Proof of Carbon

Bitcoin inadvertently created a more direct link between exchange currency and carbon, through the CPU- and hence energy-intensive process of proof-of-work mining. Can we make a better link?

Edward Dodge has proposed using the blockchain as a distributed ledger of carbon account, with mining based on a ton of sequestered CO2. Let’s follow that suggestion but make each coin represent a kilogram of carbon.

Altcoin CarbonCoin replaces distributed mining of difficult to calculate numbers with mining by an environmental trust that uses six orders of magnitude less energy and puts profits into carbon mitigation. It relies on trusting that single third party organization, though. We want to have more decentralized platform management, as in many cryptocoins, while establishing this same carbon link.

There are a couple of other projects like Dovu or Treecoin which focus on particular types of carbon sequestration, but this sketch takes a different tack.

A Design Sketch

I wrote this design sketch a few years ago and then put it in a box called THINGS TO THINK ABOUT – URGENT. I’m not launching a billion dollar crypto play using it right now, so I figured I may as well share it here. I think we should have more public design sketches in software.

Basic Protocol and Squatting

We can push these initiating ideas a bit harder. As noted, forests and oceans are major carbon sinks. Prospecting rights could be claimed for mining, with proof of work replaced with an empirical proof of carbon. For each carbon sequestering device or location, you can associate a different allowed carbon coin mining rate. A corresponding proof of carbon could require 1) making the claim first 2) providing time and location specific weather information.

For secondary tropical rainforest, Bonner et al estimate 7.5-15 tons per hectare per year (via). That’s a pretty wide band, but let’s run with the lower figure for now and make that a claim worth 7500 coins per year. That’s 20.53 coins a day, which we’re going to round down to 20 for whole coin mining. We’ll also halve it, to 10, for reasons explained later.

For other types of carbon sinks, different rates would apply, but the protocol is the same.

Once a day, a miner can claim the right to mine the claim for that day. It has to provide

  1. The location in latitude and longitude.
  2. Proof the location is still a tropical rainforest through a public satellite photograph. Initially this could be from Google Earth.
  3. The temperature and humidity at 10 am that day in that timezone, at the nearest location providing a trusted source for that information. Initially these would be bureaus of meteorology and similar institutional sources. 

Otherwise the process is the same as claiming a bitcoin – it is advertised to the network and validated by other miners.

A miner has to obtain a mining license. This can be bought for 1 coin from any miner that has minted a coin in the last month within 5km of the desired location. If there are no miners for the last month in that area, it is free, and can be self-certified. This is to discourage mining spam. Given the computational costs are much lower than bitcoin mining, there would be a possibility to create a miner for every hectare on earth, and spam a coin attempt at every possible temperature and humidity for a given day. The license mitigates this, and cost might vary over time to manage it.

A miner using the basic squatter protocol doesn’t need to demonstrate any legal connection to the land or ocean involved. It’s a mathematical mapping only, as with the large numbers in bitcoin. The carbon coin mining claim is more important to the network than the legal title to the land, because double-claiming the carbon sink would make the carbon accounting invalid. For natural assets, the computer where the mining software runs need not be in the same location as the trees, though a maturing platform demanding more precision might call for devices on the ground, linking the Wood Wide Web to the internet and the blockchain. These specifically designed sensors can also have more openly validatable code, connecting as part of the Internet of Things (IOT).

Proof of carbon definitions for a type of sequestration can be captured as public software contracts, using Ethereum or a similar platform. They would need to be more dynamic than the bitcoin protocol because valid earth data sources would vary over time. 

The local weather data requirement gives people local to the forest or ocean concerned a small co-location advantage similar to that of high frequency trading systems to stock exchanges. The world’s greatest carbon sinks are not found in rich world finance capitals: this would give a small home town edge to those local to say the Amazon or Daintree rainforests, and encourage more diverse locations and owners for miners.

Legal Title Protocol

The basic squatter protocol described above allows fast-moving mining organizations to get going with very low upfront costs and similar bootstrap dynamics to bitcoin. 

