Brown Bag Lab Lunch Series

DESIGN COMPUTING RESEARCH FORUM

Spring 2004, Tuesdays & Thursdays

12:00 NOON - 1:30 NOON @ Architecture Hall 043

add a new talk to schedule


3/30 (Tuesday) Welcome Back!

Happy Spring!

edit this talk read reviews of this talk add a review of this talk


4/1 (Thursday) Ken Camarata and Orit Shaer, People Pretzel

People Pretzel is a computationally enhanced game board that aims to encourage informal social interaction in public spaces. The People Pretzel uses sound, light and visual images to entice community members to participate in playful group interaction. To play, the players are required to collaborate, using their bodies as playing pieces. This physical interaction is computationally mediated and augmented to produce an improvisational multimedia performance for the enjoyment of players and spectators alike.

Ken Camarata is a Ph.D. student in the UW's College of Architecture and Urban Planning program. He is also a part-time lecturer for the Department of Architecture and a researcher in the Design Machine Group.

Orit Shaer is a Ph.D. student in Computer Science at Tufts University. She is currently a visiting student at the Design Machine Group.

edit this talk read reviews of this talk add a review of this talk


4/6 (Tuesday) Steven Drucker, An Overview of the Next Media Research Group in Microsoft Research

The Microsoft Next Media Research group focuses on exploring what new consumer media experiences are possible with the growth in computing power, connectivity and storage in a compelling, elegant and transparent way in the 3 to 10 year timeframe. The group envisions consumer information and entertainment experiences not available today and builds or combines technologies from other Microsoft Research groups and crafts intuitive user interfaces and compelling scenarios to deliver that experience. Rather than focus on old media or new media, the group attempts to develop working prototypes of the Next forms of media possible from new convergent technologies, hence the name.

Next Media's research focus spans the linear and interactive media spectrum from television, broadband, and gaming to combinations of traditional media forms or emerging media forms too new to have a name. The group partners with product groups within Microsoft and select outside content partners such as museums, and public broadcasting to develop conceptual and working prototypes which best demonstrate the functionality or vision for media that can anchor new businesses, services and consumer experiences.

Steven Drucker will give an overview of the group, answer questions about Microsoft Research and show some recent projects.

Dr. Steven M. Drucker has been the lead researcher for the Next Media Research Group in Microsoft Research for the last 4 years where he has been looking at how the addition of user interaction transforms conventional media. He is particularly interested in database visualization for consumers or where art meets technology for user interfaces. While in the group, he has filed 20 patents on technologies as diverse as remotely operated personal video recorders, spectator oriented gaming, and new visualization techniques for media databases.

Previously he was the lead researcher in the Virtual Worlds Group also in Microsoft Research. During his tenure there he helped architect a platform for multi-user virtual environments, filed 12 patents, and published papers in subjects ranging from architectures for multi-user, multimedia systems to online social interaction.

Before coming to Microsoft, he received his Ph.D. from the Computer Graphics and Animation Group at the MIT Media Lab in May 1994. His thesis research was on intelligent camera control interfaces for graphical environments. Dr. Drucker graduated Magna Cum Laude with Honors in Neurosciences from Brown University and went on to complete his masters at the Artificial Intelligence Laboratory at MIT doing research in robot learning.

His published papers have been in such areas as multi-user environments, online social interaction, hypermedia research, human and robot perceptual capabilities, robot learning, parallel computer graphics, and human interfaces for camera control.

His website is at: http://research.microsoft.com/~sdrucker/
http://research.microsoft.com/vwg/people/sdrucker.htm

edit this talk read reviews of this talk add a review of this talk


4/8 (Thursday) Ellen Yi-Luen Do and Mark D Gross, Three R's of Drawing and Design Computing

A drawing centered view of design process focuses on the interplay between designer expertise, domain knowledge and media manipulation. We report on our computational sketching software systems that support design recording, reasoning, and resolving.

