About King Context.

In short: on context, (streaming) (big) data & information architecture.

More info soon. Someday.

Introducing unknown data

NoSQL had to be explained afterwards as “not-only SQL”, when it became clear that it’s not SQL itself that is a problem, but rather its unavoidable usage for use cases where it’s - to say the least - sub-optimal. And of course that’s in essence not an SQL problem, but more a consequence of using an RDBMS for every use case, appropriate or not. But than again, if you only have a hammer, you tend to see every problem as a nail.

Terminology, it’s often a problem.

Big data? Same story. It’s not always all that big. Smart data? No, the data itself is not smart. If you mean “smart use of data” by that, it’s more a term for a group of use cases, which does make more sense as a matter of fact. But still, every use of data is intended to be smart, no?

Unstructured data perhaps? There’s no such thing. All data has a structure, but perhaps the structure is not apparent or not explicit. The absence of a schema or some form of structure description doesn’t mean there isn’t any. It’s simply not immediately available, and it makes using the data for humans and tools harder. Harder, but not impossible.

In the world of what’s commonly referred to as big data, it’s about use cases of using and combining data that was never used before to do something smart other than perhaps monitoring systems and do some reporting. The size of these data sources perhaps used to be a problem, so the term big data was first used. In the meantime technology is available to deal with vast amounts of data. And to sustain the hype, a new term is needed.

Terminology, it’s often problematic because we’re too eager to put a label on it. Because we are hype-addicted.

Smart usage of previously unused (big) data sources is hard because of two reasons. First of all the existence of the data source needs to be known. And secondly its structure or schema is often unknown, meaning it’s unavailable (in a format) to be easily used (in existing tools). This also implies it is unknown explicitly what is in the datasource.

Dark data is one way to call it, but perhaps unknown data is a better term?

Context: Turning Big Data Into Smart Data: 5 Lessons For Marketers From The Obama Campaign

It’s not about big data, it’s about smart data used at scale