Mongo's debut was well timed to ride the JSON crazewave.
The only "advantage" to JSON is that it maps directly to the object trees most developers use in their scripted programs (which is misguided IMHO).
It had to deliver three kinds of formats: CSV, JSON, and XML.
JSON was a one-liner, then came CSV and then XML.
But all the consultants working in customer projects said CSV was the most important one because it's industry standard.
When the consultants sent me example CSV files from like 5 customers I couldn't believe they all had a different format.