(1) Recording activities concern design and knowledge capture. At the most basic level we record the user's pen input strokes, pressure and speed using the Electronic Cocktail Napkin as a platform for diagram parsing and understanding. The Design Amanuensis and Design Recorder projects extended the capture to include speech and text and provide synchronized history for reply and search. To support shared drawing, annotation of design rationale, collaboration and negotiation, we developed the Immersive Redliner and Space Pen projects for design drawing in 3-D and also synchronous shared whiteboard drawing with design history in NetDraw.

(2) Reasoning activities in design engage domain knowledge in various ways–including the performance behavior of a proposed design, retrieving relevant design precedents or cases from a library or database, stimulating creative thought through reasoning-by-analogy, and so on. We have implemented several software systems to support design reasoning in the form of connecting intelligent systems of domain knowledge to provide design feedback. The Napkin's visual language parsers identify the context of drawing elements and configuration. It also supports recognition of perceptual figures such as emergent shapes from overlapping figures. The Stretch-a-Sketch program maintains interactive behavior among drawing elements using constraint propagation. IsoVist, Design Evaluator and Napkin-Archie programs support intelligent feedback such as performance analysis, critiquing and retrieval of relevant cases from a database. Light Pen offers design decision support guided by lighting design practice rules. LAN-Tools activates a simulation of network communication and proposes modifications based on designer's hand drawn diagrams.

(3) Resolving concerns the development of drawing over the design process from rough, sketchy, and abstract representations to specific, definite, and well-defined design proposals. Drawing can be used to beautify and specify design (as in Electronic Cocktail Napkin and WebStyler) or translated into 3-D visualization with VR-Sketchpad. The incremental formalization includes recognition of drawing symbols and configurations, to beautification of rectified figures, including the substitution of the diagram elements to specification drawing illustrated in the Electronic Cocktail Napkin.

Details and information about each project mentioned can be found on the Design Machine Group web site http://dmg.caup.washington.edu

edit this talk read reviews of this talk add a review of this talk


4/13 (Tuesday) Eun Soo Lee, Wearable Sketch Assistant

I propose a wearable design assistant with AR (Augmented Reality) which helps architect (1) get early stage design information on real tangible media and (2) to check his sketch on it according to rule-based system in real time. With this wearable system, architect can be released to work in front of computers
and he can sketch on any material, at any time, in any place, and with any traditional and conventional tools.

Current architects are more accustomed to sketch their design on ordinary paper, tracing paper, white board, and any other tangible material than electronic tools because they are not much experienced in working in front of a computer in the early phase of design. In addition, in order to use a computer as an assistant they put design data into a computer. It is troublesome for us to transfer design data on sketch material to a computer because some mediums such like camera and scanner are needed for the transfer of sketch data to a computer.

Unless letting architects work in front of a computer, we have to send a computer in front of them. In other words, architect can use existing tool such like paper and pens with wearable design checking system. Wearable design Assistant seems to be needed for architects while they are designing in the early stage of architectural design because it provides them with instantly visual supporting vision using wearable AR (Augmented Reality) system able to show architect design errors in real time.

This system has 5 major steps

1. When an architect draws sketch on actual paper, a head mounted mini camera continuously captures the sketch image.
2. The captured sketch image is transferred to a portable computer.
3. Portable computer vectorizes the image and turn it into semantic data according to large letters and outlines on sketch image.
4. Computer checks architect¡¯s design errors of space zoning.
5. Wearable display system (AR: Augmented Reality) shows architect design errors mapping virtual image on the actual paper. Virtual text annotations will
be used as error contents.


Eun Soo Lee is a student in the Master of Science in Design Computing program at the University of Washington.

edit this talk read reviews of this talk add a review of this talk


4/15 (Thursday) Golnaz Mohammadi, Geometric Shape Generator

Geometric Shape Generator (GSG) is a design tool for producing geometric shapes using simple programming codes algorithmically. The algorithm defines a shape through a mathematical equation. GSG is intended for architects and designers with minimal programming knowledge. GSG aims to enable architects and designers to generate geometric forms by altering or modifying only the numeric values and basic variables in existing and simplified codes. Architects and designers utilize GSG for conceptual form development throughout the design process. GSG includes an archiving system that stores programming codes in the Processing environment that will be used to generate geometric shapes at a later time. Basic shapes, such as points, lines, curves, rectangles, circles, and triangles; already exist within the Processing environment. GSG differs from the Processing environment in the sense that it can produce more than just the basic geometric shapes.

