IT question and answers

In the beginning


In the world data abounds. Always has and always will. The problem for mankind was first to identify ways of describing the information about specific properties of natural or man-made systems, and then to actually measure parameters associated with these properties. Record keeping systems had been successfully and unsuccessfully employed for literally thousands and thousands of years in the form of notations on clay tablets, palm leaves, rock, timber, bone, ceremony, dance, music, poetry and a variety of other media. When mankind moved into the infancy years of computers, nothing much changed except that a new media of information storage was being researched and developed. Electromagnetic media. Then in ancient days, as now, the organisations which attempted to seek better methods of recording information and data were of essentially of three different kinds: scientific, administrative and business. Codification of organisational intelligence started with the passing down, from generation to generation, of the many and varied cultural oral traditions scattered across the face of the planet which today are remnant in transliterations of the planet's most ancient texts. This information is briefy summarised and indexed elsewhere in this website, here. When the electronic, or more correctly stated, electromagnetic storage of information was broached as a research and development project, there were a number of pioneering systems developed. But perhaps the grandest and most inspiring project of them all was the one which sought a method and means of managing a process of transportation to enable man to set his first footprint upon the moon. Computation, as it was known in those days, was a far cry from what we expect of a computer system in today's world. The machines had serious memory shortcomings so that only a very very limited amount of program code and data was able to reside in the machine's memory. To accentuate this fact, contemplate this following quote, made more than a decade after the Eagle landed:
      "640K ought to be enough for anybody."
      -- Bill Gates, 1981 (Further reference: here)


In the early days, electromagnetic data (and program) storage was feasible only at an extremely high cost. In the 1960's such investment in technology was only possible with large capital expenditure. During those early years the only computers in existence were research projects in themselves, with their own R&D budget. Only when the technologies of computer memory, processor and storage advanced into the 1970's and these component prices began falling as a result of much pioneering work in the field, did it become feasible for smaller organisations to invest in the new technology. During all these earlier phases in the evolution of computers systems there was one marked characteristic about their management, which still holds true in todays modern IT environment. The management of the Information Technology environment implied the management of a queue of parallel problems, their detection and their resolution. By this I imply that although things were always theoretically simple, when this theory was placed into practice, rarely would everything go smoothly forever. Methods of problem resolution needed to be engineered, and problems resolved. In the very beginning the services of IT Management were conducted by heads of research teams, either at academic or other government institutions, or in exceptional instances, such as IBM, within a corporate infrastructure of research and development in the field of computing. Contrary to the metaphor used in the introduction to this series, the earliest computer systems were not cars, but rather large trucks with an attendance of a host of specialised mechanics.


In one corner, in many formats, weighing in at anything up to (INSERT DATA STORAGE SPACE REQUIREMENT HERE) was the actual organisational data. The nature of this data obviously changed (and of course still does) from organisational type to organisational type. Scientific organisations maintained data sets of the physical properties of materials, and molecular compounds and atomic elements, etc. Administrative organisations, such as all types of government departments, charity and cultural organisations, maintained data sets of legislations and people, registration plates, horse and land ownership, classroom facilities, utilities management, etc. Business organisations maintained records specific to the proper running and management of their business. Whether the business involved the buying and selling of goods of trade, or whether it involved the selling of professional services in some form, it had records to be kept of its business transactions (but to a lesser degree than today) Once the database had been populated with the organisation's data it became, to use the expression from the field, a production concern. The data structures housed all elements, in as much detail as required by the organisation, of its operational information. And once such structures became a production concern, they do what all living things do, they evolve. The early computer systems had by today's standard, savage database integrity problems caused by data structures exceeding specified sizes, by the problems introduced by the historical accumulation of data once a system had been in production for a number of years. Database "SYSTEMS" as such did not exist. Everything was completely and utterly specified within the program code used to populate and to maintain the organisational specific data structures. Which leads us to the other corner of the arena, and weighing in with a considerably smaller and relatively stable footprint, was the program code ...


Before the era of the RDBMS there were a variety of different species of data file structures. Each were specific to the application program code which implicitly defined it. At this early time in the history of the evolution of organisational intelligence into the electromagnetic medium, 100% of the organisation's intelligence was resident in the program code used to manipulate the data structures. Even at this early stage of development there appeared, naturally in the evolution of things, a number of separate parties upon whom the Information Technology Management within the organisation relied for goods and services of trade. The hardware requirements were for a machine and its peripherals sufficient for the needs of the organisation. Required: One hardware vendor. Additionally, separate requirements always existed for some form of database application software code specific to that organisation. As the applications did not exist in the beginning, they had to be written. Required: One software developer. This could involve literally one or two programmers, or one or two consultant programming development corporations, depending on scale. If only things were as simple. Unfortunately, it was rare that one hardware vendor could supply all of an organisations need for hardware and consequently the environment entertained many hardware vendors per organisation, and likewise, many software vendors over time. Systems and their environments evolved, complexity increased, data requirements grew year by year. The IT Problem queue was dealt with on a day-to-day basis. Database storage technology itself needed to evolve. The task of managing constantly growing data structures was becoming an occupational hazard for IT management and operational staff. Enter, in the mid to late 1960's the Database Management System (DBMS).