Google found a way to more-or-less reliably process Big Data using commodity hardware and an underpinning of failover logic. Everyone wanted to beat Google. Th

Google found a way to more-or-less reliably process Big Data using commodity hardware and an underpinning of failover logic. Everyone wanted to beat Google. The first step (a fallacy) seemed to be to copy their way of doing things. So started "cloud computing". Of course most of the players that tried bastardized the whole thing immensely. At this point the original cloud model of computing (very thinly interfaced, commodity hardware as a foundation for scalability) was bastardized into a sludge of "outsourced administration", "throwing more servers at the problem", and "shoving traditional product models into datacenters and hoping they'll magically work at massive scale". And that's why cloud computing today is meaningless. Wait a few years until the companies that "just don't get it" die out, and cloud computing will be taken back as a legitimate description of a technological approach much like AJAX.