The Evolution of Data Systems

Underlying all of today’s computer applications are Information Systems. Note that information characterizes the separations between entities. Information Systems are based on collections of entities whose types and attributes distinguish them from each other.

Information Systems are generally comprised of some form of a Query Generator, a System Integrator and a Data Store. The Query Generator is essentially the user interface that accepts requests for information and displays the result set. The System Integrator is the subsystem that knows what data elements are where and how to access them. The Data Store encompasses the location of every piece of data in the system that the end user could want to see.

Information Systems require no knowledge of the user beyond the content of the request being made and her authorization level. As such the layout and actions of its components are largely unaffected by the individual uniqueness of the user. In general, Information Systems manage data elements in a Euclidean (flat) data space. This means that, the logical distance between data elements in an Information System is generally the same for all users.

Of late, an increasing number of Knowledge Systems have begun to emerge. Note that knowledge characterizes the connections between entities. The connections between data elements in Knowledge Systems are their metadata, which is essentially data about the data. The distinguishing components of the Knowledge System architecture are its Preference Manager, Search Engine and Content Store.

The user interface of a Knowledge System is comprised of a Query Generator and a Preference Manager. The Preference Manager accepts user preference data in the form of demographic data and transaction history. Preference data are utilized by the user interface to customize its layout. The Search Engine uses preference data to shape the Content Store in terms of relevance to a given user.

The Content Store contains both fundamental data elements and the metadata that connect them on a more abstract level. The metadata of the Content Store are what distinguish it from a Data Store. Knowledge Systems manage relevance metrics that are used to generate non-Euclidean (curved) data spaces. This means that the logical distance between data elements in a Knowledge System can be the different for different users.

Information Systems are currently being engulfed into the history of data systems by the newly emergent Knowledge Systems, which represent the present. But the initial glimmers of the future of data systems are already becoming apparent. This future is the phenomenon known as the Experience System, which builds on the advances produced by the development of Knowledge Systems. The nature of Experience Systems will be the subject of the next essay.