At the beginning of seventies when microprocessors were introduced to the world very few could imagine that anything singular, small, or personal built on top of microprocessors could win over IBM 360 mainframe series. One could compute almost nothing with microprocessors and the enterprise world was overwhelmed by the promises of mainframe computers.
It took only a decade for personal computers, driven by passion of homebrew computer enthusiasts, to become the new paradigm of emerging swarm infrastructure. In the late seventies the spreadsheet application excited small businesses that personal computers will not be only toys for geeks and nerds. When, during the nineties, personal computers got all connected into world wide web it seemed that mainframe (centralized) paradigm lost the war against distributive, bottom-up, spontaneous, capillary dynamics of communication infrastructure build by individuals buying cheap computers on competitive free market of consumer electronics.
It took only another decade, the zeroes, where mainframe, centralized, monopoly paradigm got back in power. It happened through:
- use of loopholes in GNU GPL where companies could use free and open source software but not being legally forced to contribute back their modifications because their adjusted softwares were only run "internally" on their servers;
- use of cheap computer hardware to scale up thanks to victory of personal computers paradigm through few decades before;
- venture capital driven outsource of the research and development to so called start-up scenes where popularity contest or network-effect became the most important requirement for recruitment of the best engineers for oligopoly of Google, Amazon, Facebook, Apple and Microsoft;
- the enforced discipline of coding SOA (Service Oriented Architecture);
- institutional/organizational adaptation where corporation will keep the mainframe paradigm in the field of a control over allocation of resources, and decision making while technical infrastructure stays distributed;