�ݺ�ߣ

Editor's Notes

• #2: Using Graph Theory to inform Placemaking Big Ideas in Place, Place Leaders Asia Pacific Conference, Vivid Ideas, 30 May 2017 A copy of the transcript plus reference links is available at: https://www.linkedin.com/pulse/using-graph-theory-inform-placemaking-john-dobbin
• #3: Back in the sixteenth century, the residents of K?nigsberg, Prussia, spent many a fine Sunday afternoon trying to solve an interesting problem: how could they start at one point, cross each of their city's seven bridges once and only once, and arrive back at the same point. Many a cobbler was kept in business as the residents wore out their boots trying every imaginable route.?
• #4: Eventually, the brilliant mathematician Leonhard Euler put them out of their misery, and several cobblers out of business, by proving that it was mathematically impossible. (Euler was no ordinary mathematician, by the way; back then they used to refer to equations as either being before or after Euler).
• #5: Euler stripped away the clutter and viewed the problem in its simplest form: destinations and paths (or vertices and edges in the mathematical lingo). He determined a general rule: only if destinations are connected to an even number of paths can such a circuit be made. This is now called a Euler Circuit.?
• #6: He unknowingly gave birth to the field of mathematics known today as graph theory, which is used in fields as diverse as linguistics and condensed matter physics.? The founders of Google used it to score the relative importance of web pages and became billionaires
• #7: My colleagues and I are using it for highly precise targeting of marketing messages.
• #8: How might it be used for Placemaking? Well, we can use it to uncover patterns of human interaction, and see if we can improve upon those patterns. For example, in designing innovation districts we should be interested in the flow of knowledge and resources around a region.?
• #9: Studies of innovation ecosystems show that better results are obtained when random collisions between people that would not ordinarily meet increases. 'Collisionable hours' has been suggested as a metric.
• #10: I don't actually like the term 'Collisions', it seems a misnomer. What we are trying to facilitate is knowledge spillovers. This is when knowledge flows beyond its normal boundary and into another domain. For example, by taking a calligraphy class, Steve Jobs was able to absorb knowledge about typography and ten years later pour it into computer design. We might not have multiple typefaces or proportionally spaced fonts today if it wasn't for that spillover.
• #11: It takes time, and trust, and play for tacit knowledge to spill from one brain into another. It very much favours face-to-face socialisation?(alcohol sometimes helps too). Therefore I posit the term "interaction hours" as a better metric for innovation precincts.
• #12: In societies, clusters of activity tend to form along teleological boundaries, around our goals. For example, nano-technologists tend to cluster around quantum physics stuff, while rich people tend to cluster around hedge fund and property development stuff. Rarely do they intersect. Rarely do they exchange knowledge. This is a problem for innovation.
• #13: From a network design perspective, we want to create nodes that connect clusters together. These are said to have a high betweenness centrality or high importance. They shorten the distance between points on the outer branches and create flow between otherwise isolated parts.
• #14: For example, Incubators and VCs can bring nerds and rich people together.
• #15: E.g, ATP Innovations incubates over 70 start-ups and has helped raise for them over $150 million. It is very successful.
• #16: There are many other network assets that might contribute to a vibrant ecosystem: kickstarters, network builders, investment bridges, accelerators & incubators, international bridges, community builders, industry-research bridges etc -- programs and places that bring parts of the network together.
• #17: When these start to form connections between each other, it is referred to as a "rich club phenomenon". These are highly desirable because there develops a level of resilience. If one node fails, the network does not collapse.?
• #18: New data sets are becoming available all the time. Indeed, we are all walking around with data collection devices in our pockets. In addition, new tools are being released to make it easier to interpret the data. We can quite easily build little apps or programs to uncover patterns. For example, very simple graphs of innovation districts can tell us a lot. They tell us the types of activity being undertaken in a location, what type of services are taking shape, who the investors are and where workspaces can be found. Importantly, they can show what is missing from an area.?
• #20: My hypothesis is that the better we understand the network of human interactions, the better we can design improvements. Just as Place once informed mathematics, maybe mathematics can now inform Place.?
• #21: If you are involved with Placemaking, Precinct, City or Regional Planning and would like to understand more about how Data Science can inform the design process, Link In and let's chat. A copy of the transcript plus reference links is available at: https://www.linkedin.com/pulse/using-graph-theory-inform-placemaking-john-dobbin