Advertisement

Blog

A short history of the PDF

August 27, 2026

In 1991, printing a document on someone else's computer was a gamble. Fonts you didn't have were silently swapped for ones you did, or came out as boxes. Layouts shifted, tables broke, and the safest way to send a formatted document across town was still a fax machine. John Warnock, Adobe's co-founder, wrote an internal memo about this. He called the project Camelot. The goal: a file that displays and prints the same way on any machine, regardless of what software made it or what fonts the receiving computer happens to own.

Adobe was well placed to try. Its first product, PostScript, was already the language laser printers spoke. But PostScript was built to drive printers rather than to be read on screen. A PostScript file is a program that a printer executes; you can't jump to page 40 without executing the first 39, and files were heavy for the modems of the day. Camelot kept the parts of PostScript that describe a page, threw out the programming language around them, and added compression, random access to any page, and a way to carry fonts inside the file. The result shipped in June 1993 as the Portable Document Format, PDF 1.0, alongside a program for making and viewing these files: Acrobat.

The years when it nearly failed

Adoption came slowly at first. Adobe charged for everything, including the reader, which cost around fifty dollars, at a time when a format war was on. WordPerfect (and later Novell) had Envoy, No Hands Software had Common Ground, and each faced the same chicken-and-egg problem: nobody wanted to send files their recipients couldn't open, and nobody wanted to install a viewer for a format they never received.

Two things broke the deadlock. Adobe made the reader free in 1994, which took care of the recipient. Then the US Internal Revenue Service began distributing tax forms as PDFs, and millions of people suddenly had a concrete reason to install the viewer. Once enough readers existed, publishing a PDF stopped being a risk, and the competitors faded within a few years.

The web was the bigger threat. By the mid-90s HTML was free, open and everywhere, and for a while fixed-layout documents looked like a dead end. But the two formats turned out to want different things. HTML reflows, which suits a screen of unknown size and ruins a contract, an invoice, or anything that has to print the same way twice. The web became the medium for reading; PDF stayed the medium for paperwork, and that division has held.

What the versions added

The format accumulated capabilities in layers, most of which you can date by the problem they solved. Interactive form fields arrived in 1996 with PDF 1.2, turning the printed form into something you could fill on screen, a story painful enough to get its own article. Encryption and digital signatures came for the businesses that wanted contracts. PDF 1.4 in 2001 added transparency for designers. It also added tagged PDF, an optional layer of structure that tells software what's a heading and what order to read things in. That layer is the foundation of PDF accessibility to this day. Specialised profiles split off for archiving, printing and accessibility.

The biggest change was to who controlled the format. In 2007 Adobe released the full PDF 1.7 specification for standardization, and in 2008 ISO published it as ISO 32000-1. Adobe granted a royalty-free licence to the patents needed to implement it, and the format's evolution moved to a standards committee. PDF became an open standard rather than one company's file format. PDF 2.0 followed in 2017, mostly tightening and clarifying rather than reinventing. Microsoft launched a competitor, XPS, alongside Windows Vista; it is remembered mainly by people who once accidentally printed to it.

Why it's still here

Plenty of 1993 software is gone, and the PDF survived on merit. The format promised that the page you send is the page they see, and it has kept that promise for three decades, across every operating system, printer and phone. Fonts travel inside the file, layout is baked in, and a PDF made twenty years ago opens today, which matters enormously to courts, archives and anyone whose documents outlive their software.

The same promise explains the familiar frustrations. A layout fixed to the page is hard to rearrange after the fact; text nailed to coordinates copies out strangely; a fixed page reads badly on a phone. A format that guarantees a page looks identical everywhere has to fix the layout in place, and fixed layouts are awkward to edit. That trade is covered separately, because understanding it makes every PDF tool easier to choose.

Warnock's memo assumed the tools would be desktop software. Rendering, parsing and rebuilding now all fit in a browser tab, which is how this site works: you can edit a PDF or merge two of them without installing anything.

Found this useful? The tools and the writing are free and stay that way. ☕ Buy us a coffee