International Journal of Communication
ABSTRACT:
Part I of this two-part article reviews methodological and statistical challenges involved in the estimation of humanity’s technological capacity to communicate, store, and compute information. It is written from the perspective of the results of our recent inventory of 60 technological categories between 1986 and 2007 (measured in bits and MIPS [million-instructions-per-second]). In Part I, we summarize the results of our inventory, and explore a series of basic choices that must be made in the course of measuring information and communication capacities. The most basic underlying assumptions behind our estimates include—among others—decisions about what is counted as (1) communication, (2) storage, and (3) computation; if technological capacities or consumption of information is measured; and if unique information is distinguished from duplicate information. We compare our methodological choices with different approaches taken in similar studies. The article shows how the particular question on the researcher’s mind, as well as the availability of source data has and will influence most of the methodological choices in different exercises. Part II focuses on the adequate unit of measurement for quantifying information. We propose an information theoretic measure that approximates the entropy of the source (which we call “optimally compressed bits”). We explain the interpretation, creation, usage, benefits, and limitations of this unit of measurement. A more coherent understanding of information volumes and magnitudes starts with a thorough understanding of the methodological choices involved in related inventories. We also discuss statistical lessons learned in our exercise (which is informed by more than 1,100 sources) in the roughly 300-page supporting online Appendix that is available at http://www.martinhilbert.net/WorldInfoCapacity.html .
Part I of this two-part article reviews methodological and statistical challenges involved in the estimation of humanity’s technological capacity to communicate, store, and compute information. It is written from the perspective of the results of our recent inventory of 60 technological categories between 1986 and 2007 (measured in bits and MIPS [million-instructions-per-second]). In Part I, we summarize the results of our inventory, and explore a series of basic choices that must be made in the course of measuring information and communication capacities. The most basic underlying assumptions behind our estimates include—among others—decisions about what is counted as (1) communication, (2) storage, and (3) computation; if technological capacities or consumption of information is measured; and if unique information is distinguished from duplicate information. We compare our methodological choices with different approaches taken in similar studies. The article shows how the particular question on the researcher’s mind, as well as the availability of source data has and will influence most of the methodological choices in different exercises.
Part II focuses on the adequate unit of measurement for quantifying information. We propose an information theoretic measure that approximates the entropy of the source (which we call “optimally compressed bits”). We explain the interpretation, creation, usage, benefits, and limitations of this unit of measurement. A more coherent understanding of information volumes and magnitudes starts with a thorough understanding of the methodological choices involved in related inventories. We also discuss statistical lessons learned in our exercise (which is informed by more than 1,100 sources) in the roughly 300-page supporting online Appendix that is available at http://www.martinhilbert.net/WorldInfoCapacity.html .
