foreword
Welcome, dear reader, to this book about DuckDB. It feels somewhat surreal to write a foreword for this book about DuckDB because it seems like everything has happened so quickly. The world of data management systems moves slowly—software projects started in the 70s are still in strong positions on the market.
It has only been a few short years since we sat at the Joost bar in Amsterdam one evening in 2018 and decided we were going to build a new system. We had been toying with the idea previously but had been hesitant, as we knew it was a daft idea. The common wisdom is that it takes “ten million dollars” to make a new database system successful. But we decided on an equally daft plan: we would create a new kind of data management system—one that had never been built before—an in-process analytical system. Maybe the usual rules did not apply to this new kind of system. After some more beers, we had pretty much decided on the first rough draft of DuckDB’s architecture. The very next day, we started hacking.
Only a year later, in 2019, we opened up our repository and started telling people about it. We showed our first demo of DuckDB at the 2019 SIGMOD conference, coincidentally in Amsterdam. Since we co-organized the conference, we snuck stickers in the goodie bags in an early attempt at a type of viral marketing. At the same time, we also opened up the source code repository to the public. The “duck was out of the bag,” so to speak.