There are advantages in linking legal title to the land to mining rights in the network, though. Miners have a financial stake in the carbon sequestering income of the land they claim – if trees are cleared, proof of carbon is lost. Owners of land have much more direct control over what’s growing there. Mining rights would even be an incentive to reforest cleared land.

Legal systems are complicated systems varying widely by location. There are problems of language and legal expertise. Legal title is often hard to validate in software, and even where such interfaces exist, title searches have significant charges, which could easily multiply with independent validation by network participants. Imposing these as barriers to entry for all mining would make participation uneconomic until the coin value was relatively high.

The solution in this protocol is to treat the two types of miners as complementary and have both. 

With both proof of title and proof of carbon, a miner can mine a second coin for each corresponding kilogram of CO2 sequestered by the underlying hectare of land. This gives no squatting protocol rights. The first coin is still determined by speed.

Title rights would often be shared, and any proof that does not rely on a central trusted source seems implicitly tied to proof of identity by an authority, and impossible to be anonymous, if published on a publicly verifiable blockchain, or through intermediaries such as banks or governments. Techniques for doing this in general, and the codification of proof regimes for each jurisdiction, will grow over time, and aren’t detailed here.

Deflation, Re-emission and Redistribution

Atmospheric carbon isn’t sequestered forever. Trees are cut down or eventually die. Ocean sinks and old coal mines leak. Tundra melts in the summer.

The simplest way to reflect this in a carbon coin is to make the coins expire. Those mined from a given type of carbon sink have an expiry date based on the ecological infrastructure that minted it. For secondary tropical rainforest, we use the example mean lifespan of 60 years.

The second way a coin can expire is if the sequestration source that backed it is destroyed, eg, the corresponding hectare of forest is cut down. This intensifies the economic incentive to preserve carbon sinks, as not just future revenue but existing wealth can be destroyed.

When this happens, it has the monetary effect of deflation. A fixed amount of commodity-like currency corresponding to the actual carbon stock is desirable in this case, as it would make market actors responsive to the actual carbon limits of the ecological layer of the economy.

We suggest the market would respond to expiry dates in a similar way it responds to expiry of options contracts or dividend rights, by value declining to zero near the end of their lifespan. Since it’s not desirable to have cash expire in your wallet, or to lose significant chunks of wealth because coins happened to come from the same source hectare, it would also create a demand for portfolios of coins balanced across many sequestration sources. Algorithmic balancing wallets seem a reasonable solution to this problem. This would also keep coins in greater circulation and discourage hoarding, which is more of a feature than a bug.

Linking Emissions

At this point you already have a commodity-based exchange currency platform equivalent to Bitcoin, including distributed mining. All of the usual financial and software infrastructures can be built on top of it. The main missing feature is money supply management available in central banking. That is deliberately designed out of Bitcoin too, out of libertarian grumpiness with the state. For carbon cryptocurrency it would be omitted for a more sincere representation of the foundational geophysics the whole planetary stack runs on.

That the coin is based on carbon allows extensions which reinforce carbon homeostasis. Carbon-emitting endpoints such as power stations, petrol service stations or factories could have corresponding IOT devices requiring spending carbon-backed coins to operate, basically acting as IOT smart meters connecting to a carbon exchange. Governing such a mechanism, and avoiding tampering to evade it, would likely involve both taxation enforcement and digital rights management, whether implemented by state or corporation. Because carbon emission would result in a transaction on a public blockchain, it would also be publicly auditable, depending on how much detail the emitting device is configured, or mandated, to disclose. This latter consumption piece isn’t necessary for the currency to work, but it does look like a good feature.

Permissive Licenses Are Gifts

Harrison Ainsworth, who is always interesting on software matters, has a recent note arguing permissive software licenses are unethical. I disagree. Such software is a non-toxic gift, like free cake.

All open source licenses allow access to source code and some kind of copying and modification. Permissive licenses, such as the MIT license, have no restrictions. You may copy, alter and share the source code and anything built with it as you please, including making other things that are not shared back to the community. This is in contrast to “copyleft” licenses like the GNU Public License (GPL). If GPL code is changed, is built into other things, that code must also be released.

