Exponential Technological Progress

[AWPrime]

From: http://forums.randi.org/showthread.php?s=&threadid=55386


Originally posted by Kevin_Lowe
Technology advances exponentially, so it will be a few centuries at most before the rate of change approaches infinity and we hit whatever physical limits the universe will impose on us.

That only applies when population and available energy is exponentially as well.

And currect trends are showing that this might not be the case.

[AWPrime]

This thread is about if Technology will always progress Exponentially.


My position is NO. It would require endless exponential growth of resources.

[drkitten]

Originally posted by AWPrime

My position is NO. It would require endless exponential growth of resources.

Not necessarily, if resources are fungible. For example, "energy" is a largely fungible resource -- when we started to run low on whale oil, we switched to kerosene, then to gasoline, and may in the future switch to hydrogen, or unobtanium, et cetera. Since many resources are the result of technology, as long as the resource-creating technologies can keep pace with the resource-consuming ones, it's not really an issue.

Of course, your question admits of a silly naive answer of "of course not, NOTHING can persist forever." But any particular hard limitation you would like to suggest can probably be overcome by other technologies.

[phildonnia]

You seem to be confusing "technology" with "technological artifacts." It is true, the number of giant plasma screens cannot increase exponentially. But "technology" is nothing more than knowledge, which doesn't follow the same laws as matter.

I found this one day, which is pretty fascinating:
http://en.wikipedia.org/wiki/Omega_point
Frank Tipler apparently discovered that, with certain assumptions, it is possible to model infinite "computational power" in a finite universe.

[AWPrime]

Originally posted by new drkitten
Not necessarily, if resources are fungible. For example, "energy" is a largely fungible resource -- when we started to run low on whale oil, we switched to kerosene, then to gasoline, and may in the future switch to hydrogen, or unobtanium, et cetera. Since many resources are the result of technology, as long as the resource-creating technologies can keep pace with the resource-consuming ones, it's not really an issue.
The replacements might not give the same rate of energyproduction.

[pgwenthold]

quote:
--------------------------------------------------------------------------------
Originally posted by Kevin_Lowe
Technology advances exponentially, so it will be a few centuries at most before the rate of change approaches infinity and we hit whatever physical limits the universe will impose on us.
--------------------------------------------------------------------------------


Infinity does not exist. The rate of change will never approach infinity. No matter how high it goes, the rate of change will always infinitely slower than infinity.

[AWPrime]

Originally posted by phildonnia
But "technology" is nothing more than knowledge, which doesn't follow the same laws as matter.

I was taking about knowledge. And it is bound by physical limits:

- Research requires resources.
- Education requires resources.
- As knowledge increases also the number of specialists will need to increase as well.


Another thing that is assumed is that civilization will remain stable enough to support all the knowledge.

[drkitten]

Originally posted by AWPrime
The replacements might not give the same rate of energyproduction.

Doesn't matter if we have technological methods for compensating for the differences.

There's more energy density per liter in gasoline then there is in alcohol -- so one simply burns more liters of alcohol.

[drkitten]

Originally posted by AWPrime

- Research requires resources.
- Education requires resources.


Of course, technology makes it possible to do both with fewer resources (or alternatively makes resources more widely available). I can, for example, do research anywhere I have an internet connection; I need no longer locate myself near a world-class library.

I can use modern production techniques to purchase and acquire equipment much more cheaply than I could have even ten years ago.

As I said, as long as the resource production technology stays ahead of the resource consumption technology, it's not a problem.

[AWPrime]

Originally posted by new drkitten
Doesn't matter if we have technological methods for compensating for the differences.
IF

There's more energy density per liter in gasoline then there is in alcohol -- so one simply burns more liters of alcohol.
And what if you can't produce enough alcohol? (physical limits)

[drkitten]

Originally posted by AWPrime

And what if you can't produce enough alcohol? (physical limits)

The same thing that happened when we couldn't produce enough whale oil.

We produce something else instead.

