mommysage

Gradually, machines were designed to replicate human action, but replicating human mental processes remain the stuff of dreams. and this began to change with the development of machines that could add, multiply, and eventually make decisions. At first, even the most advance machines were limited to one's specific task if you needed something else done, you had to build a new machine. However, around 200 years ago, a great industrialist thinker, Charles Babbage, was dreaming of a general symbolic machine. One that could answer complex questions by breaking them down into smaller questions of logic and arithmetic and essentially braiding them together. His final dream was never realized, but his driving question remained. Was it possible to build a general machine that could answer any question? Now, by the 1900s, mathematicians and philosophers were posing this question in different ways. Mathematicians asked, what are mechanical machines capable of? How powerful could they be? While philosophers asked, what are the limitations of mechanical machines? What will machines never do? And in 1936, Alanouring bridged this divide with a paper which revolutionized our understanding about what machines can and cannot do. He outlined blueprints for what he called a universal machine, a machine that could answer anything that was answerable. Part of his great insight was that the power of the universal machine would always reside in the instructions it followed, later known as software, not the physical design of the machine, or hardware. and using a simple language, his machine would run any instructions that you could imagine. and the year after Turing's paper, a young Claude Shannon completed a master's thesis describing a clever insight he had about telephone relays. He realized he could arrange electrical switches in various ways to perform the fundamental operations of logic automatically using electricity. Suddenly, it was practically possible to build the universal computer powered by electrical clocks that buzzed away at near the speed of light and followed any instructions you provide. In the decades to follow, computing machines grew in their speed of operation and memory capacity. Suddenly, many hard questions humans faced became easy or very practical for computers to answer quickly. but deeper problems emerged. There seemed to be a growing set of seemingly easy problems, such as as a given number prime, that were computable on our machines, but took so long when the questions were large, such as as 140 trillion trillion trillion one prime that it could take thousands or even millions of years for the computer to give you an answer, or halt. So these problems were practically impossible to solve. Think of these as heart problems. and people consider drawing a line in the sand between problems that were easy, practical to solve, and problems that were hard, practically impossible to solve and the attempt to precisely define this division of easy and hard, practical and impractical problems leads to the most important unsolved question in computer science today. What makes hard problems hard? Is it a result of some underlying mathematical pattern? Or is the perception of hardness merely an illusion? Will new insights make these hard problems easy, and this question is not just an intellectual curiosity, the backbone of the Internet depends on a set of problems being out of reach or practically impossible for our machines. But to begin, we must go.