Superexponential long-term trends in information technology

Moore's Law has created a popular perception of exponential progress in information technology. But is the progress of IT really exponential? In this paper we examine long time series of data documenting progress in information technology gathered by [1]. We analyze six different historical trends of progress for several technologies grouped into the following three functional tasks: information storage, information transportation (bandwidth), and information transformation (speed of computation). Five of the six datasets extend back to the nineteenth century. We perform statistical analyses and show that in all six cases one can reject the exponential hypothesis at statistically significant levels. In contrast, one cannot reject the hypothesis of superexponential growth with decreasing doubling times. This raises questions about whether past trends in the improvement of information technology are sustainable.

► We analyze six historical time series documenting progress in information technology for over a hundred years.
► We show that functional performance metrics have not been improving exponentially in time with constant doubling times.
► Instead, progress has been superexponential with shrinking doubling times.
► Some trends may be hyperbolic, leading up to a finite time singularity.
► We also propose a possible explanation based on information theory and extreme value theory.