"As long as the resource-creating technologies can keep pace with the resource-consuming ones, it's not really an issue."

[AWPrime]

Originally posted by new drkitten
The same thing that happened when we couldn't produce enough whale oil.

We produce something else instead.

"As long as the resource-creating technologies can keep pace with the resource-consuming ones, it's not really an issue."

There are physical limits into producing energy, it has to come from somewhere. And not all forms of energyproduction may be easy/efficient to use.

[Rolfe]

Is it possible, practically speaking, to distinguish the first part of an exponential increase from the first part of a sigmoid curve? Or even a bell curve, if one were being a pessimist?

I ask because there has also been talk about exponential growth in population, but when you look at populations you can study under controlled conditions (bacterial cultures) you find that although the growth curve starts off like that, as it progresses it becomes clear that in an open system, where nutrients can come in from outside, it's a sigmoid curve, while in an open system with finite resources it's a bell curve, as the colony runs out of nutrients.

I always thought that the trick with the Earth was to make sure that one never exceeded the energy input from the sun, and so achieved the sigmoid curve, rather than running out of resources and finishing up with the bell curve.

I'm not saying that a real exponential increase in something abstract like "technology" is impossible, but if you're actually observing the early part of a curve like that, could it not just as well be a sigmoid curve?

Rolfe.

[drkitten]

Originally posted by AWPrime
There are physical limits into producing energy, it has to come from somewhere. And not all forms of energyproduction may be easy/efficient to use.

These are technological issues and can be overcome technologically.

How many times need I repeat myself? "As long as the resource-creating technologies can keep pace with the resource-consuming ones, it's not really an issue."

You are assuming a continual increase in the demands placed upon resources by increased technology, but at the same time assuming that the technology used to create resources is not increasing. You don't get it both ways, sir.

There's a famous bet between Julian Simon and Paul Ehrlich regarding this. Ehrlich, a famous population theorist, predicted a resource crunch as human demand grew. Simon predicted, instead, that the availability of resources would grow as demand grew. You can read a fuller description on this Web page (http://www.overpopulation.com/faq/People/julian_simon.html), but the capsule summary is that Ehrlich named five strategic metals (copper, chrome, nickel, tin and tungsten) in 1980 that he expected would be seeing shortages through resource depletion.

In 1990, all five had actually fallen in price (in constant 1980 dollars). In fact, even taking inflation into account, the total cost of the metals was lower in 1990 dollars than in 1980 dollars. If you look at the track record of human history, we are getting better, not worse, at finding materials or making better use of substitutes, whether the substitutes are fuel, metals, food, energy, et cetera.

[drkitten]

Originally posted by Rolfe


I always thought that the trick with the Earth was to make sure that one never exceeded the energy input from the sun, and so achieved the sigmoid curve, rather than running out of resources and finishing up with the bell curve.


Only if you assume that we will always be dependent upon the sun for our energy supply. At this point in our technological progress, this may seem a reasonable assumption. If it ever gets to the point where we are likely to exceed the energy input from the sun, I expect a large investment in technology to figure out a way to extract energy from non-solar sources.

[Rolfe]

Originally posted by new drkitten
Only if you assume that we will always be dependent upon the sun for our energy supply. At this point in our technological progress, this may seem a reasonable assumption. If it ever gets to the point where we are likely to exceed the energy input from the sun, I expect a large investment in technology to figure out a way to extract energy from non-solar sources. Well, I nearly said "or nuclear sources", but then I remembered that my school teacher included these as "from the sun" because the matter that the Earth consists of originally came from the sun in the first place.

If you look at it that way, then you have to postulate interstellar travel to move outside that, in which case you just substitute the energy of the universe. If you have a sufficiently large Winchester, your bacterial culture can go quite a way before the sigmoid or bell-curve end-ponts start to show.

Rolfe.

[drkitten]

Originally posted by Rolfe
Well, I nearly said "or nuclear sources", but then I remembered that my school teacher included these as "from the sun" because the matter that the Earth consists of originally came from the sun in the first place.