There might be times when a gift is unethical. Giving someone an animal they can’t look after is unfair to the animal. Giving something leaking toxic chemicals could hurt people. Some gifts create an obligation of work.

Open source software doesn’t seem to be in any of these categories. People may adapt and contribute back if they wish – or choose not to use it at all. With a permissive license, people not in a position to contribute back today can still use the tool, and perhaps be able to contribute back tomorrow. Many developers working for conservative corporations or governments were and are in that position.

The exception would be open source code embedding adware and trojans, like Clipgrab (which claims to be GPL but took down its open repository years ago), or malicious NPM packages, or cryptominer embeds. These are all the equivalent of gifting your sister a nice bucket of plutonium slurry for her birthday. But this is not most open source software, and it’s not what Ainsworth has in mind.

Ainsworth says the permissive license boils down to “I should share, but others should not.” But it’s really “I should share, but those who cannot need not”. I’m sure this basic argument is already known to Ainsworth, and so his primary point is about consequences: permissive licenses create an ecosystem of freeloading corporations. But the alternative, if everyone were a GPL purist, is likely not a world of free GPL software; it’s a world of worse software, with less good tools available, and corporations and other rich bad actors continuing to skive anyway. The corporations that contribute back to open source are mostly more technically sophisticated ones; the ones who understand software development dynamics, though a Google or Facebook may mix the two together and often skirt the freeloading line.

Python is an interesting test case for permissive licenses. Would Python be as dominant in data science today if it hadn’t been given away, with no license headaches? I doubt it would, or that data science would really exist in its current productive form at all. This continued to foster openness in the rest of the Python ecosystem. Likewise for maven, the Apache webserver, and so on. It is good that the GPL and open licenses exist. The famous GPL ratchet, that forces openness on software using GPL libraries, is a good thing. The two work together to some degree, permissive licenses softening a reuse culture up, copyleft forcing it open. There was, and still is, a problem of basic literacy for open source values in many companies. This way we did not all need to hold our breath waiting for them to learn.

Is it unethical to take free gifts, over and again, make money off the use of it, and never give back? Sure, and that freeloading should socially embarrassing and bad for business. For software and content developers: it’s your gift to the world. Leave the strings unattached if you want.

Surprised To Find Them There

There are neat parallels, where assumed supremacy left old property rights intact, in both McGirt v Oklahoma, the case where East Oklahoma is recognized as a Muscogee (Creek) Indian reservation, and Mabo v Queensland No 2, where ‬Australian native title was first recognized.

In both cases the court held that the relevant power could have extinguished title at any time by passing a new law. But the government never got around to it, so the court found the balance of continuity lay with the original, indigenous claim.

In the Australian case, Queensland parliament had the power, and there was no explicit treaty. So it turned on recognizing something like ordinary title property rights based on continuous occupation from before British settlement, and that they were not changed by transfer of the sovereign power.

In Oklahoma, Congress, ie the US federal government, had the power, there was an explicit treaty, and the case turned on rights of self-government not being undermined by sale of ordinary property title within the granted reservation.

The success of these moments of self-determination depend on an inherent tensions in conservatism: fidelity to old written law versus preservation of settled arrangements versus interests that happen to hold power today. Conservatism’s devotion to history is also its vision of the future: the idea that certain old practices and values, perhaps hard to explicitly name, are part of the world to come.

You see this in both cases. The arguments in Mabo include a long history of different forms of property title across the world. McGirt rests on originalist fidelity to the legal text, argued by the conservative Justice Gorsuch. Note too that justice and self-determination here emerges from the messiness of the law: the legislative remnants of a colonial project not carried to its logical rectilinear endpoint.

Oklahoma replies that its situation is different because the affected population here is large and many of its residents will be surprised to find out they have been living in Indian country this whole time. But we imagine some members of the 1832 Creek Tribe would be just as surprised to find them there.

