Dispatches from SFI’s Interplanetary Festival

Image Credit: MR.TinDC from Flickr

Last week’s inaugural Interplanetary Festival — organized by the Santa Fe Institute brought together a range of experts from multiple scientific backgrounds and artistic disciplines to discuss the key theme of the festival — What challenges does humanity face in order to achieve Interplanetary Travel?

I decided to attend this festival as a way of introducing myself to the study of complex adaptive systems. It struck me that this would be a prudent step given the types of challenges society is facing today; unintended side effects of digitized societies, runaway capitalism and global climate change. I am particularly interested in learning and understanding if it’s possible to apply some of the same thinking to the new breed of digital complex adaptive systems we are building today, particularly machine learning based systems.

The festival’s interplanetary theme was chosen to be provocative and to frame the question from a higher level perspective;

  • What challenges does society face when considering off-world space travel and habitation?
  • What do we need to build here on earth in order to achieve the former?

The lineup of this festival was aligned with the mission of SFI itself — to seek and uncover knowledge about the dynamics of complex adaptive systems that pervade our environments and society at large.

This summary post mainly focuses on the talks that I attended. As with many panel discussions — there is no central narrative, so my notes mainly focus on the most thought provoking aspects of the discussion.


Panel: Sandboxes to think and play with

Shawn Douglas & Chaim Gingold

Simulations make invisible phenomenons visible and let you explore the dynamics of the system.
  • Chaim Gingold — one of the designers behind the game Spore discusses that Spore was originally inspired by the Drake equation.
  • He’s always harbored an interested in the concepts of simulation as a way to visualize complex adaptive systems.
  • Decided to carry on with his work on simulations, and eventually created a educational simulator called Earth Primer as a side project. This simulation models many of the natural processes on earth — including geology, plate tectonics, the water cycle and erosion.
  • Shawn Douglas — Biophysicist and Computer Scientist discusses how in his practice, he has employed simulation software (CADnano) methods into his research, as it helps visualize and communicate core concepts to his research teams and students.
  • Using CADnano Shawn explains that he can simulate the Gel electrophoresis process which allows the study of dissected strains of DNA. Simulation allows his research to undertake experiments that would ordinarily take much longer using more conventional research methods.
  • This simulation process also allows users to scaffold learning, eliminate the pain of absorbing abstract concepts, bootstrap mental models and internalize the relationships within the system.

Autonomous Ecosystems

Jen Dunne, Kate Greene, Mark Nelson, & David Stout

HI-SEAS habitat on Flickr — Photo Credit: University of Hawaii

This panel discussion surfaced many discussions about the challenges around creating self-sustaining ecosystems. Two of the panelists Kate Greene and Mark Nelson have been inhabitants of the HI-SEAS and Biosphere 2 simulation projects respectively. This panel covered a lot of ground, so I’ve tried to group most of my notes around key questions.

Earth is the only autonomous ecosystem we know of.

What are the biggest challenges in running and creating and autonomous ecosystem?

  • You can count on life to do what life does. Technology will break down. Very important need for technological and ecological redundancy.
  • Better understanding and account of the interplay between technological, ecological and social systems.
  • There are many challenges in building a fully hermetic autonomous systems. A good example of this was Biosphere 2 which eventually failed due to the presence of too much organic matter in the biosphere’s soil, which fueled oxygen-eating bacteria, which was not predicted by the designers.
  • It’s difficult to predict unintended consequences. Everything is connected.
  • Once you have interconnected networks of organisms and technology, there’s a dramatic increase in chances of unintended consequences.

What assumptions do you include to model complex ecosystems?

  • Feeding interactions in a food web.
  • A balance between specialists and generalists (you need a mix)
  • Food systems in biosphere 2:
  • Low calorie high nutrient diet
  • Cooks had to work extra to make new meals
  • Putting some aside consciously to have feast days.
  • Bad meals were terrible for team morale
  • Problems with invasive species (morning glories were rampant in the biosphere)
  • Only takes a few to succeed to completely change the dynamic of the system

Do we have the capability to go to another planet?