I believe current scientific thinking is that your school teacher was wrong (conservation of angular momentum or something -- 99% of the mass of the solar system is in the Sun, but 99% of the angular momentum is in the planets, and that Just Ain't Right.)


If you look at it that way, then you have to postulate interstellar travel to move outside that, in which case you just substitute the energy of the universe.

... unless we can find some extra-universal source of energy, such as being able to tap into quantum fluctuations on the other side of the Trousers of Time.

That's the problem with technological predictions, especially about the distant future. If history is any guide to go on (and I think it's the only guide we have), I feel confident stating that technology 100 years from now will include both things we currently consider to be impossible as well as things no one alive today has yet considered at all. For this reason, I feel very, very confident in rejecting any one particular projection of "this kind of thing cannot be done."

AWPrime quotes Clarke's Third Law in his signature.

I would like to quote his First Law:


When a distinguished but elderly scientist states that something is possible he is almost certainly right. When he states that something is impossible, he is very probably wrong.

[AWPrime]

Originally posted by new drkitten
These are technological issues and can be overcome technologically.
There are limits like you can at best only get 100% efficieancy, no more.

No amount of technology is going to change the laws of physics.


You are assuming a continual increase in the demands placed upon resources by increased technology, but at the same time assuming that the technology used to create resources is not increasing. You don't get it both ways, sir.
Buzzzzz, Wrong.

I am stating that increase of demand is growing faster then the resource tech.

[drkitten]

Originally posted by AWPrime

I am stating that increase of demand is growing faster then the resource tech.

And I'm not letting you.

Because it's false, has been false for most of human history, and will almost certainly continue to be false for a very long time to come.

[AWPrime]

Care to back it up?

[AWPrime]

Anything involved in the real world is finite; forms of energy, number of substances, efficiency improvements, possible knowledge per person, amount of food grown on land, amount of fresh water even the stability of an civilization is limited.

-

As we gain more knowledge we must support that knowledge. And because the human mind can only integrate a part of science, specialization will become necessary. The relevant information will then be more splintered into more fields and therefore harder to use.

Then there are diminishing returns on scientific progress. This means the more complex it gets the more resources it will take to take a similar step forward. So at some point, some experiments will be deemed too expensive and trivial to be performed by anyone.

We can see this happening as professional scientists concern themselves with more and more narrow area of specialist expertise. And your amateur scientist who performs useful science is a rarity nowadays.

-

We can always hope for technology to come along and make the technology required for scientific progress less resource-intensive, but since this technology is also subject to fundamental diminishing returns, it is pretty much inevitable that we will come to a point where everything grinds slowly to a halt.

-

However you could maybe reset the point of diminishing returns with (new base values); genetic manipulation for higher intelligence and more energy available with fusion energy. But that still leaves limits and I have a feeling that we would waste most of the new resources on trivial matters.

In the end the most realistic scenario would be up and down periods of technological advancement. With a general upward trend so long as nothing truly apocalyptic happens.

[drkitten]

Originally posted by AWPrime
Care to back it up?

Already done.

[drkitten]

Originally posted by AWPrime


We can always hope for technology to come along and make the technology required for scientific progress less resource-intensive,


Always has so far. Dr. Ehrlich, meet Dr. Simon.

[Drooper]

Originally posted by AWPrime
From: http://forums.randi.org/showthread.php?s=&threadid=55386

quote:
--------------------------------------------------------------------------------
Originally posted by Kevin_Lowe
Technology advances exponentially, so it will be a few centuries at most before the rate of change approaches infinity and we hit whatever physical limits the universe will impose on us.
--------------------------------------------------------------------------------


That only applies when population and available energy is exponentially as well.

And currect trends are showing that this might not be the case.

Misunderstanding here.

Exponential growth means that the rate of growth will be contstant, not increasing.

Technological progress is the underlying driver of economic growth (adjusted for available labour and capital). It has been remarkable stable over time across countries around the world.