– Neil Gorsuch, McGirt v Oklahoma

He Was Not At All Afraid To Be Killed In Nasty Ways

I ‪was at a family-age party the other night (I know: does my corona-privilege know no bounds?) where pre-teens were running around in knight dress-up gear and throwing out lines from Monty Python and the Holy Grail.‬ It occurred to me that The Holy Grail may be the only Python thing to really last as a cultural artifact – to be a recognizable reference in say, a century’s time. There’s an Alice in Wonderland quality to its deep silliness and texture. The sex jokes are giggly rather than deep blue, the violence is silly red paint rather than grotesquerie, and the religion comedy is mainly surface level‬ jibes at self-serious authority figures. ‪

Consider the coconut running joke. Its a visual joke, an auditory “hook”, an “infinity times” meta-joke, and a very English class commentary on the upper class literally riding on the backs of the peasantry‬.

Though it isn’t the work of Monty Python alone, I’d argue the depiction in The Holy Grail sits at a locus of our current common sense understanding of Medieval Europe. The Victorian images of romance and chivalry are not without their appeal, but also hard to take straight. The idea that the king has power for absurd and hard to justify reasons is much more accessible. It makes sense to us for peasants to be muddy and diseased, for everyone to be pretty anti-scientific and stupid, and Arthur a somewhat overwhelmed duffer. The Holy Grail can’t help including visions of full chivalric romance within it, though. Consider the beautiful passage of Arthur across the misty lake to the castle Aargh. You can tell it was directed by one medievalist and one animator. The moments of vision are setups for jokes, but it doesn’t matter. They are in there all the same.

‪Everyone, Pythons included, name Life of Brian as their masterpiece, and it’s a more coherent work of art. But the institutional Christianity it mocks is much rarer in the UK and other places now than when the film was made. The leftwing political jokes have an eternal quality, though they could go out of fashion quick, and the gender jokes are now weirdly off-colour. Likewise The Meaning of Life is more offensive, more sexually and violently grotesque, but without much “high art” aspiration to justify the investment of time for someone who didn’t grow up somewhere in that cultural milieu.

I find all the movies and most of the sketches very funny. I just don’t expect them to last. At most one might sneak through as a piece for older children, an anarchic discovery where you don’t need to understand all the words at first, because another song, or a silly joke about a killer rabbit, will be along in ten seconds. The other sketches and films then might pulled along in the wake, like A Tale of a Tub trails Gulliver’s Travels.

‪Time is always brutal to comedy. It depends so often on momentary references and sensibilities. To stay around it needs to be attached to some broader body of work, as in Chaucer, or Shakespeare, or Wu Cheng’en 吴承恩. Or it needs to be both surreally funny and deeply textured, like Alice … or The Holy Grail.‬

The Platform Biostate

Quarantine – perhaps we should say the first quarantine – has lifted here, and children have gone back to school. This is of course a matter of tremendous individual good fortune, which I am grateful for, while also wanting to snapshot the alien mindset of this moment into text before new consensus realities congeal and solidify. Perhaps the moment has already gone; the text is here nevertheless.

Governments did act during the crisis according to a different idea of the state. The nascent ideology was only partially realized, but does stand in contrast to what we might call financial neoliberalism. They acted as if the state was an extension of its hospital system, that the health of their people was the most precious asset to defend, and they used networked computing infrastructure as means to that end. They acted as Platform Biostates.

Declaring the death of neoliberalism has of course been a recurring hobby of many political professionals and dilettantes. Declaring COVID-19 the end of capitalism or even just liberalism seems very wishful thinking. But the specific financial neoliberalism that defined public policy from 1970 to 2020 seems intellectually over. This saw the state providing gardeners to cultivate and harvest taxes from markets that were desirable and pervasive. Nowadays even a lively True Neoliberalism Has Never Been Tried book like Radical Markets isn’t much about using money any more, but introducing tokens to voting or immigration to create markets which do not trade in everyday currency.

The basic causal model of financial neoliberalism was that the health of the state was downstream of the free market, free trading economy, and that the health and prosperity of people is downstream of the state and the market economy. Of course there is some truth to this, or neoliberals wouldn’t have been as effective as they were in creating wealth and gaining power for their nations, from China, to Germany, to Chile, to New Zealand. Financial neoliberalism was also so successful because it is a partial ideology. It dictates how to organize markets and where the state should switch from mover to referee, but it also mixes with Chinese Communism, army juntas, social democracy, or corporate liberalism.

