Short works

Books : reviews

Andreas S. Weigend, Neil A. Gershenfeld, eds.
Time Series Prediction: forecasting the future and understanding the past.
Addison-Wesley. 1993

Most observational disciplines, such as physics, biology, and finance, try to infer properties of an unfamiliar system from the analysis of a measured time record of its behavior. There are many mature techniques associated with traditional time series analysis. However, during the last decade, several new and and innovative approaches have emerged (such as neural networks and time-delay embedding), promising insights not available with these standard methods. Unfortunately, the realization of this promise has been difficult. Adequate bench-marks have been lacking, and much of the literature has been fragmentary and anecdotal.

This volume addresses these shortcomings by presenting the results of a careful comparison of different methods for time series prediction and characterization. This breadth and depth was achieved through the Santa Fe Time Series Prediction and Analysis Competition, which brought together an international group of time series experts from a wide variety of fields to analyze data from the following common data sets:
• A physics laboratory experiment (NH3 laser)
• Physiological data from a patient with sleep apnea
• Tick-by-tick currency exchange rate data
• A computer-generated series designed specifically for the competition
• Astrophysical data from a variable white dwarf star
• J. S. Bach’s last (unfinished) fugue from Die Kunst der Fuge.

In bringing together the results of this unique competition, this volume serves as a much-needed survey of the latest techniques in time series analysis.


Tim D. Sauer. Time series prediction using delay coordinate embedding. 1993

Neil A. Gershenfeld.
The Nature of Mathematical Modeling.
CUP. 1999

Neil A. Gershenfeld.
When Things Start to Think.
Coronet. 1999

rating : 3 : worth reading
review : 16 April 2001

I like living in the future -- that's one of the reasons I read science fiction. Now I realise that bits of that future are rather closer than I had thought. Gershenfeld is a director of MIT's Media Lab, and this is the story of some of the technology that he and his colleagues have been working on over the last few years -- technology that should help make computers usable (science fiction indeed!), smarter, ubiquitous, pervasive, and invisible.

Whether he's telling us of designing smart paper, building an electronic cello for Yo-Yo Ma, putting computers in shoes, using pens that remember what they've written, turning a high-tech magic trick into a child car seat sensor, or using 3D printers, Gershenfeld opens up a world of pragmatic possibilities, for most of the technology he describes already exists as some form of prototype in the Media lab. It "only" has to be productised. And he tells it all with a wonderfully wry style.

My office has ten things that include a clock, and each one reports a different time with equal confidence.

Towards the end of the book, he moves from describing new technologies, to explaining the educational and research medium that made them possible. The Media Lab is a different kind of learning environment. Students have a very broadly based practical, hands-on education. This helps them to stay grounded in reality, and to appreciate the context of any theory they learn. This theory is delivered in a "Just in Time" format, when and where they need it. His two supporting technical texts, The Nature of Mathematical Modeling and The Physics of Information Technology, provide the basis of that theoretical backgrounding. In addition, he describes how the Media Lab's partnership with industry works so well.

The Media Lab certain sounds a fun place to learn and work.

Neil A. Gershenfeld.
The Physics of Information Technology.
CUP. 2000

Neil A. Gershenfeld.
Fab: the coming revolution on your desktop---from personal computers to personal fabrication.
Basic Books. 2005

rating : 3.5 : worth reading
review : 15 May 2011

Gershenfeld's vision of "personal fabrication" isn't just about 3D printers. As well as printing "stuff", this is about cutting, milling and routing stuff. And adding intelligence to it, in the form of very simple circuits. The point here is that all these technologies are getting affordable enough, and compact enough, that individuals or small communities can have access to them (although it does require some software support to get all the kit working in harmony). And when they do, they build artefacts that no industry would think of, or would make if they did.

He illustrates all this with several case studies, of where "fab labs" were put in an Indian village, a Finnish community, a Ghanaian village, and a Bronx suburb. The people there learned to use the systems rapidly, taught each other, and built things that directly related to their lives. Part of the story is that this empowers the people, and also encourages cooperative learning. Also, probably unsurprisingly, it is the children who lead the way, leaping in to using the systems without worrying exactly how it all works, or if they are doing it "right".

There's more support here for the "Just in Time" versus "Just in Case" style of teaching. Interesting, inspiring, and thought provoking.