[Iconoclast]

Originally posted by pgwenthold
Infinity does not exist.Sure it does, in fact I'm writing a paper on it at the moment, I'm trying to find out if it's as big as everybody says. It's taking forever.

[drkitten]

Originally posted by Drooper

Exponential growth means that the rate of growth will be contstant, not increasing.


Rate of growth measured as a percentage will be constant, but not in absolute terms. In absolute terms, exponential growth does in fact mean a continuously increasing rate of growth.

For example, I put a thousand monetary units of your choice in an investment paying ten percent a year compound interest. The rate of growth will be exponential. Today I have a thousand units. Next year I'll have 1100. In seven years, I'll have about 2000. In twenty years, I'll have about 8000.

But if you look at how much money I'm making, I'm getting an income of 100/year next year, but I will be getting an income of 800 units/year in 2025. So in absolute terms, my "rate of growth" has increased eight-fold in twenty years.

In discussing resources and so forth, it's usually the absolute terms that are of interest. For example, the local airport may have 100 flights a day (and be growing at 10%/year). Twenty years from now, it will be taking in 800 flights/day -- which means that someone somewhere will have needed to essentially build seven complete new airports to handle the traffic. At present technology, this might be an unreasonable demand on local resources, money, and labor. This could cause a serious growth bottleneck.

However, if airport-building technologies improve at 20%/year over the same period, it will probably be cheaper and easier to build eight new airports than to maintain the existing one. There will be no bottleneck.

Historically, there has never been a (long-term) bottleneck.

[Ziggurat]

Originally posted by new drkitten
I believe current scientific thinking is that your school teacher was wrong (conservation of angular momentum or something -- 99% of the mass of the solar system is in the Sun, but 99% of the angular momentum is in the planets, and that Just Ain't Right.)


This claim (which is sometimes used by creationists) suffers from two fundamental problems, the assumption that there is no mechanism for the sun to lose angular momentum, and the assumption that angular momentum should be evenly distributed to begin with. Both assumptions are wrong, and the reason is magnetic fields.

The sun is constantly ejecting energetic plasma that goes flying off into deep space. As the sun rotates, the magnetic field it creates also rotates, and the charged plasma gets pulled around with this magnetic field. As the plasma moves away from the sun, the magnetic field tries to keep it rotating at the same angular velocity, but as it increases its distance from the sun, this requires that it increase its angular momentum. At large enough distances, the solar wind breaks free from the sun's magnetic field, but in the mean time there's still been a transfer of angular momentum from the sun to the solar wind.

Going back farther to the initial development of the solar system, the effects of magnetic fields on the dynamics of nebular gases can be quite significant, and provide a mechanism for the transfer of angular momentum from one part of the nebula to another. If some of that gas loses angular momentum to other parts of the nebula, that's the gas that's going to collapse into the center first. So we can't assume that there was anything approaching even distribution to begin with.

[Drooper]

Originally posted by new drkitten
Rate of growth measured as a percentage will be constant, but not in absolute terms. In absolute terms, exponential growth does in fact mean a continuously increasing rate of growth.

For example, I put a thousand monetary units of your choice in an investment paying ten percent a year compound interest. The rate of growth will be exponential. Today I have a thousand units. Next year I'll have 1100. In seven years, I'll have about 2000. In twenty years, I'll have about 8000.

But if you look at how much money I'm making, I'm getting an income of 100/year next year, but I will be getting an income of 800 units/year in 2025. So in absolute terms, my "rate of growth" has increased eight-fold in twenty years.

In discussing resources and so forth, it's usually the absolute terms that are of interest. For example, the local airport may have 100 flights a day (and be growing at 10%/year). Twenty years from now, it will be taking in 800 flights/day -- which means that someone somewhere will have needed to essentially build seven complete new airports to handle the traffic. At present technology, this might be an unreasonable demand on local resources, money, and labor. This could cause a serious growth bottleneck.

