I am a little foggy on this but thought I should throw it out there. I had thought that XML was first built as a dataform to transfer data in an agreed upon schema between two data sharing partners. As such the schema could be very complex and specialized. Some where along the way the idea came to store this probably as an audit train for online transactions. This was due to probably legal requirements to keep the original transactions as they came to the server/service.

Originally stored as string of data, the parsing and query of the data within the schema was done by the application and simply stored as string data in databases. Then came a new age of enlightenment where data servers no longer saw this data only as strings but as data within data controlled by external knowledge and standardized schema proforma.

What then was at one point orderly data parsed into the database, became numerous xml fields each acting as a data heap or unformatted data collection that may or may not conform to any published schema. Truly this added security to the use of XML because no one knew what was going on. It also added to the complexity of storing and retrieval of data, let alone the ability to query the data structures within the heap.

So what once was and really is still today a very nice instrument for carrying data from point a to point b has now become a playfield for those who wish to define their own rules and order, but what does it do the maintainer of the process/database?

M.