Will the internet survive energy contraction?

If I’m right about the end of the internet, it won’t be an immediate event — rather, costs will rise and access will diminish over time. Whether public libraries are restocked during that process, or whether private libraries become the next information nexus, is a good question What I’ve suggested is that as costs rise and pressures for control escalate, it will gradually become an expensive luxury used mostly by government, big business, and the rich, while everyone else falls back on less sophisticated methods of interaction. It could straggle on for some time before resource shortages or sociopolitical collapse or any of a dozen other things finally pulls the plug.


Obviously a very serious challenge to our assumptions here at the P2P Foundation, by John Michael Greer:

“Given the modern world’s obsession with economic issues, one of the best examples of this reshaping of assumptions by the implications of cheap concentrated energy has been the forceful resistance so many of us put up nowadays to thinking about technology in economic terms. It should be obvious that whether or not a given technology or suite of technologies continues to exist in a world of depleting resources depends first and foremost on three essentially economic factors. The first is whether the things done by that technology are necessities or luxuries, and if they are necessities, just how necessary they are; the second is whether the same things, or at least the portion of them that must be done, can be done by another technology at a lower cost in scarce resources; the third is how the benefits gained by keeping the technology supplied with the scarce resources it needs measures up to the benefits gained by putting those same resources to other uses.

Nowadays, though, this fairly straightforward calculus of needs and costs is anything but obvious. If I suggest in a post here, for example, that the internet will fail on all three counts in the years ahead of us – very little of what it does is necessary; most of the things it does can be done with much less energy and resource use, albeit at a slower pace, by other means; and the resources needed to keep it running would in many cases produce a better payback elsewhere – you can bet your bottom dollar that a good many of the responses will ignore this analysis entirely, and insist that since it’s technically possible to keep the internet in existence, and a fraction of today’s economic and social arrangements currently depend on (or at least use) the internet, the internet must continue to exist. Now it’s relevant to point out that the world adapted very quickly to using email and Google in place of postage stamps and public libraries, and will doubtless adapt just as quickly to using postage stamps and libraries in place of email and Google if that becomes necessary, but this sort of thinking – necessary as it will be in the years to come – finds few takers these days.”


Greer continues with the more general point that we have become addicted to exceptionally cheap energy, and thus a logic of abundance in the physical world, which is now deeply rooted in our worldperspectives:

” It’s been fashionable to assume that the arc of progress was what made all that energy available, but there’s very good reason to think that this puts the cart well in front of the horse. Rather, it was the huge surpluses of available energy that made technological progress both possible and economically viable, as inventors, industrialists, and ordinary people all discovered that it really was cheaper to have machines powered by fossil fuels take over jobs that had been done for millennia by human and animal muscles, fueled by solar energy in the form of food.

The logic of abundance that was made plausible as well as possible by those surpluses has had impacts on our society that very few people in the peak oil scene have yet begun to confront. For example, many of the most basic ways that modern industrial societies handle energy make sense only if fossil fuel energy is so cheap and abundant that waste simply isn’t something to worry about. One of this blog’s readers, Sebastien Bongard, pointed out to me in a recent email that on average, only a third of the energy that comes out of electrical power plants reaches an end user; the other two-thirds are converted to heat by the electrical resistance of the power lines and transformers that make up the electrical grid. For the sake of having electricity instantly available from sockets on nearly every wall in the industrial world, in other words, we accept unthinkingly a system that requires us to generate three times as much electricity as we actually use.

In a world where concentrated energy sources are scarce and expensive, many extravagances of this kind will stop being possible, and most of them will stop being economically feasible. In a certain sense, this is a good thing, because it points to ways in which nations facing crisis because of a shortage of concentrated energy sources can cut their losses and maintain vital systems. It’s been pointed out repeatedly, for example, that the electrical grids that supply power to homes and businesses across the industrial world will very likely stop being viable early on in the process of contraction.”


After showing that national energy grids that waste two-thirds of their energy may not be viable in many places, Greer then makes a plea for distributed energy infrastructures, with a maximum amount of renewable energy to be produced at the local level:

“As the age of abundance made possible by fossil fuels comes to its inevitable end, a great many things could be done to cushion the impact. Quite a few of these things could be done by individuals, families, and local communities – to continue with the example under discussion, it would not be that hard for people who live in rural areas or suburbs to provide themselves with backup systems using local renewable energy to keep their homes viable in the event of a prolonged, or even a permanent, electrical outage. None of the steps involved are hugely expensive, most of them have immediate payback in the form of lower energy bills, and local and national governments in much of the industrial world are currently offering financial incentives – some of them very robust – to those who do them. Despite this, very few people are doing them, and most of the attention and effort that goes into responses to a future of energy constraints focuses on finding new ways to pump electricity into a hugely inefficient electrical grid, without ever asking whether this will be a viable response to an age when the extravagance of the present day is no longer an option.

