Futurescaper is an online tool for making sense of the drivers, trends and forces that will shape the future. As a user interface system, it still needs development. As a tool for analyzing and understanding complex systems, it works very well and does something I have yet to see anything else be able to do. Several people asked me about this after my last post, so here is some more detail.
Following the logic of collective intelligence (as part of my my PhD), I broke up the the scenario thinking process into discrete chunks, came up with a system for analyzing and relating them together, and then distilled them into key outputs for helping the scenario development process.
Emergent Thematic Maps
One of the coolest things about Futurescaper is how it translates simple input into complex analysis, and then back again into simple insights.
To demonstrate this, I tested the system using data from an International Futures Forum project on international climate change impacts for UK Foresight. This project identified hundreds of scientific papers, journal articles and press clippings that were relevant to climate change impacts on the UK. With help from an intern, I dumped 186 of these into Futurescaper to see how it would work. This turns out to be quite useful for the scenario creation process, particularly for generating and ranking key drivers and uncertainties.
Revealing Hidden Connections
One of the outputs is the following map. This illustrates the high-level thematic relationships between subjects in the database. Each circle represents a cluster of drivers and issues. The size of the node reflecting its importance in the system. Links (and link thickness) describes relationships between the thematic clusters. Remember, there were 186 interlinked drivers and forces in the original data set, with thousands of connections between them. The system is based on the exact same procedures used in this study, which reveal the linkages between over 6.4 million citations in over 6,000 different scientific journals.
My first reaction was, “Oh great, how banal.” On further reflection, this is actually an interesting way of uncovering high-level themes and issues amongst a very complex, multi-dimensional set of issues. Thinking scenarically, six major drivers and uncertainties are more than adequate to create meaningful variation. And if you can support the choice of these drivers and uncertainties through online evidence, it is even better.
Some interesting observations emerge from the relationships between these themes, as well. The strong link between climate change and economic growth is an obvious one, but the entire “acceptance” cluster, which seems to be about political power and the ability to collaborate, was a total surprise.
It’s not so surprising that “climate change” comes out as a cluster though, given the focus of the data. So how can we get more useful insight from this?
Thankfully, each of the high level clusters above are made up of underlying sub-clusters as well. The words below the titles represent some of these sub-issues, as you can see. But with a little bit of tweaking of Futurescaper, Mapinfo.org lets you drill down to the detail of the sub-clusters to get a more revealing picture of their underlying dynamics.
Drilling down into the “Climate Change” cluster at the center, for example, reveals the following connection of issues and drivers.
This graphic shows all of the drivers within the “climate change” sub-cluster. It reveals the tremendously complex inter-linkages between them. The bigger the circle, the more influential the force, and the darker the line, the more interaction between that force and its connections.
This is a bit overwhelming, but is useful to illustrate the system complexity involved – even with only 186 discrete data points. Things get more useful when you filter down to the top 20 influences on the system, as below.
Even here, you can see that there are only about 6 or 8 major forces and issues at play. The most influential, according to the data set, are “decreasing water availability” and “decreasing water quality”. Other strong factors driving climate change impacts include:
- Increasing toxic algal blooms
- Decreasing improved water and sanitation
- Increasing diarrhea
- Decreasing agricultural productivity
- Increasing pollution
Even this output is a bit hard to read. As a ranking and sorting mechanism, it seems to work pretty well. I didn’t code these by uncertainty, but it is easy to see how you could filter these by “relatively certain” influences and “critical uncertainties”, per traditional scenario planning methods. These could then be used on their own or in a traditional scenarios workshop context.
Making Sense of the Story
In order to make its meaning more clear, I re-arranged the graphics to present an actual systems map, as above.
Now we can tell an interesting story. Starting with the big blobs on the upper right, for example, you can see how “Decreasing water availability” impacts a range of things like decreasing crop yields, increasing migration and increasing hardships for women. The small-ish blob above it, “Increasing droughts”, leads to decreased agricultural productivity, increasing food prices, and several of the other outcomes described above. “Increasing flooding”, for example, also leads to decreased water availability, increased migration, and increased contamination of the water supply.
Building scenarios around the future of climate change, for example, could use the drivers of decreased water availability and decreased water quality to tell a story about their causes and secondary impacts. You could show how these were driven by increasing flooding, and then spin a narrative around increasing algal blooms, infectious diseases, crop failures and hardships on women.
You could then link this detail with the other, higher level themes explored in the first map. There could be an interesting meta-story around increased economic and population growth, for example, combining with lack of political will to change. This could “set the stage” for forced climate change and all of the details described above, thereby providing rich ideas for detailed scenario development and story-telling.
Summing It All Up
Futurescaper is a prototype platform for collecting drivers of change from distributed sources and linking them together. It lets people explore these trends very easily, get a sense for which are the most influential and then build an understanding of how it works as a system.
This can be useful as both a high-level scenario generation tool and for more detailed scenario development. Although the UI is horrible, the system is incredibly quick. Thanks to Martin Rosevall’s online network mapping plug-in, these maps and rankings take minutes to create. Also, in theory, this step could be automated for real-time scenario generation and exploration.
In conclusion, the system is to augment the traditional scenario-building workshop, not replace it. It offers a way to engage a much larger audience in the scenario creation process, it speeds up the drivers ranking process and it provides increased transparency in the choice of key drivers. It also allows for interactive exploration of complex trends and patterns that could be difficult or impossible in the traditional workshop setting.
Over time, there are lots of opportunities for the development of Futurescaper. One is to create a more flexible user interface for specific projects. Significant improvements could also be made on the data entry and exploration process. Finally, a semi-automated data harvesting function from Twitter and Google would be amazing for key topics. Imagine setting up a Twitter account that automatically entered fragments into the system based on key hashtags, for example, or connecting Futurescaper to S4’s awesome Momentum package.
Finally, thank you to the many colleagues who have contributed directly to this; most notably Nathan Koren, Graham Leicester, Tony Hodgson, Daniel Wahl, and Martin Rosvall. If interested in learning more, please contact me directly or follow me on Twitter.