Episode 69 — September 29th, 2022 — Available at read.fluxcollective.org/p/69
Contributors to this issue: Dart Lindsley, Dimitri Glazkov, Erika Rice Scherpelz, Neel Mehta, Boris Smus
Additional insights from: Ade Oshineye, Gordon Brander, a.r. Routh, Stefano Mazzocchi, Ben Mathes, Justin Quimby, Alex Komoroske, Robinson Eaton, Spencer Pitman, Julka Almquist, Scott Schaffter, Lisie Lillianfeld, Samuel Arbesman
We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.
“Now, here, you see, it takes all the running you can do, to keep in the same place.”
― The Red Queen, Alice in Wonderland
👶🧬 The unvarying infrastructure of variety
Multicellular organisms go to extraordinary lengths to maintain genetic fidelity between generations. For instance, most human cells replicate over 18,000 times between birth and age fifty, but generation-defining germ cell lines follow an obsessively short path: the sperm-to-sperm path averages between 40 and 1000 replications, while the egg-to-egg path is even shorter at only 23 replications. This is no accident. Each cell division is an opportunity for mutations, so faithful reproduction between generations favors the fewest germline cell divisions.
Each cell displays a similar obsession with reducing variation. DNA mismatch repair mechanisms constantly inspect DNA for anomalies, snip out any evidence of error, and replace it with (hopefully) accurate code.
Evolution has selected mightily for organisms able to reproduce faithful copies, both cell to cell and generation to generation. One would be forgiven for assuming that suppression of variation is the entire story. It’s absolutely not. In fact, some of the genes that so obsessively conserve code appear to do this to enable generating genetic variation.
Sexual reproduction tops the list: organisms conserve, conserve, conserve, and then mix everything up to produce a combinatorial explosion of variation. To defend against threats emerging from a dynamic population of pathogens, B-cells mix and match DNA segments, including a step called “hyper-mutation,” to create a combinatorially enormous range of antibodies.
In a complex environment able to produce an endless stream of unpredictable threats and opportunities, in an environment full of other organisms deploying precisely the same strategy, preserving the ability to produce combinatorial hyper-diversification seems to be a winning strategy.
Can we apply this insight to non-biological infrastructure for diversification? We can find encouraging examples of variation-enhancing infrastructure built into our economy, businesses practices, and nations — even if that was not their original intent.
Antitrust laws create a competitive environment in which companies must innovate new products and services in order to remain relevant.
In the United States, the Constitution enshrines freedom of speech, which enables a diversification of ideas.
Some platforms establish micro-economies in which the duration of intellectual property rights is limited so that code sharing accelerates application innovation.
University sabbatical programs enable researchers to work with colleagues at other institutions, driving cross-pollination (an apt metaphor) of ideas.
In Talent Wants to be Free, Orly Lobel makes a compelling argument that California’s prohibition against non-compete clauses in employment contracts enabled the success of Silicon Valley by encouraging cross-pollination of knowledge when people move between companies.
The list goes on — and a few lessons emerge. Hyper-diversification is not simple chaos: it arises from a stable infrastructure encoded by genes, laws, contracts, policies, and cultural norms. It is at its best when it enables combinatorial mixing. It comes at some cost, but this cost is worth it.
🛣️🚩 Signposts
Clues that point to where our changing world might lead us.
🚏🗜 Stable Diffusion compressed images better than JPEG in one test
The AI image synthesis tool Stable Diffusion is normally used to generate images given textual prompts. But one engineer realized that one of Stable Diffusion’s neural networks includes an image compression algorithm. In an experiment, he extracted this algorithm and used it to compress some images. He found that this compressor was able to shrink images down to smaller file sizes than JPEG or WebP while introducing fewer visual artifacts — though the algorithm doesn’t do too well with faces or text, and it sometimes “hallucinates” features that weren’t in the original images.
🚏🛫 One-way air fares out of Russia skyrocketed after Putin started his mobilization
The price of flights out of Russia soared as soon as Russian President Vladimir Putin ordered a “partial mobilization” of fighting-age men and the “calling up” of reservists. The price of one-way fares to Turkey more than tripled in a week, reaching over $1,100; the cheapest flights to Dubai exceeded $5,000, nearly five times the average monthly wage. Some itineraries, like Moscow to Tbilisi or Yerevan, were completely sold out.
🚏🛢 One startup can generate cheap hydrogen gas from abandoned oil rigs
Currently, the main “green” way of generating hydrogen gas is electrolyzing water using electricity generated by renewable sources like wind or solar. But this gets expensive: the price of green hydrogen generated this way briefly spiked to over $16 per kilogram. One startup had a different idea and ran a field test that injected lab-grown microbes into “depleted [underground] oil reservoirs”; these microbes were able to convert the leftover hydrocarbons into hydrogen gas at a cost of just $1 per kilogram.
🚏🍗 The FDA made a “NyQuil Chicken” challenge go viral after warning against it
One obscure 4chan in-joke involves trying to cook raw chicken in the blue-green liquid cold medicine NyQuil. (It looks nasty and is just as toxic as it sounds.) For some reason, the FDA this month published a press release warning about the dangers of the “NyQuil Chicken” challenge. This had the unintended consequence of making the meme go viral on TikTok, where searches for the challenge grew 1,400-fold in just a week.
🚏🛒 Walmart is launching two metaverses to reach younger shoppers
Walmart is launching a pair of virtual experiences on the Roblox gaming platform: Walmart Land and Walmart’s Universe of Play. The company’s stated goal is to “experiment with new ways to reach shoppers” in a world where “Gen Z” increasingly lives on TikTok and gaming platforms. These new metaverses will feature a blimp that drops virtual toys, a music festival, and virtual merchandise (nicknamed “verch”). Walmart won’t directly monetize these games, but it hopes that Roblox gamers will play with digital versions of toys and then want to buy the real things in stores.
