Sunday, April 05, 2009

The unbearable lightness of information

This decade was the one that was supposed to usher in the era when bits and bytes would replace tons and barrels as the measure of what an economy does. The information economy would eclipse the economy of blast furnaces and railcars.

The allure of such an economy is that it was said to be less resource intense, less driven by the high-amplitude economic cycles of the industrial economy, and more driven by the need for and efficient use of information, something that is always in demand. It turned out not to be so. The tech bust of the early part of this decade highlighted the vulnerability of the so-called information economy to cyclical forces and also the reliance of that economy on the more substantial physical economy.

We mistake the lightness of electrons and the vaporous nature of the information that rides on them for the lightness of the entire economy behind them. Every person who works in the so-called information sector of the economy must be housed, clothed, schooled, provided transportation, provisioned with household goods, given opportunities for entertainment and recreation, supplied with a wide array of public services, and...well, you get the idea. And, much of the manufacturing economy which previously provided employment in the United States and other industrialized nations has simply shifted to China and other low-cost locales. As it turns out, one of the main tasks of the information economy is to direct and manage the resulting global logistical system, a system that continues to bear down with its ever increasing weight on the landscape and the environment.

Howard Odum, the great pioneer in understanding energy flows in nature and society, understood that information, far from being a feathery presence in society, is actually its most resource- and energy-intensive output except for the natural process of species formation.

To read the chart below one must know that Odum turned all measurements into equivalent calories of solar energy which he dubbed solar emcalories. Concentration of emcalories leads to their greater and greater usefulness to human society. Diffuse sunlight on a field only warms a person for as long as the sun shines. But the energy concentrated in field crops can be stored until needed for food or fuel. Such is the role of what Odum calls transformities, that is, the transformation of previously concentrated energy into more concentrated, more energy-intense forms. Transforming fossil fuels into electricity is another example.


Adapted from "A Prosperous Way Down" by Howard T. Odum and Elisabeth C. Odum

Solar Emcalories Needed Per Calorie Produced
Sunlight energy
Wind energy
Organic matter, wood, soil
Potential of elevated rainwater
Chemical energy of rainwater
Mechanical energy
Large river energy
Fossil fuels
Electric power
Protein foods
Human services
1 X 1011
Species Formation
1 X 1015

Odum is not trying to discount the usefulness of information. In fact, energy embodied in the various products of nature and of human societies generally becomes more useful, the more concentrated it gets. Energy that is more concentrated is more easily transported and used. And, energy which becomes the above-mentioned weightless information may be the most potent of all. It was Archimedes who said, "Give me where to stand, and I will move the earth." He was, of course, talking about the power of the lever to move things. The key element, however, is a piece of information, namely, where to stand.

Far from being costless or weightless or light on resource use, information comes to us at very great expense. Today, we talk about the vast volume of information that is being produced. But is this the case? Aren't we really talking about the vast quantity of copies of information flowing through the information system? Aren't we also talking about the vast quantity of gossip that moves through that system? As anyone who sifts through the information on the Internet on a regular basis knows, a good piece of solid, actionable information is not always easy to find. In proportion to the chatter and clutter on the Internet, there simply isn't that much good information. Perhaps one reason is that genuinely useful new information is so very hard to produce.

The idea that ours is a new age when people first began to grasp the importance of information is patent nonsense. When someone tells you that we are moving into an information society, you can retort that we have always been an information society: information about how the forest works and where one might find food, about how to grow crops and which ones grow best, about how to cut and stack stone upon stone to make buildings that will last for the ages, about how to float vessels on water, about nearly everything human societies value past and present. We are now copying and disseminating what information we have on a grander scale and at a faster pace than ever before. And, we certainly have a lot of information about how to make the earth, the sky, and the sea give us whatever we want. In truth, much of the modern "information revolution" is nothing more than this.

What we are lacking is the widespread understanding of how to live within the limits prescribed by the planet. Putting to rest the idea that so-called information-based industries somehow have a negligible impact on the biosphere might be a good first step in focusing us on the kind of information that we will need to become partners with nature rather than its adversaries.


Clifford J. Wirth, Ph.D., Professor Emeritus, University of New Hampshire said...

Hi Kurt,

Good post, and how true it is what you wrote:

"What we are lacking is the widespread understanding of how to live within the limits prescribed by the planet."