  • Things will start with almost no food production.
  • Humans come “bundled” with a biosphere.
  • Space system designers will create gardens, but will most likely not depend on them for crops to feed people.
  • Ecosystems are not tidy and often difficult to ‘debug’
  • People intervening adds complexity.
  • Mars dust contains salts which would make the whole planet a superfund site.
  • Space life support will be limited with robust backup. Cost of resupplying to space will be very expensive.

Planetary Policy & Regulation

Linda Sheehan & Jeff Ubois

A billboard promoting Henry George’s single tax idea
We live in a dynamic complex adaptive system where our actions have reverberations over the whole system.

What does an economic system look like that reflects the complex legal and economic world we live in? 3 options:

  • First: Use existing neoclassic models we used for the industrial system.
  • Second: Second is a commons based governance approach like we did with Antarctica (or open IP).
  • Third: Nature needs rights. All things have a right to have a chance. Extension of rights. Rights are inherent because we exist. Giving rights modulates behavior.
  • Space treaties prohibit national appropriation of space entities
  • A concept whereby we treat space like the commons might have it’s own pitfalls. The commons suffers because it ignores complexity (e.g water usage in California. Water is used for agriculture because it makes money)
  • Anticipatory philanthropy has worked well in the past (e.g. Rockefeller and the Tetons National Park)
  • Think universally act locally. Think about what you can do in your locality to make the community better.
  • Henry George’s Progress and Poverty cited as an interesting alternative to current market systems, whereby the key difference is maximizing ecological well being rather than making profit.

The Search for Life in the Universe

Pete Worden

Pete Worden, was Director of NASA’s Ames Research Center at Moffett Field, California, until his retirement on March 31, 2015. His keynote centered around 3 key questions;

  • Is there any life elsewhere?
  • Is there intelligent life elsewhere?
  • Can we travel between stars?

In order to answer the first two questions, Pete discussed that a lot of capital has gone into many initiatives in which he has personally be involved in including the Breakthrough prize series.

  • The Breakthrough prize for listening has invested a lot of capital into purchasing radio telescope time alongside a consortium of scientific agencies across the world. This funding has gone towards radio telescopes all around the world, including the FAST telescope in Guizhou, China.
  • The Breakthrough prize Starshot: This initiative aims to send a spacecraft to our nearest galaxy - Proxima Centauri.
  • Pete discusses conventional chemical reaction based rockets. To get conventional rockets to get us to 20% of the speed of light (the lowest feasible velocity for interstellar travel) we’d need as much fuel as is contained in our current galaxy.
  • We’re about 14 orders of magnitude away (at current development levels for anti-matter based propulsion systems) from having enough dark matter to create a propulsion system.
A nanocraft prototype
  • Nanocrafts seem to be our best bet yet. With a size roughly the same as an Apple Watch, propulsion becomes much more feasible.
  • Current research shows that using a combination of solar sails and earth-based high energy lasers trained on a small spacecraft might yield the best result of propelling a spacecraft to 20% the speed of light.
  • Two trends which make space sails feasible: microelectronics and photonics.
  • There is a similar dynamic to Moore’s law playing out in the world of laser development. Lasers fall in cost and increase in power every decade.

Conclusion

Overall, I thoroughly enjoyed the Interplanetary Festival. It had a well balanced mix of presentations, panels as well as artistic performances. The panel discussions were largely inspirational, and generally left me curious to learn more about the topics being discussed.

The topic that most resonated with me was the approach of using simulations as a method to understand the dynamics of complex adaptive systems. While simulations by definition cannot be a fully accurate model of any given system, they provide an interesting opportunity in giving designers and developers an early preview of the type and magnitude of challenges they might face when building a full scale system. We see this type of approach already being conducted to train deep learning systems, and I suspect we will see many more simulation-based approaches as we continue to build the toolset for machine learning based systems.

Most importantly, the festival felt incredibly inclusive to all, regardless of background or discipline— a great environment for the fostering of creative ideas. I look forward to seeing what David Krakauer and the rest of the Santa Fe Institute will do next in this series of events.