Chinese Navy Hospital Ship Peace Ark
Peace Ark 和平方舟/ Daishan Dao 岱山岛



The Platform Biostate is also a partial ideology, but the causal model is different. The power of the state is downstream of the health of its citizens. Money and employment are a means for optimizing that dependency.

Presentation

Sketching some features of the platform biostate, more descriptively than normatively:

Baseline medical welfare. The state looks after the base health of all of its residents. This is both an ideological commitment to care and a herd management technique. Keeping a persistently unhealthy group roaming around the community makes everyone much more vulnerable to communicable disease, so people in the community have to be fed well and have basic medical needs attended to.

Prestige Epidemiology. The science of disease management is the premier technical field. Other fields don’t go away, or stop providing insight, but they are politically secondary. John Hopkins’ COVID-19 dashboard is more urgently read than the New York Times, and the money-conscious watch an infection curve for signs of flattening, not (just) a zigzagging stock chart. Health is the lead department, rather than the Treasury or a Ministry of the Interior. The flawed WHO is more important than the flawed WTO. Oxbridge changes its major to Politics, Philosophy and Epidemiology. Economists already recognize this shift, and have been stunned at attempts to fix a virus with monetary and fiscal policy (consider Paul Romer‘s calls for massively expanded state-funded testing in the US). Economists also expect epidemiologists to be part of a public policy conversation they are not yet used to. 

[Distracted Man/Economists]

[Shocked Face/Explaining Sociology to Sociologists]

[Red Dress/Explaining Epidemiology to Epidemiologists]

External biosecurity. The platform biostate is a hospital system with a navy attached. It knows which people, animals, and plants cross its borders, and it imposes controls on those it wants to exclude. The Platform Biostate wears a mask. This doesn’t make it a hermit kingdom, necessarily, but one with good information about what is near its borders, able to open or close the borders quickly and effectively, and practised at filtering according to current health criteria. And yes, one prone to overgeneralizing the idea to memes, race, or other criteria less biologically grounded.

Internal biosecurity. Threats to community health inside the border can be detected and acted upon quickly. Isolation, testing and tracking are deployed specifically to maintain community health. Alcohol breath testing, mobile health apps, and the prison system state anticipate this. The tracking may be very explicitly individual tracking by a central system, or it may be distributed, but responsive to emerging threats. The social obligation to isolate when sick is a widely held value and backed by law and the police.

Asynchronous delivery / Remote synchrony. Logistics networks end at the house and the office again. Meeting is online by default, remote by default, virtualized by default, a routine piece of greenscreen theatre. Offices themselves fragment and partition. Automated supply chains are bioshock buffers.

Ecosystem complicity. Food, water and air – how casually we let them flow through us; how futile our pretence of the mind’s separation from them. So many more now know what it is like to be breathing death on people, and not able to see the effects until a fortnight later. Without doubting the human capacity for doublethink, everyone did just get an intuition crash course on the diseased exhalations of fossil fuel power.

Causes

Why now? What changed?

Well, meandering failure of the old thing, certainly. Epidemiology was an obviously relevant alternative idea lying around, and with some existing institutional forms, which is always important in a crisis. The platform elements of this form of governance – dashboards, rapid publication, gene-based tests, big data, smartphones – are also relatively new in institutional time. So in one sense we may care about COVID-19 because it is now cheaper to care about it and states may act as platform biostates because there are now technological platforms to use. Another common explanation is a shift of political consciousness – a sudden surge of social-mindedness. Perhaps. I doubt the wellspring of human sentiment has grown so very quickly without some other structural attractor.

Take a world with a wide difference in wealth and income between elites and most people, particularly the 0.1% and the remainder, but also where a top 10% of technically skilled managers and technocrats do quite well: our world of the elephant and Loch Ness monster graphs. This is also an era of dynastic capital, as Piketty has shown. Merit paths are important but so too is ensuring boomer-accumulated capital is stewarded in an orderly way to maintain the next generation’s place in society. As a purely instrumental matter, if you are a member of this 10 or 0.1%, crime, poverty, schooling, clean food, water and air can be protected against in a fairly individualized or family-level fashion. You can get bodyguards, you can live in a compound, you can have air-conditioning and an armoured limousine. It’s not a marvelous way to live, in my opinion, but it’s livable enough. Pandemic, though, is particularly hard to deal with this way; there are essentially three solutions if you’re trying to preserve dynastic capital.