In addition, many people don't know that we NEED to live within the limits. The metallurgist-philosopher Chris Shaw has noted this (a good read):

And about information, most of what we needed to know about limits was available from the 1970s in Meadows, Meadows, and Randers -- "Limits of Growth" works, and a 1980 publication by the National Academy of Sciences, "Energy in Transition 1985-2010,"
that provided the information needed to face the Peak Oil crisis. But most energy "experts" never bothered to read the NAS study that they paid for in taxes and which has been available for years in university libraries or free online. They didn't do their science homework and it shows.

As a result we have gotten off on tangents of finding renewable sources of energy, even though is is obvious from reading the 600 page NAS study that we probably won't be able to solve the liquid fuel problem (liquid fuels power agriculture and transportation that is key to our survival), and therefore emergency measures in energy conservation and population control would be necessary to reduce the mass fatalities that now loom. Kenneth Boulding saw the obvious:

When I read the NAS in 1982, I said to myself, there is really no transition to make and this global economic Titanic is "full steam ahead" and won't change course.

Meanwhile back at the ranch, today few "experts" realize that global economic collapse is very near, but some do (read my comment too):

The information that is most important to the world now is: how to survive Peak Oil and Peak Oil preparation. But that information is scant and is avoided by most everyone, even by many energy "experts."

Best regards,

Cliff Wirth

Henry Warwick said...

Wirth is correct. However, I must emphasize some points he misses that seem to run counter, but actually don't.

Basically, our Information and Communication Technologies (ICT) are very energy intensive, however - their productivity continues to explode by leaps and bounds. Example: the IBM 360 / 370 systems were huge and sucked down enormous amounts of electricity. An iPod or iPhone has more computing power, storage, and ability than a room full of 360s. The classic response by Peak Oil theory is that of Jeavon's Paradox.

I would refute the position by spinning Jeavons around. In a resource constrained situation, with decreased energy availability and rising prices, conservation does not lead to more consumption. Conservation is required just to tread water. Conservation in excess of depletion will permit "growth", but over time, one runs into issues of granularity and irreducibility.

In order to get BILLIONS of people into living underneath the depletion curve will require enormous amounts of communication, and the ICT infrastructure will have to adjust to meet these needs.

The needs are multifold, but roughly breakdown into several areas:

There is a need for ICT so people can exchange information over long distances. This is not a new need - Rome had runners. There was the pony express until telegraphy came along, there has always been a communication system for civilisations. They have been much slower than a telephone. But there were fewer people to inform... There is a great deal of value in a postal system at the least, and I'll explain the value of ICT shortly.

At a low level, people have great use for ICT - not only for communication between each other, but to get information from society.

And "society" has a GREAT interest in "informing" its people. And "entertaining" its people.

Civilisation is predicated on class structure and class struggle. Those who are the ruling elite seek to stay that way. And ICT provides advantages.

Example: the Pony Express was developed so the ruling elites could get information from the California gold country more easily. The North needed gold to finance its war against the South, and speculators on wall street were making a PILE on the mining interests and related industries. Telegraphy removed the pony express in 18 months. Messages from the Gold country could be received in NYC within a day or two of issuance, instead of weeks, and it soon evolved into same day communication, collapsing typical notions of space with it.

The ruling elites will demand high speed high quality data. That the working classes can use the system only serves to pacify them, and bring the most recalitrant of the lot (academics and theorists) into the game with the co-optive nature of ICT.

So, it is my conjecture that ICT will be maintained LONG after it ceases to be even vaguely sustainable. If one speculator can find out pivotal information one second faster than his competitors, it can give him a massive advantage.

TV, Radio, all entertainment now devolves on questions of computability: ICT. Telephones are little more than directed computers. ICT is the only game in town for contemporary industrial society. Newspapers are closing down, magazines are nearly superfluous. All these systems act as great social pacifiers as the critical establishment gets to feel like it can voice its concerns, and the ruling elites are happy they are fooled, as ICT provides the means by which the critical establishment refines and tightens the noose around itself. Criticising the system provides it with ideas on how to survive. Not criticising the system results in a repressive society. People will naturally Express Themselves, and this expression does nothing to change the relations of production or the relations of depletion. Hence: the ruling elites win, and get to put a meter on the system as it collapses, and use ICT to keep people pacified, distracted, and amused on the way down.

Geoff said...

Henry, I often read that the productivity of modern IT gear has increased at a massive rate, but I wonder whether we have outsourced a lot of the energy use for gaining that productivity, so the productivity / energy ratio remains similar but is hidden from casual consideration? How much more energy do we expend these days on ultra-pure silicon wafers than was expended previously when tolerances were lower for example?

