinder.js

@@ -0,0 +1,146 @@
+/**
+ * reprodb-artifinder.js — Chart logic for the ArtiFinder discovery page.
+ *
+ * Depends on: ECharts 5 (loaded globally via head/custom.html).
+ * Reads inline JSON from <script id="artifinder-data" type="application/json">.
+ */
+(function() {
+  'use strict';
+
+  // Consistent 2-colour scheme across every chart on this page:
+  //   AF_COLOR (indigo)   = the ArtiFinder measure (discovered / rate / heatmap)
+  //   MATCHED_COLOR (orange) = the subset also matched to an AE-evaluated paper
+  var AF_COLOR = '#4a5aa8';      // ArtiFinder indigo (primary)
+  var MATCHED_COLOR = '#e08a2b'; // matched-to-AE overlap accent
+  var AF_AREA = 'rgba(74,90,168,0.12)'; // translucent AF_COLOR for area fills
+
+  document.addEventListener('DOMContentLoaded', function() {
+    var dataEl = document.getElementById('artifinder-data');
+    if (!dataEl) return;
+    var D;
+    try { D = JSON.parse(dataEl.textContent); } catch (e) { return; }
+
+    var years = D.years || [];
+
+    /* ===== 1. Discovered artifacts over time (matched vs. unmatched) ===== */
+    var discEl = document.getElementById('afDiscoveredChart');
+    if (discEl && years.length) {
+      var matched = D.matched || [];
+      var unmatched = (D.discovered || []).map(function(d, i) {
+        return Math.max(0, (d || 0) - (matched[i] || 0));
+      });
+      var chart = ReproDB.initEChart(discEl);
+      chart.setOption({
+        tooltip: { trigger: 'axis', axisPointer: { type: 'shadow' } },
+        legend: { bottom: 0, data: ['Not evaluated (no AE)', 'Matched to AE paper'] },
+        grid: { containLabel: true, left: 40, right: 20, bottom: 50, top: 30 },
+        xAxis: { type: 'category', data: years },
+        yAxis: { type: 'value', name: 'Discovered links', min: 0 },
+        series: [
+          { name: 'Not evaluated (no AE)', type: 'bar', stack: 'af', data: unmatched, itemStyle: { color: AF_COLOR } },
+          { name: 'Matched to AE paper', type: 'bar', stack: 'af', data: matched, itemStyle: { color: MATCHED_COLOR } }
+        ]
+      });
+      ReproDB.registerEChart(chart);
+    }
+
+    /* ===== 2. Discovery rate over time ===== */
+    var rateEl = document.getElementById('afRateChart');
+    if (rateEl && years.length) {
+      var rate = (D.discovered || []).map(function(d, i) {
+        var tot = (D.totalPapers || [])[i] || 0;
+        return tot ? Math.round((d / tot) * 1000) / 10 : 0;
+      });
+      var chart2 = ReproDB.initEChart(rateEl);
+      chart2.setOption({
+        tooltip: {
+          trigger: 'axis',
+          valueFormatter: function(v) { return v + '%'; }
+        },
+        grid: { containLabel: true, left: 40, right: 20, bottom: 30, top: 30 },
+        xAxis: { type: 'category', data: years },
+        yAxis: { type: 'value', name: 'Discovery rate (%)', min: 0, max: 100 },
+        series: [
+          { name: 'Discovery rate', type: 'line', data: rate, smooth: true,
+            itemStyle: { color: AF_COLOR }, lineStyle: { width: 2, color: AF_COLOR }, symbolSize: 6,
+            areaStyle: { color: AF_AREA } }
+        ]
+      });
+      ReproDB.registerEChart(chart2);
+    }
+
+    /* ===== 3. Per-conference discovered vs. matched ===== */
+    var confEl = document.getElementById('afConfChart');
+    var confs = D.conferences || [];
+    if (confEl && confs.length) {