However, if airport-building technologies improve at 20%/year over the same period, it will probably be cheaper and easier to build eight new airports than to maintain the existing one. There will be no bottleneck.

Historically, there has never been a (long-term) bottleneck.

Thanks for the remedial maths.

A rate of change is a first derivative, or percentage change in my book.


With regard to your example, you have fallen for the classic Malthusian misconeption. An easy mistake to make.


This is all extremely well trodden territory in economic literature. "Limits to Growth" do not exist (or not in the form that you seem to believe).

[Drooper]

Originally posted by new drkitten
Rate of growth measured as a percentage will be constant, but not in absolute terms. In absolute terms, exponential growth does in fact mean a continuously increasing rate of growth.

For example, I put a thousand monetary units of your choice in an investment paying ten percent a year compound interest. The rate of growth will be exponential. Today I have a thousand units. Next year I'll have 1100. In seven years, I'll have about 2000. In twenty years, I'll have about 8000.

But if you look at how much money I'm making, I'm getting an income of 100/year next year, but I will be getting an income of 800 units/year in 2025. So in absolute terms, my "rate of growth" has increased eight-fold in twenty years.

In discussing resources and so forth, it's usually the absolute terms that are of interest. For example, the local airport may have 100 flights a day (and be growing at 10%/year). Twenty years from now, it will be taking in 800 flights/day -- which means that someone somewhere will have needed to essentially build seven complete new airports to handle the traffic. At present technology, this might be an unreasonable demand on local resources, money, and labor. This could cause a serious growth bottleneck.

However, if airport-building technologies improve at 20%/year over the same period, it will probably be cheaper and easier to build eight new airports than to maintain the existing one. There will be no bottleneck.

Historically, there has never been a (long-term) bottleneck.

Thanks for the remedial maths.

A rate of change is a first derivative, or percentage change in my book.


With regard to your example, you have fallen for the classic Malthusian misconeption. An easy mistake to make.

In economics (which is the area of discussion here) it is all about rates of growth. Economies tend to gravitate to dynamic, rather than static, steady states for specific reasons.


This is all extremely well trodden territory in economic literature. "Limits to Growth" do not exist (or at last in the form implied in this thread).

[AWPrime]

Originally posted by Drooper

Technological progress is the underlying driver of economic growth (adjusted for available labour and capital). It has been remarkable stable over time across countries around the world.

That is an illusion. Because it has been more or less like this in the last 100 years, doesn't mean that it always use to be so.

If you look in the distant past you would see that my scenario fits reality better.


In the end the most realistic scenario would be up and down periods of technological advancement. With a general upward trend so long as nothing truly apocalyptic happens.

[drkitten]

Originally posted by Drooper
Thanks for the remedial maths.

I'm sorry you need it.



A rate of change is a first derivative, or percentage change in my book.


Then you need a new book. For example, "velocity" (or more generally "speed") is often defined as the "rate of change" of position (or as the first derivative of position), but the usual units by which velocity is measured are absolute -- kilometers per hours, meters per second, light seconds per century. Look at the problems given here (http://home.alltel.net/okrebs/page205.html) Or, for that matter, type "first derivative" and "rate of change" into your favorite web search engine and see if the word "percentage" appears anywhere.

A ten-percent increase in my speed if I'm moving at 10m/sec would be a 1m/sec increase. A ten-percent increase if I'm moving at 100m/sec would be 10 m/sec. At least in my book, 10m/sec is clearly greater than 1m/sec (by 9 m/sec).



With regard to your example, you have fallen for the classic Malthusian misconeption. An easy mistake to make.


I have not. Re-read what I wrote.

[AWPrime]

Originally posted by new drkitten
Already done.

Still not seeing anything.

Maybe becasue you wishfull thinking and you argument of 'because I say so', isn't worth much.

[drkitten]

Originally posted by AWPrime
That is an illusion. Because it has been more or less like this in the last 100 years, doesn't mean that it always use to be so.

If you look in the distant past you would see that my scenario fits reality better.

