The Rise and Rise of JSON

An interesting little story about how JSON rose to its prominence today. It’s probably an illustration of the rule of least power (“choose the least powerful language suitable for a given purpose”). In fact, the article’s author states as much:

my own hunch is that people disliked XML because it was confusing, and it was confusing because it seemed to come in so many different flavors.

The author goes on to say:

XML was designed to be a system that could be used by everyone for almost anything imaginable. To that end, XML is actually a meta-language that allows you to define domain-specific languages for individual applications...And yet here was JSON, a format offering no benefits over XML except those enabled by throwing out the cruft that made XML so flexible.

The simplicity of JSON, which I’m sure is often ridiculed, is quite fascinating in comparison to XML:

The first lines of a typical XML document establish the XML version and then the particular sub-language the XML document should conform to. That is a lot of variation to account for already, especially when compared to JSON, which is so straightforward that no new version of the JSON specification is ever expected to be written.

There were a few little historical tidbits I found interesting in this story. For example, when Douglas Crockford first implemented what would become “JavaScript object notation” by embedding a <script> tag in HTML, he ran into a problem where dynamically written keys could conflict with reserved words in JavaScript, so he just required all key names to be quoted. JSON requires quoted key names to this day.

There’s also the story about the name and the spec:

Crockford and Morningstar...wanted to name their format “JSML”, for JavaScript Markup Language, but found that the acronym was already being used for something called Java Speech Markup Language. So they decided to go with “JavaScript Object Notation”, or JSON. They began pitching it to clients but soon found that clients were unwilling to take a chance on an unknown technology that lacked an official specification. So Crockford decided he would write one.

On, and there was so linked reading in the article, some of which I followed. I liked this comment on XML, which put into words my feelings based on experience with XML:

I spend a disproportionate amount of my time wading through an endless sea of angle brackets and verbose tags desperately searching for the vaguest hint of actual information