+      var names = confs.map(function(c) { return c.name; });
+      var disc = confs.map(function(c) { return c.discovered; });
+      var mAE = confs.map(function(c) { return c.matched_ae; });
+      var chart3 = ReproDB.initEChart(confEl);
+      chart3.setOption({
+        tooltip: { trigger: 'axis', axisPointer: { type: 'shadow' } },
+        legend: { bottom: 0, data: ['Discovered', 'Matched to AE'] },
+        grid: { containLabel: true, left: 40, right: 20, bottom: 50, top: 30 },
+        xAxis: { type: 'category', data: names },
+        yAxis: { type: 'value', name: 'Links', min: 0 },
+        series: [
+          { name: 'Discovered', type: 'bar', data: disc, itemStyle: { color: AF_COLOR } },
+          { name: 'Matched to AE', type: 'bar', data: mAE, itemStyle: { color: MATCHED_COLOR } }
+        ]
+      });
+      ReproDB.registerEChart(chart3);
+    }
+
+    /* ===== 4. Conference timeline heatmap (year x conference = discovered) ===== */
+    var hmEl = document.getElementById('afHeatmap');
+    if (hmEl && confs.length && years.length) {
+      var hmConfs = confs.slice().sort(function(a, b) { return a.name.localeCompare(b.name); });
+      var confNames = hmConfs.map(function(c) { return c.name; });
+      var rawHeat = [];
+      var maxVal = 0;
+      hmConfs.forEach(function(c, ci) {
+        var yrs = {};
+        (c.years_data || []).forEach(function(yd) { yrs[yd.year] = yd.discovered; });
+        years.forEach(function(y, yi) {
+          var v = yrs[y] || 0;
+          rawHeat.push([yi, ci, v]);
+          if (v > maxVal) maxVal = v;
+        });
+      });
+
+      function afCellColor(v) {
+        var dark = ReproDB.isDark();
+        if (v === 0) return dark ? 'rgba(50,55,65,0.5)' : 'rgba(220,220,220,0.3)';
+        var t = maxVal > 0 ? v / maxVal : 0;
+        return dark
+          ? 'rgb(' + Math.round(40 + 60 * t) + ',' + Math.round(50 + 60 * t) + ',' + Math.round(90 + 120 * t) + ')'
+          : 'rgba(74,90,168,' + (0.15 + t * 0.75) + ')';
+      }
+      function afLabelColor(v) {
+        var dark = ReproDB.isDark();
+        var t = maxVal > 0 ? v / maxVal : 0;
+        return (v > 0 && t > (dark ? 0.25 : 0.6)) ? '#fff' : ReproDB.themeColors().text;
+      }
+
+      var hmChart = ReproDB.initEChart(hmEl);
+      function setAfHeatmap() {
+        var dark = ReproDB.isDark();
+        hmChart.setOption({
+          tooltip: { formatter: function(p) { return confNames[p.value[1]] + ' (' + years[p.value[0]] + '): ' + p.value[2] + ' discovered'; } },
+          grid: { containLabel: true, left: 20, right: 20, bottom: 20, top: 30 },
+          xAxis: { type: 'category', data: years, splitArea: { show: false } },
+          yAxis: { type: 'category', data: confNames, splitArea: { show: false }, inverse: true },
+          series: [{
+            type: 'heatmap',
+            data: rawHeat.map(function(d) { return { value: d, itemStyle: { color: afCellColor(d[2]) }, label: { color: afLabelColor(d[2]) } }; }),
+            label: { show: true, fontSize: 10, formatter: function(p) { return p.value[2] > 0 ? p.value[2] : ''; } },
+            itemStyle: { borderColor: dark ? '#333' : '#fff', borderWidth: 1 }
+          }]
+        });
+      }
+      setAfHeatmap();
+      ReproDB.registerEChart(hmChart);
+      ReproDB.onThemeChange(setAfHeatmap);
+    }
+  });
+})();