What scenario? Large-scale economic growth has really only existed in the past hundred years or so. In the more distant past, periods of relatively stagnant economic growth were also marked by relatively stagnant technological growth (and vice versa). The most marked improvement in the economic situation in the past two thousand years or so started in the Rennaissance, when technology also started to improve. Prior to that Europe had a thousand year dark age when both technology and economics were largely at a standstill, after the fall of Rome, which resulted in a net loss overall, which in turn followed the Golden Age which was a time of both technical progress and economic improvement.

How far back do you want me to go?

[drkitten]

Originally posted by AWPrime
Still not seeing anything.

Maybe becasue you wishfull thinking and you argument of 'because I say so', isn't worth much.

You're right. "Because I say so" isn't worth much.

But "Because Dr. Simon says so, and demonstrated it conclusively it in an experiment jointly with Dr. Ehrlich" is worth substantially more.

Can you name a single instance in the past two thousand years of history where a resource has run out without technology providing an acceptable substitute?

[AWPrime]

Sure the Viking colony in Greenland.

Fertile land eroded away and they starved. I guess technology didn't save them.

[AWPrime]

Originally posted by new drkitten
Large-scale economic growth has really only existed in the past hundred years or so. In the more distant past, periods of relatively stagnant economic growth were also marked by relatively stagnant technological growth (and vice versa). The most marked improvement in the economic situation in the past two thousand years or so started in the Rennaissance, when technology also started to improve. Prior to that Europe had a thousand year dark age when both technology and economics were largely at a standstill, after the fall of Rome, which resulted in a net loss overall, which in turn followed the Golden Age which was a time of both technical progress and economic improvement.

Now can you imagen that I object to your claim: that the trend of technological advancement will continue forever, if it has only been around for such a short periode?

[CBL4]

This is one of the sillier threads I have read.

First of all, what does "exponential growth in technology" mean? Do we have a measure a measure of growth I am unfamilar with? How many technos is the invention of the airplane worth? Is this more or less technos than the invention of the microprocessor?

As far as limited resources, why are we assuming the sun is some limitation? We can we get energy from Alpha Centauri when we get there?

The biggest resource needed for technology growth is the human brain. I could make the argument that this resource is going to top out somewhere around 10 billion. This will be a limitting factor in technological advances - at lease until we colonize Mars and Alpha Centauri III. Then brain growth will increase again.

CBL

[pauldmin]

Exponential growth in technology, population, whatever will simply not happen unless resources are infinite.

Things like population follow a 'logistic' curve. i.e. An initial exponential growth is following by a slowing down, then an approach to a fixed level. ( Either approaching it gradually or oscillating )

This suggests that technology will have a limit where no further progress can be made without some fundamental shift in the conditions. Even if some new breakthrough pushes the boundary further, a new steady state limit will occur sooner or later.

e.g. Moores Law has still remained true, but we are now reaching the limit of silicon's capability and without any new form of technology, the law will start to follow the logistic model.

[Jyera]

Originally posted by AWPrime
From: http://forums.randi.org/showthread.php?s=&threadid=55386
quote:
--------------------------------------------------------------------------------
Originally posted by Kevin_Lowe
Technology advances exponentially, so it will be a few centuries at most before the rate of change approaches infinity and we hit whatever physical limits the universe will impose on us.
--------------------------------------------------------------------------------
That only applies when population and available energy is exponentially as well.

And currect trends are showing that this might not be the case.
Firstly, we need to know the unit of measure.
Unless a unit of measure is given to "Technology" (what ever it means here), it cannot be measured. It is then meaningless to talk about any measure of advance. We cannot even show that it advanced linearly.

Secondly, there is more than one possible "unit of measure".
Simply changing the unit of measure changes the relevance and perspective. If the unit of measure of time is aeon, the technological advancement would be very slow and flat.
(A unit of duration equal to 1E9 YEAR. )

[AWPrime]

To bad we can't define technical knowledge in bites.