xBRL-CSV and xBRL-JSON
XBRL International’s new specifications were front and centre in a number of presentations in Frankfurt this week. The xBRL-CSV and the xBRL-JSON specifications are designed to help make the standard more accessible to more systems. They form part of the Open Information Model initiative, which aims overall, to simplify and modernise the standard.
In a nutshell, the two specifications provide ways to represent XBRL data in alternative syntax. The web friendly JSON syntax can be very easily consumed by a number of modern programming languages. Facts represented in JSON are self-contained, or atomic, making it easier to extract just the data that you want without having to consume an entire XML-based instance document. At XBRL International we fully expect that the JSON format will quickly become the syntax of choice for publishing XBRL data.
The CSV syntax is older, simpler and compact. But it remains extremely popular and continues to be relevant. The new xBRL-CSV specification provides a way to collect or publish large quantities of data, including granular data. The xBRL-CSV syntax uses the W3C’s tabular metadata standard to bind facts to their definitions, in an accompanying JSON file. This approach is simple and in initial testing compacts down XBRL documents by a factor of 20, making it extremely helpful for some of those environments dealing with large data sets.
Both specifications build on the existing, proven XBRL specifications and allow data to be validated against taxonomies and business rules without additional work.