Even worse when we consider the signal to noise ratio, and then set that against the need to sell excessive product just to finance R&D & production.

If, for example, 75% of computers are used purely for entertainment, but every one of those computers needs to be sold to make it worthwhile for the company making computers, then aren't we effectively wasting 75% of our resources subsidising the 25% of computers doing useful work?

As for the need for ICT to communicate in order to move people below the depletion curve, the majority of current communication is aimed squarely at driving people to consume more. We would be far better off ending communication so no-one knows what the Jone's have got so they don't bother trying to keep up!

Henry Warwick said...

Hi Geoff!

you wrote:

Henry, I often read that the productivity of modern IT gear has increased at a massive rate, but I wonder whether we have outsourced a lot of the energy use for gaining that productivity, so the productivity / energy ratio remains similar but is hidden from casual consideration?

At present, ICT eats about 6% of American energy consumption. In 1993 I bought a computer (apple centris 650) that screamed along at 25mHz. With a 17 inch colour monitor it cost almost $3000 and was "large". It also required a CRT, which ate huge amounts of juice.

The computer I am using now is a dual 2.2gHz laptop - literally 176 times faster. The Centris had 16 megs of RAM, my laptop has 2 gigs, 125x as much. The Centris came with an 80 meg HD. My laptop has a 160gig HD, literally 2000x the storage. It uses a fraction of the materials my laptop is composed of. The power consumption differences are also dramatic. The Centris ate 230w @ 1.9amp. My laptop uses about 19 watts sitting around, 53 watts doing some heinous rendering, so I would guess the power consumption is around 23 watts - 1/10 the consumption of the Centris and its monitor.

However: I frankly don't need even that much most of the time, and this is where it gets interesting...

The iPhone runs a 620mHz ARM processor, which is VASTLY more powerful than the 25mHz 68030 processor in the centris. It's poewr consumption is a fraction of my laptop. The power drain comes in with the screen.

How hard would it be to make a phone that was also a computer with 2 USB points (one for kbd/mouse, the other for hard drive) and a video out for a monitor/projector? Pretty simple, I'd say... AND a music/media center? Through the handsfree earphone jack.

So, I could write posts to this site, listen to music, and watch video, and even (alors!) make a telephone call.

Massive convergence into a small device.

Laser video projectors are being developed that are the size of a small book. No lenses. Miniscule energy requirements.

Now, don't get me wrong: NONE OF THIS IS A SOLUTION. What it is, is a path of conservation to maintain ICT into the future.

So, pretending that ICT continues to use 6%, it will be 6% of a shrinking pie. These kinds of developments will be the first "go to" for the system in order to maintain the ICT infrastructure.

Even worse when we consider the signal to noise ratio, and then set that against the need to sell excessive product just to finance R&D & production.

Much of that will get winnowed out as resources get tighter. It may not be possible to maintain a 24/7 ICT infrastructure. In which case, you will see a collapse of the internet into a set of balkanised meshnets. These meshnets will come up and go down as power permits. So email from me to you or this site might not happen in seconds. It might take a day or so as the mail is stored and forwarded.

Also, the cost of messaging could exist. Instead of me typing all this for free, I might be charged per k of transmission, and each hit on the site might also be dinged by k of transmission. I would submit that would be the end of Adobe Flash sites, forever. But it would also eliminate spam. you can tell I am shedding bitter tears over all of that...not.

Entertainment is NOT WASTE. Entertainment is utterly critical to the maintenance of a civil society. Even the poorest neolithic aboriginals party. A lot. Entertainment - i.e., nonproductive activity for the amusement and/or edification of an audient - is vital to maintaining dominant ideological memes in order to form a more perfect hegemony. And if there is this device convergence (and I beleive it is only a few year away) society will INSIST on these technologies, even when it is clear they cannot be maintained into the future.

The working class will insist on their TV shows, music, games, interweb, etc. The ruling class will insist on it as well because they also like entertainment, but they also know that nothing keeps people off the streets like TV in the living room.

I hope that explains it better for you.

I do agree: ICT has a limited future, but I am suggesting that its future will be extended far beyond its maximal competence thansk tothe advantages it provides and the entertainment and social management it provides.



Rice Farmer said...

Your comment about quality vs quantity was interesting. Yes, there is a lot of duplication. Of course, not all of that is useless or wasted, and should be considered "redundancy," which is useful. For example, I might want a big of valuable information that is unavailable from one source (server down, etc.), but available from another. Same goes with network routing.

