Twenty Years of Pandoc Twenty Years of Pandoc On August 3, 2006, I uploaded the first version of pandoc to my website, releasing it under the free GPL license. Pandoc 0.1 consisted of about 3000 line
By Coderz Club · 2026-08-04 · Tags: html
Twenty Years of Pandoc
Twenty Years of Pandoc On August 3, 2006, I uploaded the first version of pandoc to my website, releasing it under the free GPL license. Pandoc 0.1 consisted of about 3000 lines of Haskell code, with no dependencies aside from GHC’s standard library. It could convert Markdown, reStructuredText, HTML, and LaTeX documents into any of these formats, plus RTF or S5. I had no idea at the time that this would just be the first of over two hundred releases over the next twenty years; that the project would become the most popular program written in Haskell; that I would spend countless hours on bug-fixes, improvement, and project management; that I would collaborate with programmers in many other countries; that pandoc would come to support over fifty document formats; that it would allow automatic generation of citations and bibliographies; that it would become integrated into academic writing tools like Quarto and Jupyter Notebook; that it would be installed on millions of computers around the world. How did this happen? I want to take advantage of pandoc’s birthday to tell the story of the project, as best I can remember it. John MacFarlane August 2, 2026 Prehistory People often ask: Why is pandoc written in Haskell? There could have been good answers to this question: Haskell is a very good language for writing this kind of application. But in fact, I didn’t decide to write a document converter, then decide to use Haskell for it. I decided to use Haskell, and then decided to write a document converter in it. I had heard about Haskell from the blog of a philosophical logician friend, Greg Restall. Of an introductory book on Haskell, he said: “I’m glad that this wasn’t the textbook in my introductory computer science course, long ago in 1986. If it were, I may have fallen in love with computing and never become a philosopher” (consequently.org). Intrigued by this (and not heeding Restall’s warning about the potential effects on my future philosophical productivity), I read A Gentle Introduction to Haskell to get a basic understanding of the language. But the only way to really learn a programming language is to write something in it. I saw that Haskell was good for writing parsers and compilers, and it came with a really nice parser combinator library (parsec), so I decided to write a Markdown parser. At that time, there were implementations of Markdown in Perl, Python, Ruby, and PHP; they all transformed Markdown directly to HTML through a sequence of regex transformations. Pandoc took a different approach. It parsed the Markdown using parser combinators and produced a real abstract syntax tree (AST), which it could then render to HTML or another format. This was a more reliable architecture (avoiding many quirks of the regex versions). It was also a more extensible one: by writing N parsers (“readers”) and M renderers (“writers”), one could support N × M conversions. Soon I added a reader for reStructuredText, because I kept a lot of my lecture notes and handouts in that format. And I added a writer for LaTeX, because I wanted to be able to produce PDFs. Then I added a writer for Markdown, so I could start to convert my reStructuredText notes to Markdown. And from there the project just snowballed. Thus, a project that started out as nothing more than the product of procrastination was nurtured by the joy of writing in Haskell and by its increasing usefulness for my own academic work. First releases (2006–8) In August 3, 2006, I decided to make the source code available on my website. By now pandoc supported HTML, LaTeX, RST, and Markdown as input and output formats, and RTF as an output format; also PDF via LaTeX. The first release I made no attempts to advertise the project, other than emailing two friends. This was before social media (which I’ve never used anyway), before GitHub, and before Hackage, the Haskell package repository. But apparently some people stumbled across it on my website and started using it. In October I was contacted by a Turkish developer, Recai Oktaş, who was trying to get certified as a Debian developer and wanted to package pandoc for Debian linux. So I worked with him to do that. This was a great learning experience for me and it greatly increased the visibility of the project. During 2007, I continued to improve pandoc, largely guided by my own needs. Version 0.3 added the DocBook writer and the now-standard syntax for footnotes in Markdown. Version 0.4 added support for Markdown tables, definition lists, super/subscript, strikeout, and enhanced ordered lists, as well as writers for groff man pages and ConTeXt. This was the first release to go on the Hackage Haskell package repository, which was started in 2007. The Hackage archive and the new cabal-install tool, which automatically resolved and fetched dependencies, opened up the possibility of depending on external packages. Pandoc 1 (2008–17) Pandoc 1.0 was released in September 2008, with new writers f
