Evolution and Development of the Universe
The Physics of Performance Curves: Nature and Limits
Technology performance curves, also known in engineering, economics, and manufacturing as progress functions, and in cognitive science as learning curves or experience curves, in which technological capacity or efficiency improves by exponential, power law, or other fashion with cumulative production, often over very long time periods, have been studied by a small group of scholars since the 1930's. Given their accelerating impact on the technology environment, they are among the most important topics of technology innovation, strategy, and policy, and to improving our economic models of production functions. Yet in spite of their importance we do not have a good understanding of the physical basis of these curves, and many open questions remain. Fortunately, performance curve scholarship is on the rise, and the opportunity for high-impact collaboration and publication in this area has never been better.
Topics of Investigation:
- What models do we have for the physical basis of technology and complexity performance curves?
- Can we develop unifying theories for any classes (physical, efficiency, computational, informational) of performance curves today?
- Why are scale reduction processes persistently exponential in performance improvement, and which physical processes are candidates for continued scale reduction?
- Why are virtualization (dematerialization, simulation) processes persistently exponential in performance improvement, and which physical domains are candidates for further virtualization?
- To what degree are automation and machine learning virtualization processes? Efficiency processes? As they advance, how can we model their global exponential performance effects?
- For exponential curves, learning is based on a fixed percentage of what remains to be learnt. For power laws, learning slows down with experience. When is each valid?
- What explains the long-term smoothness and predictability we find in many technology performance curves in our Performance Curve Databases?
- Why are long-term progress forecasts in certain fields, such as computing, communications, and nanotechnology, significantly more predictable than in other fields?
- When are smoothness and predictability due to physical law, averaging, scale, collective learning, economic or psychological expectations, or other physical processes?
- Standard deviation and skew in performance times often show power law decreases with cumulative experience. Why and when does this occur?
- Why are tech product outliers so often market failures? Are outliers normally or log-normally distributed vs. the curve? Can such data improve R&D timing, strategy, and policy?
- How do we differentiate non-persistently exponential performance curves (market-limited, etc.) from persistently exponential (scale reduction, FERD, etc.) curves?
- How do non-computational (physical process, efficiency) performance curves differ from computational (computing, memory, communication) performance curves?
- How do computer hardware and software performance curves differ, and why does hardware exhibit consistently better long-term exponential performance improvement?
- When does technology substitution (creating a composite technology performance curve) occur in any technology platform? Under what circumstances can we predict it?
- When does exponential performance end in any performance curve? Under what circumstances can we predict it?
- What explains state switches (transitions to steeper or flatter exponential modes) in several technology performance curves?
- What physical processes differentiate superexponential, exponential, logistic, life cycle, and other performance curves?
- What do exponential and superexponential performance and efficiency curves imply for the future of technological innovation and sustainability?
We are seeking physicists, computer scientists, process engineers, technology performance curve scholars, technology substitution scholars, virtualization and scale reduction scholars, management and learning theorists, economists, complexity theorists, technological evolution and development scholars and their critics. Scholars who approach performance curve study from materials science, thermodynamic, computational, informational, evolutionary, developmental, economic, competitive, cognitive science, social science, systems theoretic and other perspectives are welcomed. We will seek to compare and critique a variety of performance curves data sets and models, and consider first-order implications of these models for technology innovation, strategy, sustainability, and policy, underscoring the great technical, political, economic, and social value of better scholarship and science in this area.
- Bills, Albert G. (1934) General Experimental Psychology, Chap 10, The Curve of Learning (pp. 192-215), Longmans.
- Wright, T.P. (1936) Factors Affecting the Cost of Airplanes. Journal of Aeronautical Sciences, 3(4):122–128.
- Newell, A. & Rosenbloom, P.S. (1981) Mechanisms of skill acquisition and the law of practice. In: J.R. Anderson (Ed.), Cognitive skills and their acquisition. 1-51.
- Dutton, John M. & Thomas, A. (1984) Treating Progress Functions as a Managerial Opportunity. Academy of Management Review, 9(2):235-247.
- Hartle, J.B. (1997) Sources of Predictability, Complexity 3(1):22-25.
- Wernick, I.K. et.al. (1997) Materialization and Dematerialization: Measures and Trends. In: Technological Trajectories and the Human Environment, Ausubel, J.H. and Langford, H.D. (eds.), National Academies Press.
- Triplett, J.E. (1999) The Solow productivity paradox: what do computers do to productivity?, Canadian J. of Economics 32(2):309-334.
- Heathcote, Andrew et.al. (2000) The Power Law repealed: The case for an Exponential Law of Practice. Psychonomic Bulletin & Review. 7(2):185-207.
- Limpert, E. et.al. (2001) Log-normal distributions across the sciences: Keys and clues. BioScience 51(5):341-352.
- Batty, M. & Torrens, P.M. (2001) Modeling complexity: the limits to prediction. UCL Working Paper 36, CASA.
- Nordhaus, W.D. (2001) The Progress of Computing. Cowles Foundation Discussion Paper No. 1324, 61 p.
- Albright, R. (2002) What Can Past Technology Forecasts Tell Us About the Future? Tech. Forecasting & Social Change 69(5):443–464.
- Ritter, F.E., & Schooler, L. J. (2002) The learning curve. In: Int. Encyc. of the Social and Behavioral Sciences, 8602-8605, Pergamon.
- Chaisson, E.J. (2003) A Unifying Concept for Astrobiology, International Journal of Astrobiology, 2:91-101.
- Devezas, T.C. and Modelski, G. (2003) Power law behavior and world system evolution. Technol. Forecast. Soc. Change 70:819–859.
- Jenkins, Alastair D. (2005) Thermodynamics and economics, Arxiv.org.
- Koh, H. and Magee, C.L. (2006) A functional approach for studying technological progress: Application to information technology. Tech. Forecasting & Social Change 73:1061-1083.
- Aunger, R. (2007) Major transitions in 'big' history, and A rigorous periodization of 'big' history. Tech. Forecasting & Social Change 74(8):1137-1178.
- Koh, H. and Magee, C.L. (2007) A functional approach for studying technological progress: Extension to energy technology. Tech. Forecasting & Social Change 75:735-758.
- Nordhaus, W.D. (2007) Two Centuries of Productivity Growth in Computing. The Journal of Economic History 67(1):128-159.
- Gantz, J.F. et.al. (2008) The Diverse and Exploding Digital Universe: A Forecast of Worldwide Information Growth Through 2011, IDC.
- Arthur, W.B. (2009) The Nature of Technology: The Past and Future of Human Innovation, Free Press.
- Clauset, A. et.al. (2009) Power-law distributions in empirical data. SIAM Review 51:661-703.
- Magee, C.L. (2009) Towards quantification of the role of materials innovation in overall technological development. Working Paper 2009-09, MIT Engineering Systems Division, 31pp.
- Kelly, Kevin (2010) What Technology Wants, Viking.