  • Hygienic fortress. The normal compound solution isn’t enough for a contagious pandemic, especially one with an asymptomatic period. You need very strict lockdowns with all your servants and a biosecure epidermis. This is affordable for the 0.1% but not the 10%. For both classes it is in tension with the need to maintain a professional and social network to stay rich. You can’t go to restaurants or fly across the ocean. This era of capitalism works by assembling profit constructs from globally separated opportunities; people need to socially work a cosmopolitan network to navigate and construct that. You can do this online, and reach opportunities you previously couldn’t, but you can’t exploit all the senses for social advantage online; you can’t read body language as well, or catch people in the break of a three day conference. How do you decide when to let people into the fortress? – do they spend two weeks in the citadel hotel? It’s also inaccessible, in our current urban infrastructure, to the technocrats that actually manage the world, rather than owning it.
  • High fertility backup kin. The traditional farmer and landed aristocrat insurance against catastrophe is diversification through redundancy. Have plenty of kids, over multiple generations, and so have plenty of aunts and cousins as well, so that when war or disease does sweep through some of you survive. This is in tension with the multi-decade trend of urbanization and drops in fertility rates. The tendency is to have only a few kids, then hothouse-parent them into a career that will keep dynastic fortunes alive for the next generation (and your geriatric care). Perhaps it will flip the trend and people will start popping out more kids again, but even if it does, it will take decades, and there will have to be some default political worldview in the meantime.
  • Rawlsian contract. If you still want to leave the house and be near strangers, you are faced with a veil of disease ignorance. Especially with contagious diseases with asymptomatic periods, you don’t know whether you will be in the diseased or the healthy role after twelve days, when the veil is lifted. Rawls’ solution of distributive justice then applies – make sure that no-one is ever too sick or uncared for, and you reduce the risk of your own misery or death when the outcome is finally revealed.

Essentially the third solution, familiar in shape due to the historical welfare state, is what rich governments tried to implement in the first half of 2020. Some did it quite well, due to fortunate timing, advantages of population size and geography, or previous experience in biosecurity. For others – like the US or UK – it seemed like the levers of administrative government were no longer connected to any wires underneath. Those states (and sub-national states, like California) that did function, and treated population health as causing wealth, are already in a more powerful position than six months ago. They kept capable people able to work, protected their families, and limited their distress.

John Hopkins COVID-19 dashboard screenshot

Prognosis

Optimizing for overall population and ecosystem health as a generator of power will still create losers (like all state power). A hospital is very good at treating sickness. It is also a bureaucracy that can be indifferent to people in need and make stupid mistakes. Medicine remains very fixated on treatment and on big centralized buildings of sickness control, all put under tension by the needs of keeping a population healthy rather than reacting only to sick people queuing for doctors. The epidemiological view of people as a population – a herd – also leans a manager towards the idea that a herd might be culled for the greater good. 

The platform biostate is a partial ideology: it can be combined with luxury airport quarantine stratification by passport and frequent disease flyer status, or drone-delivered Universal Basic Mask provision, or ideological hygiene death squads, if you try hard enough. It could have a neoliberal variant too: it’s pretty technocratic. Still, I am sympathetic to centreing the state on health, and think it’s perfectly compatible with liberal democracy, which I remain fond of. Financial neoliberalism was pretty good at creating wealth, markets and inequality. The platform biostate could be pretty good at cleaning up industrial toxins, governing by health dashboard and nagging like a harridan when individuals choose fun today over health a decade from now. Some Leviathans may rot; others may send the corpses to pathology for some bloodwork and an updated infection model.


((This piece owes something to Benjamin Bratton’s 18 Lessons of Quarantine Urbanism, which has a wider scope.))