This is why attention to the economics of energy in the wake of peak oil is so crucial. Could an electrical grid of the sort we have today, with its centralized power plants and its vast network of wires bringing power to sockets on every wall, remain a feature of life throughout the industrial world in an energy-constrained future? If attempts to make sense of that future assume that this will happen as a matter of course, or start with the unexamined assumption that such a grid is the best (or only) possible way to handle scarce energy, and fixate on technical debates about whether and how that can be made to happen, the core issues that need to be examined slip out of sight. The question that has to be asked instead is whether a power grid of the sort we take for granted will be economically viable in such a future – that is, whether such a grid is as necessary as it seems to us today; whether the benefits of having it will cover the costs of maintaining and operating it; and whether the scarce resources it uses could produce a better return if put to work in some other way.

Local conditions might provide any number of answers to that question. In some countries and regions, where people live close together and renewable energy sources such as hydroelectric power promise a stable supply of electricity for the relatively long term, a national grid of the current type may prove viable. In others, as suggested above, it might be much more viable to have restricted power grids supplying urban areas and critical infrastructure, while rural hinterlands return to locally generated power or to non-electrified lifestyles. In still others, a power grid of any kind might prove to be economically impossible.

Under all these conditions, even the first, it makes sense for governments to encourage citizens and businesses to provide as much of their own energy needs as possible from locally available, diffuse energy sources such as sunlight and wind.”


One of the most interesting answers in the comments is from Mash:

In a nutshell, he argues that:

If we end up with a true dark ages of NO technology, maybe we’ll lose the internet. But the very nature of it makes it one of the most resilient networks we’ve created. Because it’s not really “a” network. It’s a lot of little networks connected together, forming something that people find massively useful for information transfer.

and explains why this is so:

” “The internet” is not really “gmail” or “blogger” or “facebook” or “wikipedia”. It’s not even the sum of these things. It’s a way in which you can get sophisticated networking algorithms given some very simplistic mechanisms. (eg: “if you have too many packets, just drop whatever. make the ends figure out when and what to retransmit”)

And “the internet” isn’t really about universal connection. Anyone who has lived in a remote country knows how “the rest of the world” can disappear, but your local sites can continue to function.

So, the devolution of the internet may happen in a very similar manner to the way it originally grew.

Back in the dim dark ages even some of the “central” servers connected to each other on an irregular schedule, and during the time they were connected, exchanged information, and then disconnected once more. (eg: UUCP).

Remember, part of the goal of “the internet” was to continue to provide useful service even in the face of damage. So maybe that damage is widespread, maybe it gets worse, but there is something useful to be gained just from having your computer connected to your neighbor’s computer.

And that’s another part I think you’re missing. Information will always be transmitted SOMEHOW. You think we’ll go back to paper. I suppose it’s possible – if we lose the capability to create ANYTHING electronic (and maybe that will happen someday). But before we get that far, we’ve got a lot of intermediate steps. Think about the telegraph. Places connected with unreliable wires that required a lot of effort to maintain – but it worked because people wanted to be able to say hello.

Now take all the knowledge we now have about using simple networking methods (“the internet”), and reduce what we have available to use it with in steps.

First of all we might lose the “available everywhere”. Undersea cables get cut, satellites fail, the infrastructure as a whole turns into continents of connectivity separated by vast oceans of distance.

Next maybe we lose the reliability. The longer distances (like between cities) have connections – sometimes. Maybe an unreliable wire. Maybe a weak radio link.

But somewhere along that line we have local areas of connectivity – maybe something like a local telephone exchange (except now that we know how, why would we make a dedicated voice service instead of a data service? Now we know how to make voice look like data, and make that voice easier to handle by doing so)

And we have unreliable connections over the long distances. But in “the internet” we have created the exact formulas that allow us to use those connections to continue the information transfer in the most useful way possible.

If we end up with a true dark ages of NO technology, maybe we’ll lose the internet. But the very nature of it makes it one of the most resilient networks we’ve created. Because it’s not really “a” network. It’s a lot of little networks connected together, forming something that people find massively useful for information transfer.

Once you see those intermediate steps of devolution, your questions of “3 essential economic factors” have a lot more answers. It can still go totally black – but that’s a LONG way down.”