The observation about entertainment by Mr. Warwick is also instructive.

At the same time, the problem of quality vs quantity still deserves consideration. For example, let's say your purpose is to be well informed about world events and what is going down on the global chessboard, and you watch TV with that in mind. Now there is a vast amount of information belching forth from the TV, but with the above-stated purpose in mind, the quality is extremely low. In fact, much news is actually just propaganda disguised as news. Here in Japan, too, the TV and newspapers are more propaganda organs than accurate information providers. Thanks to that, the average person actually has no understanding of why the global economy is falling apart, and have never even heard of peak oil.

So in that sense, the signal-to-noise ratio is something like 99:1, and most of the energy consumed is wasted.

But that may be just a value judgment, because if your purpose is to laugh at hours of inane jokes, the energy is well spent.

atomicat said...

I can't tell you exactly where to stand but I have a good idea what it might take to stand and be much more resource efficient. It's not going to be easy.

Goal should be: for each US citizen to use 1/10th the energy/money we are currently using.

This means long term efficient goals would look something like this.

Small plots something like 100 by 50 mile for a large eco friendly suburbian farmtropolios of sorts. Each one having a plot of land, a small structure very efficient and some means of low energy transportation, think enclosed battery powered velomobile trike. 500 watts of power for your cycle.

The house (small shell) may be like a small room or trailer in a earth ship that is perhaps partially sunk below ground. Superinsulated, with very small heat cooling needs. Goal 1000 watts energy or less. The entire homestead may have other rooms all passively heated or cooled. You have a very small laptop low energy requirement and the big appliances are things like a 150 watt crock pot. Your energy needs for the entire day are 2000 to 4000 watts per day for your household. I found this to be my 21k scenario for the ideal suburbia like farming community. We could house perhaps the entire 20 million folks in Michigan in such a land mass and configuration. Since we are in a 50 by 100 mile stip, the entire thing can be traversed with a small bike in a day. If your using the 500 watts to motor your bike up to 40 MPH max. You can get across the long site on an expensive energy trip with 4 hours of energy or 2000 watts.

You can swap batteries. They are charged with solar and wind. We use compost and have solar greenhouses. Greenspaces and small industry exists. The only users of fossil fuels and old styled vehicles are military. Plenty of the youth want to join to drive and have the thrill of an engine driving experience.

Most of the time could be spent at home with some work. Some small industry would exist, but the goal is to be efficient. It would be mostly a poor farming life like India using the Induri method of farming. Some specialists work full time, but most work part time and grow most of their own food. There's some industrial farming and industries for some items, mostly recycling.

Downside, it will only work if all nations do it. If some compete and competition is in the nature of the natural world, then this won't work and we cannot pay off the debts because GNP output is tied to energy. So all world debts must be forgiven for this to work and much of the competiton even between nations must cease. The alternative is many resource wars and the four horsemen of the apocapylse.

atomicat said...

Also all the talk about large server farms, computers and the like are foolish and a waste of energy in a sustainable future. The best we could likely hope for is something like small XO laptops. Micro laptops for small entertainment and home based learning experience. Some portable USB drives and a wifi serverless based mesh network tying networks of home systems together.

Get rid of all the huge power sucking server hardware. Have what you need on local drives and do distributed processing of information via sharing across the mesh networks.

An XO laptop will run on 1 to 2 watts of energy an hour. Power requirements for charging are 20 watts an hour however which means the AC charger is pretty inefficient.

Other subnotebooks of the future should be at least that efficient. I need the power for more important things like cooking, heating cooling, pumping water from my watercatchment cistern, and other necessities. Not massive computer sucking power.

We won't have washing machines and dryers like we do now in a house of the future. Hang the laundary to dry in the greenhouse.

You have to think very harse in cutting back as an extreme goal. Then figure how to get that energy from a small production of renewable and scale up a little bit. This means things like "tent beds" or a "tent bed trailer inside a small trailer room, inside your earth bermed house". Not a lot of extra wood even for cooking.

A 4000 watt lifestyle doesn't even have enough energy to provide a single horse per family. Even an Amish farmer with one horse is not sustainable globally. We have to think "tiny" really small and efficient. Talk about server farms, huge HDTV screens, flying on an airplane with hydrogen fuel, are all nonsense for high efficiency. Think small.

Those who plan early and build early can survive in the northern states. Those who don't will migrate south, where they will face water shortages. The masses of lazy folks will move southward to crime infested high population areas.

