[chirau]

The article is about the internal storage mechanics of ClickHouse and how it optimizes handling JSON data behind the scenes. The data types like Dynamic and Variant that are discussed are part of ClickHouse’s internal mechanisms to improve performance, specifically for columnar storage of JSON data. The optimizations just help ClickHouse process and store data more efficiently.

The data remains standard JSON and so standard JSON parsers wouldn’t be affected since the optimizations are part of the storage layer and not the JSON structure itself.

[chipdart]

> The data remains standard JSON and so standard JSON parsers wouldn’t be affected (...)

No, not really.

The blog post talks about storing JSON data in a column-oriented database.

The blog post talks about importing data from JSON docs into their database. Prior to this, they stored JSON documents in their database like any standard off-the-shelf database does. Now they parse the JSON document when importing, and they store those values in their column-oriented database as key-value pairs, and preserve type information.

The silly part is that this all sounds like a intern project who was tasked with adding support to import data stored in JSON files into a column-oriented database, and an exporter along with it. But no, it seems an ETL job now counts as inventing JSON.