In the 1970s, while I was working for the publishing division of the Xerox Corporation, the company engaged in something called the Records Retention Programme. The purpose of this was to distinguish between records and files (loads of paper) that needed to be kept for legal or operational reasons, and to decide which could be disposed of, which could be stored more efficiently on microfilm, which moved to storage facilities that were cheaper than filing cabinets located in prime city centre offices, and which needed to be kept near to the people who generated it. It was a major operation and one that would, it was calculated, save the corporation as a whole more than fifty million dollars.
Now that such records are mostly kept in digital form, I wonder if the expense associated with retaining, organising and archiving organisational data is fully understood and taken into account by the companies that generate them. I suspect not and that the cost of keeping data that is surplus to requirements is running out of control. Does anyone know, or care, how much time, energy and money is squandered building and maintaining unnecessary server farms, mountain caverns, desert bunkers and tundra outposts? Someone certainly knows how much money is to be made by the companies doing the data warehousing job.