Geoff said...

"Entertainment is NOT WASTE"

Unfortunately it comes down to a question of the purpose of existence, doesn't it? Are we here to sit in front of the television for 6 hours a night with our mind numbed? Our prevailing philosophy would allow that as a valid use of our time, for we are considered free to do whatever we want in this age.

I personally disagree, to an extent, though I'm certainly not one to insist that we should be undertaking productive work 24/7.

Entertainment currently seems to be more about the consumption of energy for the pointless gratification of the senses, a gratification which seems to leave people disappointed and wanting more, rather than leaving them satisfied in any way.

Entertainment as a way of binding societies together and sharing a cultural narrative seems to be long gone from our western culture. Most entertainment seems to be more divisive than inclusive. If 90% of news is propaganda then we don't even gain education from the entertainment we immerse ourselves within.

This is contrary to genuine physical entertainment like sitting around having a beer and a yarn on a summer afternoon with good friends, seeing a live band, attending a working bee etc.

I realise all this is ultimately subjective and down to personal opinions & belief systems, but we're talking about the future of our & other species, faced as we are with "peak everything" and a collapsing global ecosystem. Do we cling to, and continue to promote, the dubious social value of 99% noise for that 1% signal in the face of all that?

The 1.7kg Microchip paper by Eric Williams & Co (proper article is behind a paywall, but a power point with interesting figures is available here: indicates that a 2g 32Mb SDRAM chip uses 1.7kg of resources to produce it. Modern chips would not be much different, if they weren't worse, due to the fact they require higher purities and more intricate fabrication. As the powerpoint version points out, the lower the entropy required the more energy expensive it is.

Ultimately this means that whilst the iPhone runs magnitudes more quickly and with magnitudes less power than it's predecessors, it's also using more energy to produce than is consumed in it's lifetime.

The PPT indicated that for the average desktop system with CRT production energy was 83% of total lifecycle energy. That's for an old style computer of course, but as the devices become more efficient the manufacturing becomes more complex so I don't believe we are going to see a reversal of that fundamental relationship.

This situation is only going to get worse, and when we multiply it by the number of people who want access to this dubious entertainment it seems we're setting ourselves up for even more pain in the future.

Henry W originally said "In order to get BILLIONS of people into living underneath the depletion curve will require enormous amounts of communication"

In a nutshell, billions of people will be moved to living under the depletion curve simply by virtue of the fact that there is no other option. No communication will be required. The less communication there is, the less exotic wants people will have, and the happier they'll be with what they've got.quet

Henry Warwick said...

I agree with those writing that the big systems are not sustainable. But *how* they evolve is of great interest to me, and I believe of vital interest to society.

Example: the Telegraph was implemented for the interests of the ruling elite, not the proletariat or peasantry. the amount of energy expended per message in a telegraph was very high. Even higher for the Pony Express. Right now, it has been estimated that a search on google eats enough juice to heat a cup of coffee. It's a lot when you think about the trivial amount of effort a search requires, but it's nothing when you compare it to a message on a telegraph in 1860.

Rather than giant server farms providing instant access, "seeded" viral search algorithms could be implemented over a variety of linked meshnets. so, the meshnet of my neighbourhood might have my 183 gigs of music mp3s available. Bounced out at some future speed (faster than today) and these 180gigs are parked on some multi-petabyte storage drive. so you search for "Dead Kings" by "Espers" and "Electricity" by "Orchestral Manoeuvres in the Dark" late in the afternoon. Electricity is behind you in Hawaii, so the link to that shows up instantly, and you DL it in seconds. Dead Kings is on my drive, but it's dark where I am so the meshnet is down. Other meshnets know this, so the query is stored until my neighbourhood meshnet comes up and the next day, you get your link emailed to you.

the massive storage is predicated on Free Admission. If inter-meshnet email and messaging suddenly had a charge attached to it (similar to the postal service) you would see the volume collapse. So, intra-meshnet might be free - allowing communities to quickly communicate and organise and entertain themselves. But comunications between gets expensive, as your email would require storage until the target meshnet was available.

These are just a few of the ideas for transition period ICT.

Note, this is not a discussion of postCarbon ICT. That may prove to be an oxymoron. We don't know yet - it is a highly dependent variable, whose dependence is on a factor that is largely unknown.

About Entertainment:

People Will Always Have Fun. Period.

Right now it is largely through media technology. This may not always be true. But people WILL HAVE FUN.

It's not a negotiable question. It's what people do.