🚏🇮🇳 Apple is moving some iPhone 14 production to India
Apple announced that some of its new flagship iPhone 14s are being assembled in a factory in southern India; this marks the first time that Apple is making the newest iPhone model in India. Analysts believe that Apple will move 5% of its global iPhone 14 production to India by the end of this year, a number that might rise to 25% by 2025. Potential reasons for the shift include China’s draconian COVID lockdown policies and Apple’s desire to make inroads into the Indian market.
🚏💱 The US’s derivatives trading regulator is suing a DAO
In 2021, the founders of a crypto lending platform called bZx transferred control of the protocol to a newly-founded Decentralized Autonomous Organization called bZx DAO, bragging that this made the platform “enforcement-proof.” The US’s CFTC didn’t think so, though. It’s now filing a lawsuit against the DAO (now known as Ooki DAO) for illegally offering margin trading and failing to register with the agency. Said the lawsuit, “DAOs are not immune from enforcement and may not violate the law with impunity.”
📖⏳ Worth your time
Some especially insightful pieces we’ve read, watched, and listened to recently.
June Huh, High School Dropout, Wins the Fields Medal (Quanta) — A portrait of a poet-turned-mathematician whose character is charming, placid, and in many ways the polar opposite of a prototypically “successful” person who moves fast and gets things done. There are so many ways to be.
New Business Boom and Bust: How Capitalism Experiments (Michael Mauboussin) — Finds surprising parallels between the growth of new industries and the development of children’s brains: many avenues are explored early on, the less-used ones are pruned, and the frequently-used ones are reinforced. While it may seem wasteful on the surface, it’s a flexible strategy that allows quick adaptation to the environment.
What I Miss About Working at Stripe (Brie Wolfson) — A nostalgic piece about a time when work had a strong gravitational pull on the author. She does a good job of capturing the feeling that a kind and balanced working environment may be at odds with the latent desire to do the best work of one's lifetime.
Why Subways in the US Are Set Up to Fail (RMTransit) — Uses concepts from systems thinking to argue that American transit systems are “frail.” For instance, economies of scale (like a metro system using the same train car model for all its trains) lead to brittleness (like when all those trains need to go offline at once after a flaw is discovered). What’s more, radial networks like Boston’s T can’t handle disruptions as well as mesh networks like New York’s subway, which can route around failures.
Estonia: Warning the World About Russia (Newlines Magazine) — Examines why the small Baltic country is so fiercely helping to defend Ukraine, donating 40% of its annual military budget and more than 0.8% of its GDP — more per capita than any other country. Argues that Estonians don’t need to be reminded of what Russia under Stalin did to their ancestors in 1941 and 1949. Unlike much of the rest of the world, Estonian PM Kaja Kallas believed that Putin would invade Ukraine on February 24th.
Metacrap (Cory Doctorow) — Argues that trying to get everyone to agree on a single metadata schema is a high-modernist fool’s errand. People are bad at explicitly labeling data, and everyone brings their own preferred idea hierarchy to the table. Instead, we should use implicit metadata, like how Google exploited the Web’s network of hyperlinks.
Simulation Storytelling: Impact and Value Dynamics (Emergent Outcomes) — Introduces two useful variables to consider when evaluating the benefits of democratic governance: the coordination cost discount (how much energy is taken up by the mechanics of voting and debate) and the coordination opportunity multiplier (how much adjacent opportunity is unlocked by effective group coordination).
🕵️♀️📆 Lens of the week
Introducing new ways to see the world and new tools to add to your mental arsenal.
This week’s lens: vetocracy.
Imagine you’re with a group of friends trying to decide where to go to dinner. People suggest options, give feedback, reject some, seem to assent to others. Yet, somehow, the group can’t reach a decision — until someone takes charge and says, “Alright, we’re going to <place>!” Up until that point, you’re experiencing a vetocracy.
The term vetocracy was coined by Francis Fukuyama to describe governments where no one can acquire enough power to move forward. Essentially, too many people have the power to say “no” and no one has the power to say yes. If we soften this a little, to say that no one has acquired enough power rather than that no one can, this pattern applies frequently in group decision making… hence the focus on identifying the decider.
Another variant of vetocracy is when a decision is litigated again and again. People find themselves repeatedly called upon to justify their decision, sometimes restarting the process from scratch. If those implementing the decision don’t have the authority to ignore these interruptions, they’ve been set up for constant energy-sapping distractions. It may be possible to reduce the cost with robust documentation, but only when the interruptions are primarily requests to understand the reasoning. However, where relitigation is a manifestation of underlying disagreement, it can be hard to overcome this repeated soft veto.
Whenever you find yourself in a vetocracy situation, look for those who have the power to make a decision and who have the power to enforce the decision. These may not be the same person — an executive may make a decision, but it takes the day-to-day efforts of someone closely involved to enforce it. Identifying these people may not be sufficient to make progress, but it’s a necessary first step.
© 2022 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.
On Vetocracies, the CIA's (formerly OSS) Sabotage Field Manual [0] seems to describe a very similar method for sabotaging an organization's effectiveness:
- "Refer back to matters decided upon at the last meeting and attempt to re-open the question of the advisability of that decision,"
- "When possible, refer all matters to committees, for "further study and consideration." Attempt to make the committees as large as possible - never less than five."
[0] https://www.hsdl.org/?abstract&did=750070