Most conference proceedings
have PDF quality defects
Blank equations, missing fonts, non-searchable text, and almost no archival compliance. Our research shows these problems are the norm.
The problem
We scanned 3,237 conference papers from 431 DOI-registered proceedings publishers on Crossref (2024–2025 publications, mid-size organizations publishing 150–500 articles). For each paper we ran automated compliance checks (margins, font embedding, page size, equation integrity) using pdfinfo, pdffonts, pdftotext, and mutool. PDF/A conformance was checked both at the paper level and, where available, at the assembled-volume level. The results are consistent across engineering, computer science, humanities, medicine, and natural sciences.
The most common defects
| Defect | Impact | Common cause |
|---|---|---|
| Margin violations | Content cut off in print; inconsistent appearance across papers | Authors editing the template page setup; export scaling differences |
| Non-embedded fonts | Text renders as blank or substituted on other machines | Partial font embedding; system fonts referenced by name instead of bundled |
| Blank or corrupt equations | Readers see empty space or garbage characters where formulas should be | Word equation editor PDF export bug; font encoding loss during conversion |
| Non-searchable text | Paper invisible to Google Scholar; cannot copy-paste text | Missing Unicode mappings in font encoding |
| Inconsistent page sizes | Print imposition fails; pages shift between A4 and Letter | Mixed author locales (US vs. European default page sizes) |
The blank equation problem
The most damaging defect is invisible until a reader opens the paper. When an author writes an equation in Word's built-in equation editor and exports to PDF, the equation sometimes disappears entirely, replaced by a line number like "(1)" with nothing before it, or by a string of garbage characters.
This happens because the PDF export converts equations to font-mapped glyphs. When the encoding table is missing or corrupt, the renderer has nothing to display. The equation is gone and cannot be recovered. The only fix is re-typesetting from the source document.
Real-world impact: In one proceedings volume of over a hundred papers we checked, roughly one paper in eight had blank or corrupt equations. None had been caught before publication. Read the detection recipe →
Most publishers have no quality gate
The typical proceedings workflow looks like this: authors submit camera-ready PDFs, the organizer collects them, and they are posted online unchanged. No one checks whether fonts are embedded, equations render correctly, or pages are the right size. Even when a post-processing step exists (compression, page stamping), it does not inspect the content. There is no quality gate.
Almost no proceedings are archival-ready
PDF/A is the ISO standard for long-term archival of electronic documents. Libraries, institutional repositories, and funders increasingly require it. Yet fewer than 3% of proceedings papers declare any PDF/A conformance, and among the publishers that produce assembled proceedings volumes, none are PDF/A compliant.
Conference output that is defective today is also at risk of becoming inaccessible as archival requirements tighten.
What a proper quality gate catches
Automated PDF compliance checking at the point of submission, before the paper enters the proceedings, prevents every defect in the table above. The checks are mechanical and take seconds per paper.
- Page size consistency: all pages match the template format
- Margin boundaries: text block stays within the printable area
- Font embedding: every font is bundled, not referenced by name
- Text searchability: Unicode mappings present for all glyphs
- Equation integrity: no orphaned equation numbers, no garbage glyphs
- Image resolution: figures print at 300+ DPI
When a paper fails, the author gets specific feedback and can fix it while the paper is still fresh instead of weeks later during a panicked email campaign.
How Ready.Pub solves this
Ready.Pub, the proceedings tool in the texlode ecosystem, is the production layer between "we have 100 camera-ready PDFs of varying quality" and "we have a compliant, consistently formatted proceedings volume with DOIs." It works alongside your existing submission workflow rather than replacing it.
- Automatic compliance checking on upload: authors see pass/fail with specific error messages and can fix problems immediately
- Mixed-source handling: Word and LaTeX authors submit side by side and get the same quality checks
- Production-quality assembly: uniform headers, page numbers, and formatting across the entire volume, with PDF bookmarks and metadata
- DOI registration: metadata export and DOI minting are built in
- PDF/A and PDF/UA by default: every volume and every offprint is archival-ready and accessible without extra steps
Ready.Pub delivers production-vendor quality at self-service speed and is already in production for IAHR in 2026: HIC 2026 (Zaragoza), the 9th IAHR Europe Congress (Luleå) and ISE 2026 (Lausanne). Every volume is print-ready, and the ISE 2026 papers and HIC 2026 full papers carry Crossref DOIs. Learn more about proceedings support →
Get notified at launch
Ready.Pub ships with the initial texlode release in October 2026.