Data lake or data warehouse? Stocking or stacking? Is one of these your usual approach to data management?

Because there is an alternative. It’s not a lake or a warehouse, but an infinite Structured Data Space where you neither have to cast your net in hope, nor know exactly where to look to retrieve your data:

Traditionally, these are the ways of storing and retrieving data.

On the one hand, you could be stocking your lake with shoals of data – and maybe one particularly handsome specimen – that you may some day hope to retrieve. On the other, you could be stacking the shelves of your warehouse with the boxes of 10 data widgets that you will expect to find because you have placed them on the ’10 data widget box shelf’ in the widget aisle you built specifically to store them.

But when you come fishing for your desired data you can never be sure it will ever take your bait, or that it won’t slip through a hole in your net when you try to fish it out. Even then, you may catch SOME data, but is this really THE data you wanted?

And, if someone removes the sign from the end of the widgets aisle, how will you ever know where they are unless you meticulously search row upon row, column upon column until you’ve found them?

That doesn’t mean, however, that the Structured Data Space is entirely liquid, or that it doesn’t have order. What it does mean is you can literally have the best of both lake and warehouse.

Even an infinite data space has rules, but they’re not complicated and they are instantly recognisable in our everyday lives.

We can say that all data is one of four types: It is either a person – a Who – an object – a What – a location – a Where – or an event in time – a When.

And all data can interact in that combination of ways we think of as logical opposites – All / None, And / Or.

The Structured Data Space is therefore made up of the interactions of the element types and their logical opposites, and these can be connected in endless combination.

If we’re to continue our earlier analogy, this means that far from having to take an educated guess when dipping your hook and line into the data lake, or being constrained into retrieving only what you specifically ask for from the shelves of your data warehouse, you can be assured of finding not only what you were looking for, but all of it’s connected data when you take a wander through the data space.

In terms of models, the difference is this:

Data Lake – no model.

Data Warehouse – one model.

Structured Data Space – Four stable dimensions interacting in infinite models.

Your journey through the structured data space is neither a fishing expedition, nor a single trip to a specific destination. It is a journey of discovery that can take you from a point of your choosing to a point where either you have learned all you need to know or that the connections logically run out.

Chances are, your journey will never take you to the end of the trail, but you will certainly go significantly further than you ever thought possible.

by Michael Brands

Michael Brands is CEO of Dynactionize NV and the inventor of six patents related to efficient data analysis, usage and storage. He holds degrees in linguistics, philosophy and computer science.