ESRI #Geodesign Summit Day 2 AM
Michael Goodchild began day 2 of the Geodesign Summit noting that “the measure of a meeting is how quickly my own ideas change.” And that, he said was already happening after day one. This, he noted, could be a meeting where a community forms and there is follow up. At this event he said, we have the right mix to make that happen and move geodesign forward.
Listening to Carl Steinitz, Research Professor at the Graduate School of Design at Harvard University, made me feel like I was back in school. Nothing he said was “hard” but it went by fast and I barely had time to digest it before he moved on.
He started with a great quote: “I want to use design as a verb to create design as a noun” and later dropped this pearl: “design is research, make no mistake.”
The heart of the talk focused on what design is: it’s getting from time 1 (situation 1) to time 2 (situation 2). He essentially detailed five ways to do that, with some interesting examples along the way.
(1) anticipatory - The designer sees a whole design and uses deductive logic to get back to the present
The processes was document by Hammond in a 1990 paper on how chefs make up new recipes! It’s in fact how about 50% of design is done, per Steinitz. It actually sounded like something I know I’ve done in other problem solving situations: - you use your “case memory” (stored knowledge) for a close fit to the challenge and modify it for current situation. As you collect feedback from the client and continue to modify the design, you store new info in your “case memory.”
(2) sequential - a series of steps to get where you are going (pretty sure you know path)
Steinitz took his students to Bermuda to design a solution to move an existing dump. The process involved not only talking to the people but in the end a referendum to select the “best” alternative. The process involved collecting projects that would result in some king of goal, then prioritizing them. The final selection included most of the top priorities.
(3) combinatorial - not sure of what to do - lots of options - try to pick best
The idea here is to study top options for each of the factors and make the right choices at the start - there’s more risk if you choose wrong. There is less risk if you choose the wrong path on something less important, further along.
(4) constraining - there are so many options, so the design process involves constraining them into a best choice
The project of redesigning an industrial zone in italy near Padua, ultimately came out of the selection of the lowest level elements - the smallest “bits” of the design.
(5) optimizing - design makers know what they want and have a metric to compare different options - and thus go right to the answer
In La Paz, Mexico, data was poor and resulted in 10 “optimal designs” based on different things being optimized. In the end, these helped to determine what design “not to do” and revealed some interesting patterns (and plans for the mayor to build a resort on the beach front, if I followed correctly). The maps lead to a community “discussion” and that plan was changed.
This whole discussion fried my brain a bit, but in a good way.
Ron Stoltz, Professor and Director, School of Landscape Architecture and Planning, University of Arizona and Karen Hanna, Professor, Landscape Architecture Department, California State Polytechnic Institute, Pomona took on the topic of geodesign in education.
Stoltz started with the statement that if geodesign is a discipline it should be able to be taught. He went on to discuss the mix of formal aspects of a curriculum (courses) and the informal ones, sometimes called the co-curriculum (learning outside of class - field trips, study abroad internships, etc.)
He then went on to detail how you develop a curriculum using a “fast way” (the one I believe we used to develop our MGIS curriculum at Penn State). It relies on defining learning objectives and fitting them into the formal and informal parts of the curriculum. He then went on to detail curriculum domains and other “nitty gritty” of educational design. Why was this presentation included? For me this discussion felt like putting the cart before the horse. I for one am not yet convinced geodesign is a discipline and should be taught. Matt Ball of Vector1 suggested that maybe the idea was simply to discuss how something (in this case a curriculum) is designed.
Hanna gave examples of how the skills of geodesign fit into Bloom’s taxonomy of learning objectives.
She concluded with a valuable list of challenges for those moving forward with geodesign education. They are important to me less as action items and more as indicators of how geodesign is being defined.
- need to teach hand drawing
- explore 3D tools
- deepen database use for “evidence based design” including BIM, integrated practice
- develop better terrain tools
- intergrating other tech
- integrating analysis from science
- integration of vetting tools
Finally, she got to a big question plauging at least some at this event: Is geodesign a unique profession or a skill shared by many?
Chris Overdorf, Principal, Jones & Jones, Architects, Landscape Architects, and Planners ended the keynotes with a discussion of how to take geodesign to individuals for consumption. He argued that we need to take such design to the people because public land protection is not enough. His example involved a pilot effort to link two conservation properties in Washington state separated by 8 miles via a path/trail using public land or land with easements. He detailed the technical details (ModelBuilder) and human interactions (tapping in to local knowledge).
Wacom’s Mike Dana and a team from ESRI showed off the company’s pen on screen interface tools used in a fire response situation. They sketched on a display to add new data to a map service which was then displayed in a browser.
Steve Irwin of Harvard’s GSD spoke on the importance of diagrams. They have relationships to real world stuff (like maps and sketches) but are different: high levels of abstraction, emphasize elements and relations (like topology) and are used to explain or argue. Diagrams can be derived from and can become maps and plans. A truly great short presentation.
John Przybyla, a VP at Woolpert discussed vertical design and its technology of choice (BIM) and geodesign. He highlighted the difference between the design community’s closed (fully defined and detailed) IFC classes for BIM and the geospatial community’s more open BISDM. He went on to discuss use cases for integrating BIM and GIS and the challenges therein.
Brad Ball from NASA spoke about space allocation planning at NASA Langley.
Prashant Hedao, Auroville Planning and Development Group spoke about the Auroville Universal Township, an experimental township built on a 1968 “galaxy” (spiral armed, 4-zone) plan.
Dennis William, Civil Design Team, LLC shared his experience with live field design (LFD) using ArcPAD. The idea of LFD was driven by business needs: saving time for survey, engineering and reduced construction schedule and thus costs. Basically, you get the design to the field with the owner and make the changes there. Needed: geodesigner, project scope, GIS maps and data, surveys and plats. You input local design requirements, dump data to ArcPAD/ArcMap, view and work with development team in the field. William noted he’s been going geodesign for 11 years.
Andy Bennett, Telvent addressed geodesign in electric utility.
Shannon McElvaney, of Pacific GPS detailed GIS use in designing a sustainable city, in this case Masdar City in Abu Dhabi. Goals: carbon neutral and zero-waste. Among other things designers look to older cities for ideas and looked at orientation to the sun for better cooling. He finished off with a visualization in “6D” (3D + time + money + carbon).
Jason Lally of PlaceMatters discussed what he called high fidelity 3D landscapes. We have good 3D tools but still have gaps, so the company is looking for tech to fill the gap. He note CityScape as one possibility that can enhance visualization, encourage feedback and shorten the process to save money. The takeaway: we need to look at these tools to enable clear communication with an input by the public.
Paul Cote of Harvard spoke on his favorite topic: CityGML which he suggests: enables loss-free exchange of place-based knowledge between the tools of specialized content domains.
ESRI helped cover some travel and lodging costs related to Directions Media’s attendance at the Summit.