The PDF format is one of the most successful file formats ever made. Every tax authority accepts it. Every printer understands it. Every smartphone can open it. Yet most people who use PDFs daily have no idea where the format came from, why it looks the way it does, or why some of its quirks have stuck around for thirty-plus years. The history explains the present.
The Camelot Project (1991)
The story starts at Adobe in 1991. John Warnock — Adobe co-founder — wrote an internal proposal called "The Camelot Project." The goal: a format that could capture documents in a way that printed identically on any system, on any printer.
The problem he was solving was real. In the early 1990s, sending a document from one computer to another was an adventure. Different operating systems used different fonts. Different printers had different drivers. The same file looked different in every place it landed.
Warnock's vision: a "digital paper" format that locked the layout in place. Once a document became a PDF, it would render the same on every machine, forever.
PDF 1.0 (1993)
The first public release of PDF was version 1.0 in 1993, alongside Adobe Acrobat 1.0. It was met with confusion.
The market wasn't sure why anyone would want a format that you couldn't easily edit. Acrobat 1.0 was expensive — Adobe charged $50 for the reader, which seems absurd today but was normal at the time. Adoption was slow.
Two things saved PDF:
- Adobe started giving away the reader for free in 1994.
- The internet started growing, and "send a file that looks identical everywhere" became a problem millions of people had.
By 1996, PDF was the de facto format for downloadable government forms, academic papers, and product manuals.
The PostScript heritage
PDF's design came directly from PostScript, Adobe's existing page description language. PostScript was a programming language for describing pages — a printer would run the PostScript file to produce output.
PDF kept much of PostScript's model — the same coordinate system, the same approach to fonts and graphics — but stripped out the programmatic parts. A PDF page is described declaratively: "draw this glyph at X, Y, in this font, in this colour." No loops, no conditional logic.
This is why PDFs are reliable. There's nothing to interpret, nothing to go wrong at render time. The page is a static description.
The interactive years (1996-2007)
PDF added features rapidly through the 90s and early 2000s:
- Forms (PDF 1.2, 1996). Users could fill in fields and save data.
- Hyperlinks and bookmarks (PDF 1.2). Made PDFs navigable, not just printable.
- JavaScript (PDF 1.3, 1999). Documents could now run scripts. This produced both useful interactivity and a long history of security problems.
- Encryption with AES (PDF 1.6, 2004). Real protection replaced weak RC4.
- 3D content (PDF 1.6). PDFs could embed manipulable 3D models. Mostly unused, occasionally useful in engineering.
- Embedded video and audio (PDF 1.7, 2006). Mostly removed in later readers for security reasons.
This was PDF's "everything format" era. The vision: a single PDF could be a multimedia, interactive, signed, encrypted, accessible package replacing entire web experiences.
In retrospect, this was overreach. JavaScript in PDFs introduced enormous attack surface. Embedded media often didn't play. Most readers stripped most of it out for security.
The standardisation era (2008)
In 2008, Adobe handed PDF over to the International Organisation for Standardisation (ISO) as ISO 32000-1. PDF became a true open standard.
This mattered. Before standardisation, "PDF" was effectively whatever Adobe Acrobat produced. After standardisation, anyone could implement a fully compliant PDF tool without permission. The result: an explosion of PDF tools — open source, commercial, browser-based, mobile — all interoperating reasonably well.
ISO has since released:
- ISO 32000-2 (PDF 2.0) in 2017, modernising encryption and accessibility.
- PDF/A (ISO 19005), the archival subset.
- PDF/X (ISO 15930), the print-production subset.
- PDF/UA (ISO 14289), the universal accessibility profile.
Each subset locks down features that cause problems in specific contexts (no JavaScript in PDF/A, embedded fonts required in PDF/X, accessibility tags required in PDF/UA).
Why PDF is still relevant
PDF was supposed to be killed by:
- The web (which renders documents directly in the browser).
- HTML5 (which can do interactive documents).
- Modern word processors (which sync between devices).
- EPUB (which reflows for any screen size).
None of these replaced it. PDF survives because:
- It locks layout in place. No other format does this as completely.
- Printing works. Everyone who needs to print to paper, print legal documents, or send to a print shop reaches for PDF.
- Signatures work. Digital signatures on PDF have legal weight in most jurisdictions. See signing a PDF online.
- Archival works. PDF/A is the recommended format for long-term retention by national archives worldwide.
- It works offline. A PDF is a single file that doesn't depend on a server, a browser, or anything but the file itself.
The PDF/A and PDF/X subsets
Two subsets matter for specific use cases:
- PDF/A is the archival subset. It bans features that don't survive long-term: external content references, JavaScript, transparency. It requires embedded fonts and accessibility tags. Good for legal records, government documents, tax filings. See PDF/A explained.
- PDF/X is the print production subset. It requires colour management, embedded fonts, and resolved overprints. Mandatory for many commercial printers.
Most users never need either. When you do, the constraints are usually obvious — your tax authority or your printer asks for it specifically.
The accessibility journey
PDFs have an uneasy relationship with accessibility. The format is fundamentally page-shaped, while screen readers want a flow of content. PDF accessibility tags (tags explained) bridge the gap by adding a semantic structure (this is a heading, this is a list, this is a table) on top of the visual layout.
PDF/UA (accessible PDF) requires this structure. Most regulated industries — government, education, finance — now require PDF/UA compliance for public-facing documents.
What's coming next
Active work on PDF includes:
- Better accessibility — PDF 2.0 cleans up many of the legacy issues.
- Cleaner document semantics — making PDF easier for AI tools to parse and understand.
- Encrypted media for digital signatures — strengthening the cryptographic core.
- Streaming-friendly PDFs — letting browsers start rendering before the full file downloads.
The format is still evolving, slowly. Its boring stability is exactly what its users want.
Conclusion
PDF started as a niche solution to a genuine problem (documents that print the same everywhere) and grew into a universal format because nothing else met the same need. The quirks that frustrate modern users — fixed pages, awkward editing, complex accessibility — are mostly there because the format prioritises rendering fidelity above all else. After thirty years, that priority hasn't changed, and the format hasn't either, except in carefully chosen ways. For more on its modern uses and trade-offs, see PDF vs Word and PDF vs HTML.