Golnaz Mohammadi is a M.S. in Design Computing student at the University of Washington.

edit this talk read reviews of this talk add a review of this talk


4/20 (Tuesday) Bill Beaty, Unwise Microwave Oven Experiments

Since he worked on some microwave oven physics in 1985 for an encyclopedia entry, the author has been collecting interesting "lecture demonstrations" easily performed with any kitchen microwave oven. The Lunch Lecture will be a series of these demonstrations, some appearing on the website listed below, plus a few that are a bit too dangerous for the general public. See http://amasci.com/weird/microexp.html for more information.

Bill Beaty is a research engineer in the University of Washington's Department of Chemistry. He is also a science enthusiast and maintains the website Science Hobbyist.

His website is at: http://amasci.com

edit this talk read reviews of this talk add a review of this talk


4/22 (Thursday) Claudia Valdes, Measuring Time and Inducing Memory

Claudia Valdes will be talking about her research from the past 4 years investigating themes of time, memory and perception. In particular, she will focus on her work in digital video investigating the history of nuclear arms devolopment.

Claudia X. Valdes is an intermedia artist concerned primarily with issues of time, memory, and perception. Recent projects in have been realized in formats of experimental cinema and digital video installations investigating the history of nuclear arms development. She was born in Santiago, Chile, in 1972. Valdes received an MFA and was awarded the Eisner Prize in Art from the University of California, Berkeley in 2001. Her work has been exhibited internationally: at the Institute of Contemporary Arts, London; MCA Chicago; WRO Center for Media Art, Wroclaw, Poland; the Armory Center for the Arts, Pasadena, CA; the UCR/California Museum of Photography; Art in General, New York; Centro Multimedia/Centro National de las Artes, Mexico; and the Werkstätten und Kulturhaus, Austria. Valdes taught digital media at UC Berkeley in 2001, was an Affiliate Artist at the Headlands Center for the Arts from 2001-2003, and is a Research Associate in the Center for Digital Arts and Experimental Media at the University of Washington, Seattle for the 2003-2004 year. At UW, Valdes will be teaching core curriculum theory and studio based courses from which the first DXARTS majors will be selected.

Her website is at: http://www.claudiaxvaldes.com

edit this talk read reviews of this talk add a review of this talk


4/27 (Tuesday) Alex Anderson, Human Limb Objects and the Law of Mechanical Selection

In the early 1920s Le Corbusier and Amedee Ozenfant introduced “The Law of Mechanical Selection” which rules the evolution of human-made objects in the industrialized world. One set of highly-refined objects that interested them particularly included what they called “human-limb objects”, that is, objects that extend human capabilities -- tools. They proposed that such objects play an essential role in the shaping of modern environments, with the implication that the creative process of design at all scales is bound not only to the human body but also to ineluctable processes of manufacture and mass marketing. My talk will examine these claims, their applicability at the time Le Corbusier and Ozenfant made them, and their potential relevance now.

Alex T. Anderson, Ph.D., is an Assistant Professor in the Department of Architecture at the University of Washington, where he teaches in the areas of architectural history, theory, representation, and design.

Professor Anderson received his undergraduate education in Civil and Environmental Engineering, with an emphasis on building structures, from Cornell University in Ithaca, New York. He received his M.Arch., M.S.Arch., and Ph.D. in Architecture from the University of Pennsylvania in Philadelphia. He developed his doctoral thesis, Modern Architecture and the Commonplace, under the guidance of Professors David Leatherbarrow, Joseph Rykwert, and Marco Frascari.

Before assuming his current position in the fall of 1998, Professor Anderson taught architectural history, theory, and design for two years in the College of Architecture at the University of North Carolina at Charlotte. From 1993 to 1995 he taught building structures and architectural theory as an Adjunct Professor of Architecture at the Philadelphia College of Textiles and Science. During those years he also taught architectural drawing studios in the Graduate School of Fine Arts at the University of Pennsylvania.

Professor Anderson is currently working on an annotated translation of Étude sur le mouvement d'art décoratif en Allemagne, written in 1912 by Charles-Edouard Jeanneret (Le Corbusier). He has presented papers on architectural history, theory, representation, and pedagogy at conferences in the United States and Europe. He acts regularly as a peer reviewer for conference papers and has contributed a number of book reviews to academic journals.

edit this talk read reviews of this talk add a review of this talk


4/29 (Thursday) Markus Eng, FlexM

Architectural designers often struggle with standard CAD software interfaces because manipulating the screen user interface with a mouse bears no tangible relationship to their training in manual drafting and physical model building. The absence of tactile feedback from the digital model and the shift from the gestural expression of sketching to the commitment of action further strips away architectural design for the
architect. Finally, the pleasure of creating is diminished due to the loss of haptic stimulus, which alone can be the source of satisfaction, independent of the design process.

FlexM, a computationally enhanced construction kit, addresses these deficits of conventional modeling interfaces by integrating computer graphics with construction toys. Despite their strength as a tool for productivity and creativity, computer graphics applications require a steep learning curve to master them. Toys, on the other hand, are easy to use, intuitive and fun. Construction kits, like K'nex(tm), LEGO(tm) Technic, or Zometools(tm) are designed for all ages, have flexible, moving pieces, and allow endless combinations for creativity [2].

FlexM is a flexible physical interface for manipulating and generating digital models, similar to the pioneering research by Aish [1] and Frazer [3]. In place of clicking and dragging with the standard mouse interface, the designer builds the digital model using FlexM hubs and struts. With flexible joints on the hubs, the user can sculpt models that transform into a dynamic form. The user builds a cube with the hubs and struts. An image of the box is rendered on the screen. The user then squishes the corners of the cube, deforming it. The corresponding changes update on the monitor.

The FlexM hubs are connected to microprocessors, which collect and send the model's topology and geometry to a recipient graphics application. Each hub has high intensity LEDs, which cast light through an aperture in each of the sockets. The connecting, acrylic struts direct the light from one hub to a photosensor in the attached hub. Light is the medium used to identify the connections between the hubs. The communication protocol consists of each hub taking turns casting light to the other hubs. The other hubs in return, check their light sensors to identify if they see the "casting hub." The topology of the model can then be deduced from this coupling protocol. The geometry of the model is calculated from the angles of the connecting struts. Bend sensors or potentiometers measure the bending angle at the flexible joint.

Markus is a graduate student in the Master of Architecture Program at the University of Washington.

Info about FlexM can be found at http://students.washington.edu/markuse/FlexM/00flex.html

edit this talk read reviews of this talk add a review of this talk


5/4 (Tuesday) Golnaz Mohammadi and Babak Ziraknejad, Geometric Shape Generator and Tangible Interfaces at Home

Golnaz Mohammadi will present on Geometric Shape Generator (GSG). GSG is a design tool for producing geometric shapes using simple programming codes algorithmically. The algorithm defines a shape through a mathematical equation. GSG is intended for architects and designers with minimal programming knowledge. GSG aims to enable architects and designers to generate geometric forms by altering or modifying only the numeric values and basic variables in existing and simplified codes. Architects and designers utilize GSG for conceptual form development throughout the design process. GSG includes an archiving system that stores programming codes in the Processing environment that will be used to generate geometric shapes at a later time. Basic shapes, such as points, lines, curves, rectangles, circles, and triangles; already exist within the Processing environment. GSG differs from the Processing environment in the sense that it can produce more than just the basic geometric shapes.

Babak Ziraknejad will present Tangible Interfaces at Home (TIH). The THI project is concerned with the development and integration of computationally enhanced interactive devices and smart environments. Components of TIH include "Window Seat", "Twister Game", "Energy Conservation Tutor", and "eFrame". The Window Seat is a computationally augmented rocking chair that operates as an interface to control the pan and tilt motion of a web cam in a virtual or remote location. The Twister Game is a computationally augmented interactive game board that encourages collaborative interaction with sound, lights and images. Energy Conservation Tutor is an ambient display designed to promote awareness of energy consumption levels of the house. eFrame is a large ambient information display designed for a quick glance and overview of information. This ordinary wall frame is equipped with intuitive command buttons and is positioned in a common area of the house to promote family collaboration and information sharing. Additionally, eFrame automatically detects family member's arrival and then displays their online calendar for review.

Golnaz Mohammadi and Babak Ziraknejad are M.S. in Design Computing students at the University of Washington.

edit this talk read reviews of this talk add a review of this talk


5/6 (Thursday) Greg Bowman, Users as Designers: The Influence of User Feedback on Software Design

Video Traces is a system that makes it easy to capture a piece of rich digital video and annotate it both visually (using a pointer to record gestures) and verbally. The resulting product is a "video trace": a piece of media plus its annotation--in essence, a recorded "show & tell". Traces can be viewed by their creator, exchanged with others, and further annotated for a variety of teaching and learning purposes.

The project is the work of UW's Program for Educational Transformation through Technology. A key part of PETTT's mission is to implement the projects they develop in classroom situations and study how the technology changes and improves the learning within that environment. The results of that research in return influences future changes in the technology.

Video Traces is designed and coded by PETTT's software developer Greg Bowman. Greg will talk about the history of the project, and the important effects the use of the tool in different learning environments has had on the evolution of the the software.

Greg Bowman is a Multimedia Designer at PETTT. More information on PETTT is available at: http://depts.washington.edu/pettt/

edit this talk read reviews of this talk add a review of this talk


5/11 (Tuesday) Terry Brooks, Semantics and the Web

'The Semantic Web' expresses for the web environment the long-standing ambition of global semantics -- people everywhere would contribute information to the Web and web bots could harvest meaning from the Web. The success of this agenda depends on the cultural frame of the information technology. The original cultural frame of the Semantic Web initiative has been to extend legacy information technologies such as the Dublin Core Metadata Set. In this cultural frame, meaning is asserted on the open web. Recent experience, however, shows that the lack of trust in the open web makes meaning assertion indistinguishable from spam. An alternative approach aggregates semantics from the open web. Aggregators such as Google parse HTML pages for semantic content and use hyperlinks as a plebiscite for the most important web pages. Aggregation strategies create a culture of lay indexing where spam is controlled by maintaining the secrecy of parsing algorithms. It may be that Google has already created a semantic web that is appropriate for the open web while closed systems, such as digital libraries, are appropriate for the closed web.

Terry Brooks is an Associate Professor in the Information School, University of Washington. He has a Ph.D. from the University of Texas at Austin, a M.B.A. degree from York University and a M.L.S. degree from McGill University. He concentrates on the construction and presentation of information with web technologies. His most recent publication is "The Nature of Meaning in the Age of Google" to appear in Information Technology, April 2004.

His vita is located at
http://faculty.washington.edu/tabrooks/vita/vita.html

His website is at: http://faculty.washington.edu/tabrooks

edit this talk read reviews of this talk add a review of this talk


5/13 (Thursday) Susan Finger, Team-Based Design: Collaborative Learning Across Time and Space

Rapid advances in mobile computing and wireless communication present an opportunity for fundamental changes in engineering education, particularly for team-based courses in which collaborative learning is the dominant learning mode. These advances can be exploited in two ways: first, by integrating research on mobile computing into graduate and upper level undergraduate engineering design courses; second, by having the students in these courses work in teams to develop new applications for mobile computing, particularly applications that help them collaborate on their team-based projects. By working in multidisciplinary teams to develop collaboration tools using mobile computing and wireless communication, the students become involved in state-of-the-art research and they facilitate their own learning by developing the collaboration tools they need.

Our goal is to facilitate computer-based collaborative learning in project-based design courses by developing collaboration tools for mobile computers. These tools also enable faculty to monitor and understand where and how communication breaks down during the design process. Successful design teams are those that communicate frequently and keep all teammates up-do-date on the latest developments; that is, successful teams are teams that are successful at collaborative learning. Developing tools for computer-supported collaboration not only enables more student teams to be successful, but also enables more courses to use the power of collaborative learning.

edit this talk read reviews of this talk add a review of this talk


5/18 (Tuesday) Vibha Sazawal, Helping Software Engineer's Assess Ease of Change

The creation and maintenance of software is a complex activity involving social, economic, and technical factors. Software engineering researchers seek technical solutions to software engineering problems that are compatible with the social and economic realities of building software today. Technical approaches show promise for both reducing programmer effort and increasing confidence that software artifacts possess desirable properties.

In my research, I am interested in the software design property of "ease of change''. The software engineering community has numerous design rules intended to result in software that is easy to change, but strict adherence to these rules in practice can be difficult. I have built a tool that analyzes software to produce partial design representations called "Design Snippets''. Design snippets are intended to help programmers identify problems related to ease of change as they code.

Vibha Sazawal is a Ph.D. candidate in Computer Science at the University of Washington.

Her website is at: http://www.cs.washington.edu/homes/vibha/

edit this talk read reviews of this talk add a review of this talk


5/20 (Thursday) Dieter Fox, Distributed Multi-Robot Exploration and Mapping

Efficient exploration of unknown environments is a fundamental problem in multi-robot coordination. As autonomous exploration and map building becomes increasingly robust on single robots, the next challenge is to extend these techniques to large teams of robots. In this talk I will give an overview of our approach to multi-robot exploration and mapping, which is part of the CentiBOTS project. This project aims at fielding 100 robots in an indoor exploration and surveillance task.

A general solution to distributed exploration must consider some difficult issues, including limited communication between robots, no assumptions about relative start locations of the robots, and dynamic assignments of processing tasks. The focus of this talk will be on our current solutions to the problems of robot localization, map building, and coordinated exploration. Most of these solutions rely on particle filters for efficient state estiamtion. I will also describe a hierarchical Bayesian approach that enables robots to estimate the structure of an environment during exploration. Our experiments indicate that structural information can greatly improve the reliability of multi-robot map merging.

Dieter Fox, Assistant Professor, joined the faculty in the fall of 2000. He grew up in Bonn, Germany, and received a B.Sc. in 1990, an M.Sc. in 1993, and a Ph.D. in 1998, all from the computer science department at the University of Bonn. Before going to UW, he spent two years as a Postdoc in CMU's robot learning lab. Dieter Fox heads the Robotics and State Estimation Lab which he established in January 2000.

His research interests lie in artificial intelligence and its application to mobile robotics. He believes that building systems is an important part of research, especially in robotics. To demonstrate the reliability of probabilistic methods for mobile robot navigation, he deployed with a team of researchers the robots Rhino and Minerva as tour-guides in two populated museums (one of them the Smithsonian's National Museum of American History in Washington, D.C.). More recently, he and colleagues introduced particle filters as a powerful tool for state estimation in mobile robotics.

His website is at: http://www.cs.washington.edu/homes/fox/

edit this talk read reviews of this talk add a review of this talk


5/25 (Tuesday) No Class

No class today.

edit this talk read reviews of this talk add a review of this talk


5/27 (Thursday) No Class,

No class today.

edit this talk read reviews of this talk add a review of this talk


6/1 (Tuesday) Ellen Do, Report from Advanced Visual Interface Conference

Ellen Do will give a short presentation on "HomeLab" and "Ambient Intelligence" work done by Philips Design Research which she learned about at the Advanced Visual Interface Conference (May 25-28,2004). The conference website is at: http://www.di.uniba.it/~avi2004/

edit this talk read reviews of this talk add a review of this talk


6/3 (Thursday) Last Lab Lunch,

Reflections, reports, discussions of this quarter's lab lunch talks, etc.

edit this talk read reviews of this talk add a review of this talk