Post-Scarcity Economics

Discuss life, the universe, and everything with other members of this site. Get to know your fellow polywell enthusiasts.

Moderators: tonybarry, MSimon

Luzr
Posts: 269
Joined: Sun Nov 22, 2009 8:23 pm

Post by Luzr »

TallDave wrote:
In fact, by "self-emerge" I mean that the AI will not be rule based, or "programmed in", it will be based on some sort of self-programming (just like neural networks are).
We're arguably already there. Most of the work in chip design is done by computers now
As far as I undestand process, not really. Chip design is like creating software, using human designed library building blocks, computer does simulation and silicon layout (sort of equivalent of programming language interpreter and compiler). Surely, it saves a tremendous amount of time, but saying that computer "does the chip design" is far fetched.
, and lots of programs self-correct.
Indeed. Well, I work as systems programmer, maybe I have missed something?

Maybe you are confusing "self-correct" with "are programmed to gracefully handle bugs in the code".

And of course, programming tools enchance productivity. But they are still just tools. If you look under the skin, what they do is rather primitive.

TallDave
Posts: 3141
Joined: Wed Jul 25, 2007 7:12 pm
Contact:

Post by TallDave »

Surely, it saves a tremendous amount of time, but saying that computer "does the chip design" is far fetched.
Well, sure, just as it's farfetched to say a backhoe digs a hole. OTOH, the machine is doing 99% of the actual work, and the proportion is ever-increasing.
Maybe you are confusing "self-correct" with "are programmed to gracefully handle bugs in the code".
Some more sophisticated software in specialized apps rewrites its own code in response to various situations (I even see code-building apps in ERP software). On a more mundane level, your antivirus software actively removes code from your computer when it meets certain conditions.

And at this very moment, I am writing a 10 line program which in turn builds me a 1,000 line program using input from a couple different files. It's not really an exotic programming technique, but the computer is again doing most of the work.

Yes, these are all just tools, but the tools are gradually getting smarter and more self-adapting.

TallDave
Posts: 3141
Joined: Wed Jul 25, 2007 7:12 pm
Contact:

Post by TallDave »


bcglorf
Posts: 436
Joined: Mon Jul 23, 2007 2:58 pm

Work is different than thinking

Post by bcglorf »

And at this very moment, I am writing a 10 line program which in turn builds me a 1,000 line program using input from a couple different files. It's not really an exotic programming technique, but the computer is again doing most of the work.

The same could be said of running a hello world program written in C through a compiler and on down into machine code. The computer may be doing most of the work, but the thinking is still coming from humans.

Luzr
Posts: 269
Joined: Sun Nov 22, 2009 8:23 pm

Post by Luzr »

TallDave wrote:
Surely, it saves a tremendous amount of time, but saying that computer "does the chip design" is far fetched.
Well, sure, just as it's farfetched to say a backhoe digs a hole. OTOH, the machine is doing 99% of the actual work, and the proportion is ever-increasing.
See, I do not dispute that computers are useful. But to say that silicon compilers represents strong AI is really far-fetched.
And at this very moment, I am writing a 10 line program which in turn builds me a 1,000 line program using input from a couple different files. It's not really an exotic programming technique, but the computer is again doing most of the work.
YOU are writing.

Luzr
Posts: 269
Joined: Sun Nov 22, 2009 8:23 pm

Post by Luzr »

I believe you are missing the main point. I am too very happy with accelerating technological progress, with the fact that machines do more and more work. I would even say that our inteligence is now augmented by computers - in a symbiosis, the total intelligence is much higher. Add to the mix networking (a mere fact we can discuss topics here) a search engines and you can see that current civilisation as whole is now a sort of powerful godlike supersentient organism. I like that.

Anyway, the original topic was the possibility of strong AI and I think that while related, it is still something else...

djolds1
Posts: 1296
Joined: Fri Jul 13, 2007 8:03 am

Post by djolds1 »

These types of predictions remind me of the 1939 World's Fair and Fritz Lang's Metropolis. Note how 20 years ago everyone thought there were 100,000+ genes in the human genome and every conceivable trait could be altered by easily tweaking a few genes. Then Venter trumps the Human Genome Project, we find out that there are maybe 25,000 human genes, and thus detail tweaking of minute traits is much harder involving complex protein folding interactions that derive from the genes. Oops. All futurist predictions beyond a five year horizon should be taken with a VERY large salt lick.

Now, I've given some additional thought to actual post scarcity economics. Any economic endeavor involving small to medium scale durable or consumable goods does NOT look to be in good shape. As a rule of thumb, say anything up to 150 meter luxury yachts. Larger items (infrastructure, aircraft carriers, etc.) would probably have some intrinsic value, but would be the possessions of dedicated groups or polities.

My cite of Paul Romer's work was in error. He assumes most (but not all) technological innovation comes from a profit motive. An individual or firm R&D's a new technology. Design and development info on the tech (the template and archive) is openly shared, but the legal right to exploit and profit from the new innovation is held as a monopoly - patent law IOW. But with post scarcity manufacturing the urge to cheat is intense, and enforcing the patent's temporary or perpetual monopoly right over the physical item derived from the template is near impossible. Even if the First & Second Worlds were to enforce patent law like the US does its current war on drugs, the Third World would jump patents on all consumer goods, medicines, etc. like there was no tomorrow. