The problem with NoSQL databases is that developers still do not understand how to use them well. NoSQL requires relearning how to use a database. Many developers misuse NoSQL databases, relying on the SQL patterns they know so well. I think that's Oracle's main point here. For example, it's easy to have a lot of round trips if you use NoSQL badly, but if you adapt and take advantage of its strongpoints, NoSQL can easily reduce the query complexity.
The point of this paper is to scare Oracle's customers from switching to NoSQL, and I agree that companies should not blindly switch from SQL to NoSQL without the willingness to relearn and adapt.
Everything in computing requires learning and relearning. If you hired, or work with, developers who can't learn or adapt you have far more serious problems than database choices to worry about.
I was just getting excited about seeing the secretary problem reworked when applied from the perspective of both the secretary and the employer, but instead he took the romantic approach.
"USAA's investment products, most checking and savings products, credit cards, life insurance, and shopping and discounts are available to other individuals. USAA auto and property insurance is not available due to membership eligibility requirements."
Touché sir. I had looked at their insurance in the past and very unfortunately extrapolated to the entire organization. My mistake and thank you very much for correcting that misconceived notion!
What I really didn't understand is why they even decided to switch to NoSQL in the first place. I mean Urban Airship seems like the perfect example of data is completely relational and schema-driven.
The point of this paper is to scare Oracle's customers from switching to NoSQL, and I agree that companies should not blindly switch from SQL to NoSQL without the willingness to relearn and adapt.