Roundup: March '22 Edition
A new industrial revolution; new science & research models; startup spotlights; space updates; and Planet Labs
Greetings FutureBlind readers!
In this roundup edition:
⚡️ Let’s jumpstart the new industrial revolution
🧪 The new wave of science and research models
🔦 Startup spotlights: Terraform, Hypar
🚀 A few space updates
🛰 Investment: Planet Labs
⚡️ Let’s jumpstart the new industrial revolution
There is as much headroom in physics and engineering for energy as there is in computation; what is stopping us is not lack of technology but lack of will and good sense. — J. Storrs Hall
There have been three industrial revolutions. The first two spanned from the late 1700s to the early 1900s and essentially created the technological world we know today. Energy, transportation, housing, and most “core” infrastructure is pretty similar now as it was at the end of this period — especially if you extend it into the 1970s. The third revolution, the “Digital Revolution”, started around this time and as anyone reading this knows has made computing and communication ubiquitous.
There were bad things that came from these revolutions: pollution, environmental destruction, war, child labor, etc. But the good overwhelmed the bad, leading to GDP per capita (”resources per person”) doing this, which we can use as a proxy for progress in a host of other areas like longer/healthier lifespan, lower child mortality, less violence, lower poverty, and more.
Wikipedia describes the potential Fourth Industrial Revolution as "...the joining of technologies like artificial intelligence, gene editing, to advanced robotics that blur the lines between the physical, digital, and biological worlds.”
These things are great, but we need more. Much more.
As just one example, it’s become abundantly clear over the past few weeks the importance of energy independence. But why don’t we already have it?
The cost of PV cells has collapsed over the past few decades. We also know it’s possible to build nuclear reactors far safer and more productive than any in the past. There should be solar panels on every home, geothermal wells in every town, and multiple nuclear fission (possibly fusion?) reactors in every state. A setup like this would lead to redundant energy at every scale, not reliant on geopolitics or over-centralization.
We should want to consume more energy, not less. (And unlike the second industrial revolution, it can be clean energy with minimal externalities.)
What else could a new industrial revolution bring? Just imagine what you’d see in a typical sci-fi movie:
Space parks/hotels/colonies, limb regeneration, flying cars, supersonic jets, same-day shipping to anywhere on Earth, self-replicating nanobots, new animal species, plants everywhere, infrastructure made out of GM trees, universal vaccines for all viruses, mobile robotic surgeons that can save lives on-location, convoys of self-driving cars, batteries with 50x current power, etc. etc.
To build these things — or even to see if they’re possible — a lot needs to change. Here’s just a few I’ve been thinking about:
Create a pro-progress culture. Pro-progress means anti-stasis. We’ve come a long way, and things are pretty good now. But they could be better. Far more people should be optimistic about the future and what they can do now to make it better.
Find more ways to celebrate and fund scientists and inventors like we do founders, celebrities, executives and sports stars. More crazy ideas should be funded, and even if they don’t succeed, the culture should be accepting of it.
Take more risks as a society. Incremental progress is great but even over long periods it can lead to a local optimum. To get to a higher peak, we need more exploration, experimentation, and invention. With this comes risk. We should do whatever we can to be conscious of and mitigate these risks, but in the end if the precautionary principle is applied to everything, we’ll be stuck in stasis until a global catastrophe forces our hand.
Allocate more resources to efforts that have high expected return to life on Earth. Nuclear fusion, for example, may have only a small probability of succeeding in the next 10 years. But if it does, it could bring enormous benefits to the world (to humans, animals, plants, you name it). The probability-weighted return to life on Earth is thus very large, and yet minimal resources are being devoted to it. The industrialization of space is another example. Concerned about depleting Earth’s resources or peak “X”? You wouldn’t be if we could mine asteroids and move potentially harmful processes off-planet.
If you agree with any of the above or are interested in similar ideas, here’s a few good resources I’ve enjoyed recently:
Where’s My Flying Car? book by J. Storrs Hall. An amazing book that I would highly recommend to everyone. Hall uses the premise in the title to discuss why we’ve fell behind in everything from flight, nuclear, nanotechnology, batteries, robotics and cities. Remember: it is a possibility.
Faster, Please! newsletter by James Pethokoukis. James writes a really interesting newsletter, with twice a week editions on all things growth. He recently moved to a paid model (which I subscribe to) but the archive is great, and he’s still doing once a week summaries.
A Simple Plan to Solve All of America’s Problems, article by Derek Thompson.
What would a thriving progress movement look like? post by Jason Crawford.
