Nature Of Order

"The Nature of Order" is the latest book from ChristopherAlexander. More exactly: it is not a single book, it is a four volume (2164 pages) set published by Oxford University Press published 2003-2004. A web page with more information about the book may be found at http://www.math.utsa.edu/sphere/salingar/NatureofOrder.html. Alexander is selling the book online at http://www.natureoforder.com.

The above constitutes a preview/review written by NikosSalingaros (one of the editors of the book). He describes the book as follows:

: The Nature of Order has been in preparation for over thirty years, and encapsulates all of Christopher Alexander's theories. My own modest contribution has been to help Professor Alexander edit the manuscript during the past fifteen years. In this monumental book, Alexander develops a comprehensive theory of how matter comes together to form coherent structures. Paralleling, but not copying, recent results from complexity theory, he argues that the same laws apply to all structures in the universe; from atoms, to crystals, to living forms, to galaxies. Human beings apparently have a built-in (though subconscious) understanding of these laws. Man's creations have the option of following the same laws, or violating them. Those that follow them result in our greatest achievements, either as artifacts, as buildings, or as cities.
: This book promises to be of fundamental importance to computer programmers, defining in the words of some experts, "a new paradigm for programming". This is remarkable, since the book is written primarily in the interest of architects (of buildings, not software). It turns out that the same organizing principles apply to computer programs as to buildings. This connection was made recently by several visionary programmers, and has already generated remarkable insights and results. A good overview is the book by RichardGabriel, PatternsOfSoftware (New York: Oxford University Press, 1996; with a foreword by ChristopherAlexander).

JimCoplien has reviewed a draft of NatureOfOrder and has hopped around the globe giving those interested a small taste of how the book builds on Alexander's original "patterns" into a new paradigm of "centers" and "wholeness":

: Alexander, best known for his work on patterns, is currently working on a book series called Nature of Order. This book provides a unifying model of design that appeals to naturally occurring processes based on loci called centers. He aspires to give buildings, and other built artifacts, the same elegance as one can find in naturally recurring systems. A center is "something we notice" in a system. It is differentiated from neighboring centers by a fuzzy boundary. A hallway is a center; a country pond is a center. Patterns are stereotypical centers.
: One tenet of the book series is that beauty is objective. That is, given two design choices that build different structures, one is in some sense "better" than the other. This is traditionally held to be a subjective matter. But Alexander maintains that most people will usually choose a given design over its lesser counterpart, and that the choice is experimentally objective. He also notes that these experimental judgments correlate to the presence of about 15 properties such as strong centers, graded variation, local symmetries, etc.
: Design, then, is the process of intensifying centers. Existing centers can be intensified by adjacent centers. And each center intensifies centers contained within it (for example, an external building facade intensifies each of its windows, perhaps by a graded variation in size with the largest windows on the bottom and the smaller ones on top). The final system is a collection of smoothly integrated centers that reinforce each other.

Does anyone know what's up with the printing/publication of this beast? I pre-ordered and paid :-) for the four volumes from PatternLanguage.com this time last year. In February, vol. 1, "The Phenomenon of Life" appeared at my doorstep. A February Washington Post article said that Oxford was printing 50,000 copies of all four volumes in June of 2002. Now Amazon says that volume 1 won't be available until December 2002, and no other volumes are out. I would really like to know what is causing the delays and when I can get the rest of the set. -- TLR

: Can't tell. I also pre-ordered and paid. Today [Nov 5, 2003] I received the second volume "The Process of Life". I would be very much interested in discussing the first volume (now) and the second (as soon as I've read it). I'm somewhat more interested in online communities where the idea of a "living system" has a much more fertile ground than in software development. In fact I doubt that software pattern (as they are now established) have much to do with Alexander's ideas. -- HelmutLeitner (I got volume 4, then volume 3 in the first quarter of 2005)

: "This book promises to be of fundamental importance to computer programmers, defining in the words of some experts, "a new paradigm for programming". This is remarkable, since the book is written primarily in the interest of architects (of buildings, not software). It turns out that the same organizing principles apply to computer programs as to buildings."

I am skeptical that rules that apply to physical things like buildings also fully apply to computers. With computers you can create "virtual orders" or "virtual views" (I am a RelationalWeenie). You are not stuck with physical proximity or just the preconceived groupings of the original builder. Computers liberate us from many physical limitations. Many libraries no longer carry physical card catalogs, for example. Try searching for key-words using a card catalog. RealWorldModel.

The quoted comment applies equally well to TheTimelessWayOfBuilding and A PatternLanguage, which led to the recognition of DesignPatterns in software. The application of Alexander's ideas to software design is a fait accompli.

I think GOF patterns suffer a similar problem: they are not really virtual views, but more or less global statements of perceived order. I guess I just have a problem with the idea of global classification systems. EverythingIsRelative, and managing complexity requires facing this rather than keep pretending there are absolute models around which to build things. The larger the system, the more relativism applies in my experience.

See JimCoplien's wiki on NatureOfOrder at http://www.bradapp.net/docs/NoNoO.html, previously www.enteract.com/~bradapp/docs/NoNoO.html (read-only in fishbowl mode)

CategoryBook