Time to Put SQL In To NoSQL

Although SQL has been the language of choice for RDMS, it doesn't mean SQL cannot be applicable to NoSQL databases. It's really all about sets. If a collection can be treated as a set, and a collection of collections a set of sets, SQL should be an easy fit.

Consider this document from MongoDB. If you look at it in a different perspective, you should find that most of the things you can do using a vendor specific syntax can be done in SQL. That is, if only the NoSQL vendor would inherently support it.

NoSQL is not really a movement to NOT use SQL. It just happened that implementing SQL is not the first option. JSON (or BSON) is designed for a scripting language that does not include an SQL parsing engine. Developing it for the sake of on day 1 does not make sense. In fact, releasing their NoSQL products without SQL support is perhaps the smartest thing to do. I get it. But it's been years since. It's time to put SQL in to NoSQL!

Tangyuan-mongo shows it's possible. A new version of Studio 3T shows it's possible. There's no need to create a new language really. SQL is available and it fits right in.

NoSQL vendors can't be blinded by the NoSQL moniker. It's not a movement to be NoSQL. Similarly, SQL is not a movement. SQL is a solution. The technology is ready. All that's really needed is to use it, implement it and make it possible... NOW!

Comments