Infinite Progress, Techno-Optimism and the Near Future, post by Infiniti Ventures.
🧪 The new wave of science and research models
There has been an increasing amount of experimentation in the philanthropic and scientific funding space over the past few years. This is good news — as I mentioned in my last post, we need better ways to fund crazy ideas.
Here’s a sampling of some of the recent efforts:
The Astera Institute — Pursuing new tech areas through multiple models including FROs, PARPA (based on the DARPA model).
Fast Grants — An effort by Tyler Cowen, Patrick Collison and others to quickly disburse grant money to COVID-related ideas. Funded by many wealthy donors and philanthropies. Impetus Grants for longevity research was recently launched and inspired by Fast Grants.
New Science — Funding life science labs outside of academia. Partly funded by Vitalik Buterin.
Arcadia Science — Bio research institute.
Arc Institute — Funds individuals similar to HHMI, in partnership with Stanford, Berkeley, and UCSF. Founded by Fast Grants “alumni” Silvana Konermann, Patrick Hsu, and Patrick Collison.
Convergent Research — Uses focused research organizations (FROs) to solve specific scientific or technological problems. Funded by Eric Schmidt’s philanthropy.
Altos Labs — Biotech lab, another “academia outside of academia” model.
VitaDAO — A DAO-based longevity funding org where holders get a cut of IP proceeds.
Actuate — Also using the DARPA approach to fund and implement R&D.
FTX Future Fund — A non-profit fund from the FTX crypto exchange, aiming to allocate at least $100M this year to a wide variety of long-term focused areas.
In “Illegible Medicis and Hunting for Outliers” Rohit observes that:
There are two common themes here. That’s speed and autonomy. They mostly act under the assumption (the correct assumption it would seem from a betting lens) that they can identify talent, not bug them excessively, and leave them to do their thing. Instead of imposing rules and strictures and guidelines, they focus on letting the innate megalomania do the work of focusing their research.
The academic and government driven funding models have come up against their limits in recent years (decades?). These experiments provide new methods to allocate capital to research, development, and implementation of efforts that for whatever reason aren’t amenable to the startup funding ecosystem.
Prior to World War II, support from non-government or educational institutions was the norm. Patrons like Alfred Loomis ran a lab at Tuxedo Park, hosting scientists and engineers from around the world that was integral in the creation of radar. Funding was provided by philanthropies from the likes of Carnegie and Rockefeller. Or private R&D from Edison, Bell Labs or Cold Springs Harbor Lab.
These past models are still doing well of course — HHMI, the Gates Foundation, Google X, etc. — but much more is needed to expand experimentation. The government can continue to play a valuable role, particularly as a buyer of first resort.
I’m super excited to see what comes from these orgs. A few like Fast Grants have already had some impact.
For more on the topic, see:
The Overedge Catalog — an updated list of atypical research organizations.
Understanding science funding in tech, 2011-2021, by Nadia Eghbal.
🔦 Startup spotlights: Terraform, Hypar
Terraform Industries is a carbon capture startup that wants to turn atmospheric carbon into natural gas. So basically “reforming” the emissions of the past into usable fuel. This process requires more energy of course, and they’re banking on the falling cost of solar cells to get it. In the end the Terraform fuel plants will combine carbon from air + sunlight + water ⇒ natural gas. In other words, the plant is like a plant. They recently raised a $5M seed round and are hiring. If you’re interested in more details check out cofounder Casey Handmer’s whitepaper on Terraform. (Prometheus Fuels is another interesting startup with a similar goal.)
Hypar makes software that automates the building design process. It supplements the existing design software from AutoDesk. Rather than have to design a building wall by wall, you can specify the general parameters and constraints and Hypar creates the final design. So “design a 3,000 square foot, 2 story farmhouse home with 4 bedrooms, 3 bathrooms” translates into exportable blueprints/CAD files with all the necessary infrastructure, and you can further refine from there. Most architects are implicitly using design patterns anyway — this just turns those patterns into code.
I strongly believe that generative design is just beginning and that this will be a huge trend going forward. As technology progresses, complexity inevitably increases. And with complexity, humans will still be the higher level creative force but software is needed for the heavy lifting. Check out cofounder Anthony Hauck in this interview with Brian Potter for more.
🚀 A few space updates
Starship still making progress and waiting for FAA clearance. The FAA has now delayed their environmental determination a few times now, with the current ETA end of this month. Regulatory delays are frustrating, but there may be legitimate concerns, and it seems like the delay may not be actually on the critical path given SpaceX’s revamp of the Raptor engine. Worst case scenario is they move orbital launches to Florida. In Musk’s Starship update last month, he showed off the new v2 Raptor engine, targeting an eventual production rate of 1 engine per day.
