Starlink

A familiar tale of tragedy

You've probably heard of The Tragedy of the Commons. This is the idea that if a natural resource (the commons) is available to all to use as much as they like, it will inevitably become overused and ruined. This is a common economic argument which has been cited in contexts ranging from healthcare to the Internet.

However, the argument wasn't popularized by an economist. In 1968, it was American ecologist Garrett Hardin who published The Tragedy of the Commons (PDF).

Hardin's article starts out by making clear that he believes he is discussing the class of problems for which no technical solution exists:

In our day (though not in earlier times) technical solutions are always welcome. Because of previous failures in prophecy, it takes courage to assert that a desired technical solution is not possible. Wiesner and York exhibited this courage; publishing in a science journal, they insisted that the solution to the problem was not to be found in the natural sciences. […] My thesis is that the “population problem,” as conventionally conceived, is a member of this class.

So Hardin's focus was on population control. This was a popular issue at the time. Paul and Anne Ehrlich had released a book called The Population Bomb, which predicted world famine in the 1970s and 1980s due to overpopulation. Harry Harrison had written the book “Make Room! Make Room!", later turned into the movie “Soylent Green”, which portrayed an overcrowded and starving world. The story starts on the then-distant day of Monday, August 9th 1999:

On this hot day in August in the year 1999 there are—give or take a few thousand—thirty-five million people in the City of New York.

The population of New York in August 1999 was actually 18.8 million; Harry Harrison hadn't foreseen the white flight and urban decay of the 1970s. However, that's not the actual reason why we weren't all crammed together in slums surviving on a subsistence diet of soy-lentil cakes. (The book doesn't mention cannibalism; that was added to get the movie studio interested.)

There are some people who have concluded that we didn't starve because the doom-and-gloom brigade were simply wrong. These cornucopians believe that thanks to human ingenuity the population of the earth is practically unbounded.

It was actually thanks to a particular human that we didn't see mass hunger, at least in the west: Norman Borlaug. His new crop strains roughly doubled yields, and famine rates plummeted, in what became known as the Green Revolution.

Unfortunately, we now need another Green Revolution to double food yields by 2050 if we're going to keep avoiding famine. We need to achieve that even as land becomes leached of nutrients, climate change reduces crop yields, the aquifers industrial farming relies on run dry, and oil to make industrial fertilizer starts to run out.

Hardin had a different solution in mind. First he set out his view of the idea:

The tragedy of the commons develops in this way. Picture a pasture open to all. It is to be expected that each herdsman will try to keep as many cattle as possible on the commons. Such an arrangement may work reasonably satisfactorily for centuries because tribal wars, poaching, and disease keep the numbers of both man and beast well below the carrying capacity of the land. Finally, however, comes the day of reckoning, that is, the day when the long-desired goal of social stability becomes a reality. At this point, the inherent logic of the commons remorselessly generates tragedy.

After giving some more examples, Hardin begins describing his proposed solution under the heading “Freedom To Breed Is Intolerable”:

If each human family were dependent only on its own resources; if the children of improvident parents starved to death; if, thus, overbreeding brought its own “punishment” to the germ line–then there would be no public interest in controlling the breeding of families. But our society is deeply committed to the welfare state, and hence is confronted with another aspect of the tragedy of the commons.

He then calls on the US to reject the UN Declaration of Human Rights, and to begin programs to control who should be allowed to have children.

He returned to the theme many times over the rest of his life. In 1970, for example, he wrote “Parenthood: Right or Privilege?” for Science Magazine:

But in the long run a purely voluntary system selects for its own failure: noncooperators outbreed cooperators. So what restraints shall we employ? A policeman under every bed? Jail sentences? Compulsory abortion? Infanticide? … Memories of Nazi Germany rise and obscure our vision.

Yes, Hardin felt that if only people weren't distracted by memories of Nazis, America would be able to get on with reasonable eugenics. You probably won't be surprised to lear that he also had some ideas about who should reasonably be allowed to breed, and who should not; you can read about them on the SPLC web site.

Apart from being a racist eugenics crackpot, Hardin was wrong. Economist Elinor Ostrom won a Nobel prize for her work proving it. As the New York Times summarized:

Traditionally, economics taught that common ownership of resources results in excessive exploitation, as when fishermen overfish a common pond. This is the so-called tragedy of the commons, and it suggests that common resources must be managed either through privatization or government regulation, in the form of taxes, say, or limits on use.

Professor Ostrom studied cases around the world in which communities successfully regulated resource use through cooperation. Her work has important applications for climate change policy today.

