Diagrams as Information Structures
Background and Influences:
Christopher Alexander, educated both as an architect and mathematician, was widely influential during 60’s -and still is- on discussions around design methodology and the use of computers in design, architecture and urbanism. Despite being a quite controversial and “lone” scholar, he was engaged in an active conversation around these issues, due to his involvement in multiple academic collaborations, conferences and publications. In particular, during his doctoral studies at Harvard (1958+) he worked with the Center for Cognitive Research, the Joint Center for Urban Studies of MIT and Harvard, and the MIT Civil Engineering Systems Laboratory. In 1962 he participated in the first of a series of Conferences on Design Methods in London, attended also by personalities like Gordon Pask and representatives from HfG Ulm. In 1967, he founded the Center for Environmental Structure in University of California-Berkeley, where he continued his teaching career for almost 40 years.
Alexander’s work argues for the relationship between science and design methodologies: ‘Scientists try to identify the components of existing structures, designers try to shape the components of new structures.’ (Alexander, 1964) For Alexander structure means form that manifests information structures. He expresses design problems as systems. In order to synthesize a form, he first decomposes the system into subsystems and tries to define their interrelationship.
“This is what lies behind D’ Arcy Thompson’s remark that the form is a diagram of forces. Once we have the diagram of forces in the literal sense, this will in essence also describe the form as a complementary diagram of forces.”(Alexander, 1964)
This systemic approach is directly influenced by Cybernetics and specifically Norbert Wiener and Ross Ashby’s homeostasis theory. In Alexander’s case, systems are used to describe a environmental approach to design, which conceives our habitat as an open system of animate and inanimate agents (Maldonado, 1969) and design knowledge is approached holistically as the man-environment relationship. Due to his continuous effort to describe design as a diagram of forces that and his participation in Design Methods Conference, his name is associated to the nascent field of Design Methods. However, later in his life, he denounces the whole field:
“Indeed, since the book was published (Notes on the Synthesis of form, 1964), a whole academic field has grown up around the idea of “design methods”-and I have been hailed as one of the leading exponents of these so-called design methods. I am very sorry that this has happened, and want to state, publicly, that I reject the whole idea of design methods as a subject of study, since I think it is absurd to separate the study of designing from the practice of design. […] Study of method by itself is always barren, and people who have treated this book as if it were a book about “design method” have almost always missed the point of the diagrams, and their great importance, because they have been obsessed with the details of the method I propose for getting at the diagrams.” (Alexander, 1971)
Relation of problem decomposition to computational means: Diagrams and Computer Programs
Alexander’s constant redefinition of theories and visualization of information structures is even a more insightful as a narrative, than the criticism he received for being too deterministic. It is interesting to see the relationship between his information structures (diagrams) and the capacities of computer programing. It is argued here that the units of work had a significant impact on how he perceived and described design problems. The decomposition of the problem should be processed in a standardized manner, following rules, as if a computer performed it.
Tree Structure and HIDECS 2 (Hierarchical Decomposition of Systems): In his dissertation Notes on the synthesis of Form, Alexander describes design problem as sets that can be in a hierarchical tree. The computer program HIDECS 2 that corresponds to this theory uses a binary stochastic algorithm that “at each level of the tree each set of variables is broken into those two subsets with minimum information transfer between them”. Alexander realizes that this method does not take into account the holistic relatedness of system and subsystems. Usually subsystems overlap in a system.
Lattice Structures and HIDECS 3: Since his realization that structural interrelations cannot always be described as trees, he writes the City is not a Tree and HIDECS 3 that aims to tackle weaknesses of HIDECS 2, while using the same “machine representation”. He therefore introduces the structure of the semi lattice. In this program the decomposition into subsystems is not defined by a binary condition in each step, but all at once.
Later on, Alexander will write collaboratively Pattern Language, his most famous work, where he looses his control on information structures and describes systems as open networks. Patterns contain information of the forces that made them but they are at the same time instances of an open-ended approach to design. “Each pattern is a field — not fixed, but a bundle of relationships, capable of being different each time that it occurs, yet deep enough to bestow life wherever it occurs.”
Alexander employed mathematics and in particular set theory to analyze design problems and used graphs (diagrams) to bridge the gap between architectural and software representation. This relation of computational means and techniques for problem solving is probably one of the reasons that made Alexander so popular among object-oriented programming.
Bibliography & References:
- Alexander Christopher, “Notes on the Synthesis of Form”, Harvard University Press, 1964
- Alexander Christopher, “The City is not a Tree”, Ekistics Vol.23, 1967
- Alexander Christopher, “HIDECS 3: Four Computer Programs for the Hierarchical Decomposition of Systems which have an associated linear graph”, MIT, Department of Civil Engineering, 1963
-Maldonado Tomas, “How to fight complacency in design education”, Bit 4 International, 1969
- Cross Nigel, “A History of Design Methodology”, 1993, available here: https://monoskop.org/images/6/66/Cross_Nigel_1993_A_History_of_Design_Methodology.pdf
- Wright Steenson Molly, Architectures of Information: Christopher Alexander, Cedric Price and Nicholas Negroponte & MIT’s Architecture Machine Group, PhD Thesis, Princeton, April 2014