🎮 Hypothetical 3D walkthrough of Starship — explore a 3D model of a crewed Starship. This is “fan fiction” but a really awesome view into the potential of Starship once it’s human rated.
📹 Beautiful animation of Starship’s potential journey to Mars
Jared Isaacman is funding a new civilian space program. The Polaris Program plans to launch at least 3 missions, each of them testing new capabilities related to human spaceflight. Polaris Dawn will launch later this year on a Falcon 9 / Crew Dragon, and plans to test a new SpaceX-developed EVA suit for a spacewalk along with other research related tasks.
25 theses on space. Joe Carroll lists 25 theses related to Starship and the future of space. I’m skeptical of some of these, but here are my favorite though-provoking points:
Once launch prices drop ~4-fold, most customers will focus on launch vibration & reliability.
It is harder to live in the best place off Earth than the worst place on Earth (except volcanoes).
Growing food in LEO can get us closer to living off Earth than putting people on Mars can.
After decades of refining 0g health countermeasures, human health still goes downhill in 0g.
We know nothing about sustained human health between 0g & 1g (Apollo stayed 1-3 days).
There are a bunch of new Moon rovers being developed. See: The bevy of rovers heading for the Moon.
James Webb launch and journey was 100% (200%?) successful — All phases of the JWST deployment have completed successfully so far, and it has reached its destination at the L2 Lagrange point between the Sun and Earth. The precision of the Ariane 5 launch was so efficient in fact that it increased estimated lifespan of the telescope from 10 to 20 years! On Feb. 3, it detected its first photons and is currently still calibrating before being able to send actual imagery back sometime this summer. The best way to keep up with updates is to follow @NASAWebb on Twitter.
🛰 Investment: Planet Labs
Even though FutureBlind started as an investment blog, I almost never write about investments anymore. I’m still an investor of course but writing about it just isn’t as interesting to me anymore. ¯\(ツ)/¯
This is an exception. I first purchased shares of Planet ($PL) in the private markets almost 4 years ago. They went public via SPAC this past December, and shares promptly got cut in half along with many other former SPACs and high growth companies. I added more to my position between $5 and $6 as it fell (currently around a 5% allocation, a mid-size position for the fund I manage).
Planet Labs is an Earth observation (EO) company. It creates tiny imaging satellites, pays to launch them into space, collects and analyzes imagery from them, and sells that data to customers. Their largest satellite constellation (called "Doves") are built from mostly off-the-shelf components, making them much cheaper than traditional satellites. Planet is currently the only company that images the entire globe every day.
Here are the main points:
Competitive advantage comes from both the capital barriers to entry and their data flywheel. The more past imagery they have, the easier it is to build detection models for future imagery. And because it currently takes a lot of money and time to launch that many satellites, Planet is ahead of any competition by many years. (It won’t be until Starship is launching regularly that this gap can be easily narrowed.)
Market currently isn’t large but it has the potential to be huge. The biggest applications are in agriculture, defense/intelligence, climate, energy, finance, and mapping (Google is an investor and partner for map data).
Opportunity to move up the stack. Most of their prior business was in selling access to raw imagery data. This can be really useful for some (think tech cos and intelligence agencies) but a hard sell for most. Planet is now selling services like the ability to analyze and detect specific things, which has a much bigger and more profitable market. Like “give me a chart of how many cars are in these parking lots in the last year” or “what is the crop yield on these fields” or “how many Russian tanks got destroyed in this convoy”.
I see Planet as a “utility for data”, similar to Bloomberg, FactSet or Compustat. They have an installed hardware base that needs some maintenance — new satellites to replenish + ground communication costs — but most of the future opportunity lies in software.
The bear case it seems is primarily that the market won’t be able to profitably expand much. I don’t have any counter evidence but it just feels like this data is too valuable for that to happen — customers just don’t know how to use it yet. The biggest risk I see for Planet is competitors giving access to different light spectrums. This can allow customers to do things like see through clouds. But if Planet can get to this space soon it should help mitigate the risk. They just deployed 48 “SuperDoves” that have access to more spectral bands.
So there are risks. But this to me is a clear case of a lollapalooza of positive factors. Good managers, good product, huge market size potential, recently public and more incentivized for growth/profits, and optionality of moving up stack. As always, do your own research — but I think Planet is a good buy here.
Thanks for reading! If you found the above post thought-provoking and want more random musings, subscribe below. I post irregularly and usually only a handful of times a year so subscribing is the best way to catch up.