As Ostrom put it, “A resource arrangement that works in practice can work in theory.”

In short, the Tragedy of the Commons is nowhere near as universal as some would have you believe. The Internet you're using right now is an example of the Tragedy of the Commons failing to take place. However, there are cases where it occurs.

The Kessler syndrome

There are approximately 5,000 satellites in low earth orbit, of which roughly half are active. They range from military KH-series spy satellites, to GPS satellites, to Iridium mobile phone satellites. There are also around ten times as many pieces of “space junk” — small pieces of debris.

The problem is, once something is placed in orbit above the earth's atmosphere, it tends to stay there. In 1965, Ed White lost a glove during his Gemini 4 spacewalk, and it's almost certainly still up there somewhere, along with a rather nice Hasselblad camera that Michael Collins lost during the Gemini 10 mission. Oh, and 15 years of feces and urine from the Mir space station, which had a toilet that vented out into space.

In 2009, Iridium satellite #33 collided with an obsolete Russian military satellite called Kosmos 2251. The result was 2,201 detectable separate pieces of space debris, of which 315 have so far fallen and burned up in the atmosphere.

With orbital speeds up to around 50,000km/h, it doesn't take a very big piece of space debris to cause damage. In 2016, a piece of debris the size of a fleck of paint caused a 7mm impact in one of the windows of the ISS.

In 1978, NASA scientist Donald J. Kessler noted that if the amount of space debris went beyond a critical level, a cascade of collisions could occur and produce more and more debris, knocking out more and more satellites. Obviously a Kessler cascade becomes more likely the more satellites there are. That number increased at a fairly constant rate of around 120 a year until 2013.

Wishing upon a large number of stars

During the late 1990s, a lot of companies that should have known better decided that they needed to diversify into Internet and software technology. In 1999, PC clone manufacturer Compaq decided for some reason that they needed an Internet-based city guide, so they bought a company called Zip2 that had created one, paying $307 million in cash. By 2002, they had accumulated $2.2 billion in debt, and were bought by Hewlett-Packard. However, one of the beneficiaries of the Zip2 deal was one Elon Musk, who made $22 million.

After selling Zip2, Musk founded X.com, a service for handling online payments. It merged with another company to become PayPal. Elon wanted PayPal to ditch Unix and go with Windows for all of its servers, and got ejected from his CEO role. Then in 2002, online auction site eBay decided for some reason that it needed to own a credit card transaction processor, and bought PayPal for $1.5 billion. eBay realized it didn't need to own PayPal in 2015, but by then Elon had trousered another $165 million.

And that's how he got the money to start one of his current ventures, SpaceX, which has the long term goal of landing on Mars. In 2008 they became the first privately funded organization to put a satellite into earth orbit.

Elon's latest project is called Starlink. It is intended to be a worldwide satellite Internet service provider. I'm skeptical, because of the example of Iridium. Iridium service prices are still eye-wateringly high, and they only need to cover the cost of running and periodically replacing 75 satellites. Starlink need 4,409 satellites just for phase 1 of their deployment, so how expensive do you think Starlink service is going to have to be to cover their operating expenses? How about when they deploy the remaining 7,500 or so satellites for phase 2? Don't forget, the satellites will have about a 5-year lifespan, so they will be replacing 40 of them each week. But don't worry, all the rare earth metals used for the large solar panels will be scattered when the satellites burn up, they're 100% nonreusable.

No, whatever the PR web site suggests, I don't think this is going to end up being a system for grandma in her lakeside house in rural Minnesota to get on the Internet; I think this is going to be like Iridium, a super-pricey niche product for businesses, and probably one that goes bust once or twice and has to be bailed out by the government. But hey, maybe I'm missing something, maybe the economics will work better this time?

Anyway, whether it works as a business or not, in May 2019 SpaceX launched a string of 60 Starlink satellites. This month, they launched 60 more. It wasn't long before astronomers noticed. It might not be long before it's basically impossible to do long-exposure astrophotography during twilight hours.

That's bad, but it's the possibility of a Kessler cascade that worries me, and I'm not alone in that worry — just as Starlink aren't alone in their plans to fill the sky with satellites. Jeff Bezos also wants to own a worldwide satellite Internet network, because of course he does, so he has plans to launch 3,236 satellites of his own. A third company, OneWeb, plans to launch another 650.

My verdict? I like the idea of global Internet as much as the next excessively online person, but this plan is wasteful, risky, and economically stupid.