Also in the comments, Pasttense writes:

“The internet is going to survive because it is the most energy efficient technology. Carpooling to work vs mass transit? The more efficient is neither–telecommuting from your home is better. Likewise consider an email vs mailing a letter. A few electrons vs all the energy to cut the tree, move the tree to the papermill, create the paper, move the paper to the envelope factory, manufacture the envelope, move the envelope to the wholesaler, move the envelope to the retailer, your trip to purchase the envelope, the envelope’s trip to the post office, to a sorting center, to a couple more sorting/postal offices, the carrier’s trip to your mail box…

As to the efficiencies vs inefficiencies of the grid; note that the generating capacity to serve a million separate households in an off-grid manner is going to be many, many times the generating capacity you need via a grid because of the economies of load-sharing.”


A correspondent from Thailand adds:

“For over a decade I’ve lived in a region of the world where I can experience first hand how people live without an electric grid.

Except for refrigeration, it is quite possible. Most have a car battery. Some wealthy families have two. When the battery is drained, they leave it out on the roadside, and every morning a motorcycle with a trailer comes by to pick up the battery, charge it, and return it before nightfall. The cost for this service is about 30 US cents. This is all rural villagers really need. The battery provides power for lights, phone charging, a fan if necessary, and generally a television set. Except for refrigeration, electric grids are a luxury.

The question always comes up as to why an enterprising individual does not set up his own grid, or why the government doesn’t do it, and the answer is always the same. Theft of the power lines. Unless you can keep them continuously live, they will be stolen and sold for scrap. Only the mafia is immune from theft. During the commodities spike in 2008, Thailand suffered from a high tension power line that blew over in a storm. The reason was because bandits had stolen 6 of the 8 steel bolts that stabilized the tower. It cost over $300K USD to repair. The bolts probably fetched $50 in scrap value.”


We conclude with how John Michael Greer responds to the above challenges and critiques:

“the internet is very useful, therefore it will be maintained.”

That simply fails the logic test. Usefulness and maintainability and viability are separate ideas, with no necessary connection.

Doubtless the Roman messenger relay system was very useful. Important military intelligence was conveyed quickly back to Rome, which was able to use that information to direct troop movements, and adjust political and diplomatic responses to events. And the infrastructure of a vast network of all-weather roads from one end of the Empire to the other was vital not only for military purposes, but the enormous trade it enabled of raw materials and finished goods, which supported the cost of the network and made the centralization of the Empire possible.

And yet, that system ceased to work. Despite it’s obvious value as a communication and commercial resource, the cost of maintaining the roads, bridges and fortifications, and manning the fortresses was so great that eventually Rome gave it up. In urban areas, where it made sense, the network was maintained or even expanded. Outside of those areas, much of the system was simply lost, the roads swallowed by forests, mudslides, wandering rivers, fields and weeds, the fortifications falling to ruin and scavenged for building materials, the bridges succumbing to earthquake and flood.

True, telecommuting is more efficient of power than the current system of physical commuting. But perhaps the choice will be not between telecommuting and a 30 minute drive, but between walking a mile or two to work and not having a job. And what if there simply are no jobs? The current system of employment is only about 250 years old. Prior to that, most people were peasants, slaves, farmers, artisans and merchants. And frankly (I say this as a telecommuter) any job you can do via telecommuting may well be too abstract for the post-peak future.

Email may be more efficient overall than snail mail, but what if there’s no route to host? What if one end or the other of the communication has no computer, or no internet connection, or no power? What if your primary goal for the day is to split enough fence rails to keep that @#$%@# goat in, or to scrounge enough 36ga copper wire to build a new RF modulator, rather than to cruise for pron and respond to JMG’s latest attempt to appeal to the left side of your brain?

Also note that the postal system can be and has been run entirely on muscle power. The same cannot be said about email.”

Some further reactions that came in after sending a draft of this article:

Bill St. Arnaud:

“I would argue the opposite. The one institutional infrastructure that will survive any energy contraction will be the Internet. As you know in Canada we are deploying the worlds first zero carbon Internet – where all routing and cloud/storage nodes are powered solely by local renewable energy sources. – http://www.greenstarnetwork.com/ (warning: Canadians seem to be good at technology bu terrible a keeping web sites up to date. The website is woefully short of information). Greenstar now has 5 nodes in Canada, several in Europe and pending nodes in China and Japan.

As well most people are surprised to discover the biggest energy consumers in their home are not traditional appliances such as fridge and stove, but the aggregate consumption of all the electronic devices in their home. These devices individually draw very little power, but the fact that they are always on in standby mode adds up over time such that their aggregate power consumption over time is greater than the total power consumption of all traditional appliances. All electronic devices could easily be powered by local rooftop solar panels or mini-windmills delivered by 400Hz over existing copper.”
(source Arch-Econ)

Leave A Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.