Twenty Years of Pandoc On August 3, 2006, I uploaded the first version of pandoc to my website, releasing it under the free GPL license. Pandoc 0.1 consisted of about 3000 lines of Haskell code, with no dependencies aside from GHC’s standard library. It could convert Markdown, reStructuredText, HTML, and LaTeX documents into any of these formats, plus RTF or S5. I had no idea at the time that this would just be the first of over two hundred releases over the next twenty years; that the project would become the most popular program written in Haskell; that I would spend countless hours on bug-fixes, improvement, and project management; that I would collaborate with programmers in many other countries; that pandoc would come to support over fifty document formats; that it would allow automatic generation of citations and bibliographies; that it would become integrated into academic writing tools like Quarto and Jupyter Notebook; that it would be installed on millions of computers around the world. How did this happen? I want to take advantage of pandoc’s birthday to tell the story of the project, as best I can remember it. John MacFarlane August 2, 2026 Prehistory People often ask: Why is pandoc written in Haskell? There could have been good answers to this question: Haskell is a very good language for writing this kind of application. But in fact, I didn’t decide to write a document converter, then decide to use Haskell for it. I decided to use Haskell, and then decided to write a document converter in it. I had heard about Haskell from the blog of a philosophical logician friend, Greg Restall. Of an introductory book on Haskell, he said: “I’m glad that this wasn’t the textbook in my introductory computer science course, long ago in 1986. If it were, I may have fallen in love with computing and never become a philosopher” (consequently.org). Intrigued by this (and not heeding Restall’s warning about the potential effects on my future philosophical productivity), I read A Gentle Introduction to Haskell to get a basic understanding of the language. But the only way to really learn a programming language is to write something in it. I saw that Haskell was good for writing parsers and compilers, and it came with a really nice parser combinator library (parsec), so I decided to write a Markdown parser. At that time, there were implementations of Markdown in Perl, Python, Ruby, and PHP; they all transformed Markdown directly to HTML through a sequence of regex transformations. Pandoc took a different approach. It parsed the Markdown using parser combinators and produced a real abstract syntax tree (AST), which it could then render to HTML or another format. This was a more reliable architecture (avoiding many quirks of the regex versions). It was also a more extensible one: by writing N parsers (“readers”) and M renderers (“writers”), one could support N × M conversions. Soon I added a reader for reStructuredText, because I kept a lot of my lecture notes and handouts in that format. And I added a writer for LaTeX, because I wanted to be able to produce PDFs. Then I added a writer for Markdown, so I could start to convert my reStructuredText notes to Markdown. And from there the project just snowballed. Thus, a project that started out as nothing more than the product of procrastination was nurtured by the joy of writing in Haskell and by its increasing usefulness for my own academic work. First releases (2006–8) In August 3, 2006, I decided to make the source code available on my website. By now pandoc supported HTML, LaTeX, RST, and Markdown as input and output formats, and RTF as an output format; also PDF via LaTeX. The first release I made no attempts to advertise the project, other than emailing two friends. This was before social media (which I’ve never used anyway), before GitHub, and before Hackage, the Haskell package repository. But apparently some people stumbled across it on my website and started using it. In October I was contacted by a Turkish developer, Recai Oktaş, who was trying to get certified as a Debian developer and wanted to package pandoc for Debian linux. So I worked with him to do that. This was a great learning experience for me and it greatly increased the visibility of the project. During 2007, I continued to improve pandoc, largely guided by my own needs. Version 0.3 added the DocBook writer and the now-standard syntax for footnotes in Markdown. Version 0.4 added support for Markdown tables, definition lists, super/subscript, strikeout, and enhanced ordered lists, as well as writers for groff man pages and ConTeXt. This was the first release to go on the Hackage Haskell package repository, which was started in 2007. The Hackage archive and the new cabal-install tool, which automatically resolved and fetched dependencies, opened up the possibility of depending on external packages. Pandoc 1 (2008–17) Pandoc 1.0 was released in September 2008, with new writers f