It doesn't feel the author has real world experience on relational databases. Let's see : "Small records with complex, well-defined, highly normalized relation

It doesn't feel the author has real world experience on relational databases. Let's see : "Small records with complex, well-defined, highly normalized relationships." Maybe, but not necessarily. Example? Call Data Records for a phone billing application. While there may be relationships in the inserted rows, even to other databases (i.e. customer information) the data stands pretty much on it's own. The challenge is to get floods of data into a single table, so that it can be later sliced and diced for billing purposes. "The type of queries you will be running on the data is largely unknown up front." Excuse me? This is so very wrong. I don't even no where to start. EVERY good RDB application is designed in a way where (ideally) all queries are known up-front. If you have to guarantee response times (i.e. think of a cash withdrawal at an ATM) you absolutely MUST control the queries that run on the db. Ad-hoc analysing and reporting MUST be relegated to dedicated, possibly replicated databases. "The data is long-lived: it will be written once, updated infrequently relative to the number of reads/queries, and deleted either never, or many years in the future." Mostly so, but absolutely not necessarily the case. I work on an application where the entire data is toasted after a couple weeks and in fact: today and yesterday would suffice. "The database does not need round-the-clock availability (middle-of-the-night maintenance windows are no problem). Queries do not need to be particularly fast, they just need to return correct results eventually." This is so full of crap, I won't even get into it "Data integrity is 100% paramount, trumping all other concerns, such as performance and scalability" Yes, 100% integrity is paramount, but most certainly not at the cost of scalability, let alone performance. Recently, methinks, there are a lot of proponents of new and improved data management capabilities, who see their little walled environment, but seem to have no whatsoever experience running databases in a real business. A normal (even big, huge or multinational business) does not have those "cloud-data-management-requirments" that very, very few companies really have.