And Then There Was Information

Some time in the late 1970's William Kent published a book entitled Data and Reality in which he provided some philosophical grounding for the design of databases. Although it is still somewhat difficult to get through today, it was totally incomprehensible in the late 1970's. One of the zen koans that he provided was that database design should model information rather than the way information is processed. For database designers of the day, this was no more meaningful than the sound of one hand clapping.

Databases in the 1970's were design to support data processing systems. They were really little more than fancy files with some handy functional features like support for transactions, data independence and the like. The were storage and retrieval mechanisms used to support data processing. In fact, early data bases did not even have query languages.

But  a new idea was emerging in the late 1970s and that was the idea that all the data that was being processed might have a secondary purpose as information about the organization. This information could, potentially be used to better understand and more effectively run the organization. As obvious as this is today, it was a novel observation at the time and it took a  while to catch on.

Of course today well designed databases are designed primarily for information and processing is merely the means by which the information is kept up to data. But as data processing (automation) gave way to information systems there were a lot of entrenched views of data processing systems that needed to be over come.

In keeping with the expanding model I am developing here I would like to point out a few things. First, when automation was the primary usage of computer applications, it was far from obvious that the next stage would be information systems. In fact, it took quite a while for that idea to catch on. Second, as automation systems created information giving rise to information systems, information systems in turn created a greater demand for automation. That is, when you are modeling the organization in information you need more and more automation to provide more and more information.

Now we are in the era of information systems and the question that I started this thread with was - can we see what the next stage of evolution will be. If the model is consistent, it will not be obvious from what we have today and as we evolve into it, it will create an even greater demand for information. But, to find out what that next stage is, you will have to come back next week.