From Science of Design

With support from the National Science Foundation's Science of Design program, Carnegie Mellon and the University of Virginia are developing and starting to offer courses and a curriculum for PhD students for whom design is a central research issue. Such research might be about the nature of design, about design in a specific domain, or about a specific application for which the research has a significant design component. Research with the objective "design an X" is likely to be out of scope, but research with the objective "develop a method for designing a class of things of which X is an instance" is likely to be in scope.

One of the goals of these educational efforts is to begin laying a foundation for realizing Herbert Simon's long-standing vision of a science of design.

Contents 1 Carnegie Mellon University 1.1 Summer School in Design Research 1.2 Spring 2007 Course in Strategies for Research About Design 1.2.1 Course catalog description 1.2.2 Plan for the course 1.2.3 Working documents for this course 2 University of Virginia 2.1 Course description 2.2 Prerequisites 2.3 Topics 2.4 Readings 3 Humboldt State University, California 4 Other Educational Activities 4.1 Radical Design at UMass Lowell

Carnegie Mellon University

Summer School in Design Research

We will be offering a prototype Design Research Summer School with this curriculum in late May / early June 2007.

This Summer School will target PhD students doing research on the nature of design and students whose research involves a substantial design component. The research objective in the latter case will typically not simply be the design of an X, but rather a methodology for the design of a class of things of which X is an instance.

Spring 2007 Course in Strategies for Research About Design

In preparation for the Carnegie Mellon Design Research Summer School, we will offer a spring course covering much of the same material, with the objective of selecting and organizing materials for both summer school and regular course offerings.

We are principally interested in the sort of utilitarian design typical of engineering, whose goal is practical capability, where cost-effectiveness is a concern, and where the interests of a client or other stakeholders shape the objective.

Here are some characteristics of this type of design:

• Designers in all disciplines are setting up a plan for building new things.

• Design maps from a problem space to a solution space, and it is possible to evaluate and compare solutions.

• Designers optimize in the solution space by techniques such as search and exploration. (Not all designers would accept "optimize", but they do all make choices in the context of some objective.)

• Designers make tradeoffs, often to resolve conflicting constraints. To do so they must understand the objectives, or even better the utility functions, of the clients.

• Designing is not constrained by disciplinary boundaries; it is committed to solving a client's problem by drawing on relevant disciplines as appropriate.

• When exact analysis is not possible, designers use recognition, qualitative understanding, and order-of-magnitude (back-of-the-envelope) estimates.

• Designers consider a set of issues. Not all possible issues matter in a specific design (e.g., geometry and manufacturability are not always pertinent).

• The principal responsibility of the designer is to create a satisfactory solution to the client's actual problem. Tools, techniques, methods, and so on are means to this end.

Course catalog description

17-900 and 39-659 Strategies for Research About Design

TuTh 1:30-2:50, NSH 3002

Instructors: Susan Finger sfinger [at] ri.cmu.edu (Civil Engineering/ICES), Mark Gross mdgross [at] cmu.edu (Architecture/ICES), Jim Herbsleb jdh [at] cs.cmu.edu (ISR), Mary Shaw mary[dot]shaw [at] cs.cmu.edu (ISR/CSD)

The course will compare strategies used in various disciplines for research that advances our knowledge of designs or the process of designing. We will compare the kinds of questions different disciplines recognize as good design research questions, the approaches to answering those questions, and the standards for evaluating the answers. Students will examine designed artifacts and descriptions of the way they were designed to identify the strategies and methods of design; students will also compare paradigms for doing research about design and designing things. We seek students from all disciplines: computer science, engineering, science, humanities, social sciences, arts, business, public policy, etc. This is a proto-course in which the instructors and student will jointly design a curriculum in design research strategies; the initial application for this curriculum will be an interdisciplinary summer school in design research methods.

Open to PhD students. Permission of instructor no longer required. Contact any of the instructors for further information.

Plan for the course

During the semester we will develop a Curriculum Guide for preparing PhD students to do design research. This guide will include a Framework for Studying Design and an analysis of the types of Strategies for Design Research that we observe in the design research community. It will support these with suggestions about ways to organize the material, examples for study, exercises, and an annotated bibliography. The emphasis will be on the charactistics of design research that are common across disciplines.

Students will have the opportunity to understand the character of design research by helping to develop the design framework and analysis of design reserch; this should provide deeper insights that a simple presentation of the conclusions.

Working documents for this course

The Course Calendar. This shows how our activites during the semester will lead us through the raw material and shape it into the Curriculum Guide. The Course Calendar began as a preliminary sketch and is evolving over time as we add notes for each class. Assignments are found here!

A preliminary (and growing) sketch of the Curriculum Guide that we will create in the spring 2006 course. Draft material to incorporate in this guide is in working pages on Topics in Design, on Strategies for Design Research, and on Class Activities.

Candidates for Framework for Studying Design. Several candidate structures are available for organizing the material. During the first part of the course we will use one or two of these structures to direct the review of primary resources. We expect the structures to be refined as a result. The topics in Topics in Design list some of the content that will flesh out these framework candidates.

Preliminary analysis of Strategies for Design Research. We will identify the major design reseaerch communities, examine their journals, and identify and compare research strategies.

Candidates for Designs Worthy of Study. We are looking for great examples of design, either in the form of case studies or in the form of artifacts accompanied by expository material. We are also looking for illuminating examples of design -- which may offer lessons without being great examples -- and for revolutions in the design process.

Candidates for Primary Resources. There is a huge literature of books, papers, sample designs, and other resources. The early part of the course will analyze the relevance of each to the design curriculum so we can select the most effective resources to include in the curriculum. In addition, all the material in the Bibliography section of this Wiki is included here by reference (there's no point in having two copies). The first part of this course will concentrate on reviewing this material to select a subset of reasonable size that is of high quality, covers the area, exposes different points of view and different styles of design research, and is well written. We will screen and annotate these resources in order to identify the ones that are most useful as study materials; all annotations on each resource will be collected on a page in the form of CMU Annotation Template that is linked from the citation plus other pertinent places.

CMU Class Members. This is a set of links to pages describing the design research interests of class members.

Canon of Design. This is a list of big ideas we feel need to be represented in a canonical set of literature about design.

If we have an abstract ready by April 2, there's a software engineering special issue of the Int'l Journal of Engineering Education (IJEE SE Issue).

Here is a link to the web pages for a summer school in engineering design research, which has goals similar to ours. http://www.designconference.org/?menu=301. Summer 2006 was the 8th time the course was taught.

University of Virginia

Kevin Sullivan will offer a graduate seminar course, Concepts for a Science of Design, in the Spring of 2007 at the University of Virginia. The goal of the course is to give graduate students a grounding in efforts to date to develop a theory and intellectual discipline of complex system design, with a particular emphasis on relevance to the design of software-intensive, cyber-physical systems.

Course description

<Everything below this point is notional and subject to change. I.e., "Under Construction">

CS 851 ``Concepts for a Science of Design

Instructor: Kevin Sullivan

Time and place to be determined

The goal of this course is to give graduate students a grounding in efforts to date to develop a theory and intellectual discipline of complex system design, with a particular emphasis on ideas that are potentially relevant to a science of design useful in software-intensive cyber-physical systems design. The bulk of student effort will be devoted to reading books and papers from a range of disciplines, and to discussing this material and its relevance and adaptability to issues in the design of complex software and software-intensive systems. The course is open to graduate students from all disciplines: computer science, engineering, science, humanities, social sciences, the arts, business, public policy, etc.

Prerequisites

Open to PhD students by permission of the instructor.

Topics

• Needs, Designs and Artifacts
• Artifacts and Properties
• Properties and Value
• Complexity
• Coupling
• Modularity
• Hierarchy
• Abstraction
• Crosscutting concerns
• Adaptability
• Exploration
• Evolution
• Integration / Composition
• Representation
• Architecture
• Economics
• Emergence
• Wickedness
• Coordination
• Specification
• Design intent
• Formalism
• Conceptual integrity
• Patterns / handbooks
• Open source / Wikipedia
• Module systems (e.g., ML signatures, Haskell classes, Java interfaces, units, etc)
• Formative work on data abstraction
• Model-driven architecture (methods, notations, tools, etc.)
• Task structure & dynamics
• Methods and processes
• Participatory design
• Ethical dimensions
• Observational studies
• Design reconstruction

Readings

• Mary Shaw and David Garlan. Software Architecture: Perspectives for an Emerging Discipline. Prentice Hall, 1996.
• Carliss Y. Baldwin and Kim B Clarke. Design Rules I: The Power of Modularity. MIT Press, 2000.
• Fred Brooks. Mythical Man-Month (Anniversary Edition). Addison-Wesley, 1995.
• Daniel M. Hoffman and David M. Weiss, (Eds.). Software Fundamentals: Collected Papers by David L. Parnas. Addison-Wesley, 2001.
• Herbert Simon: The Sciences of the Artificial MIT Press, 1996 (3rd edition).
• Michael Jackson. Problem Frames. Addison-Wesley, 2001.
• Eric Gamma, Richard Helm, Ralph Johnson, and John Vlissides, Design Patterns: Elements of Reusable Object-Oriented Software. Addison Wesley 1995.
• Henry Petroski. Design Paradigms: Case Histories of Error and Judgment in Engineering. Cambridge University Press, 1994.
• Walter Vincenti. What Engineers Know and How They Know It. Johns Hopkins University Press, 1990.
• Charles Perrow. Normal Accidents. Princeton University Press, 1999 (updated edition).
• Mark Maier and Eberhardt Rechtin. The Art of Systems Architecting, 2nd ed. CRC Press, 2000 (2nd edition).
• N. Cross, H. Christiaans, and K. Dorst (eds). Analysing Design Activity. John Wiley & Sons, Chichester, 1996.
• The Reflective Practitioner - How professionals think in action. Avebury, Aldershot,UK, 1991.
• Donald Norman. The Design of Everyday Things. Currency/Doubleday, 1990.
• Christopher Alexander, Notes on the Synthesis of Form, 1960 (2nd edition, 2006).
• Daniel Jackson, Software Abstractions: Logic, Language and Analysis, MIT Press, 2006.
• Do and Gross, Thinking with Diagrams in Architectural Design (workshop paper), here.
• Stewart Brand, How Buildings Learn: What Happens After They're Built (Paperback), Penguin (Non-Classics); Reprint edition, 1995.

Humboldt State University, California

Humboldt State University is offering a new course titled The Science of Design for Software Intensive Systems for Spring 2007.

Other Educational Activities

Radical Design at UMass Lowell

George Grinstein, Sarah Kuhn, Fred Martin, and Damon Berry teach a course called Radical design at UMass Lowell. The emphasis is on creativity and innovation, and the reading list reflects this.

"Software development is presently in a crisis. As tool chains continue to evolve, developers are expected to build increasingly complex software systems. Yet, creative and breakthrough applications often only arise from the combination of computing resources from multiple, conflicting domains. A new approach is needed to allow teams of designers—some software engineers, others subject matter experts—to combine their expertise in the design of products based on large, heterogeneous software systems. "