 I'm speaking about ways to analyze operational workloads for NoSQL. A lot of what's out there in terms of trying to help people make decisions on it is ranges from a little bit misguided to not quite hitting the kinds of things that people really need to think about. So I'm hoping to help people start to evaluate these products in a more systematic and clear-cut way. NoSQL space is evolving very rapidly. The various types of databases are converging into sort of one major theme, at least in the operational space, and that's a large part of the talk I'm going to give. I think the industry itself is going to consolidate a lot, and the databases will flow from that. Well, NoSQL is a major part of how our company operates. Even before NoSQL databases were a thing, we were using many of the concepts of NoSQL to help our clients scale. Now that the industry has matured a lot, it's made our job a lot easier as a lot of the things we used to painstakingly do by hand has been automated for us, and the more the industry evolves, the better it is for us, and we're betting big on it. Relational databases are here for good. They're not going away anytime soon, if ever. They've proven themselves again and again as a fantastic tool and are extremely flexible. What we believe the main value of NoSQL is, is in those categories of applications that really have extremely high data loads. These operational workloads, there are kinds of operational workloads that relational databases have trouble handling. This will be a larger and larger part of the market as people start to discover all the data they could be capturing and what they can do with it, but relational databases are not going to go away or even close to it. I hope to get out of this conference a better understanding of where the industry is moving. Our company really depends on understanding NoSQL, and we've bet very big on this technology. The industry is evolving very rapidly, and it's critically important for us to know where it's going.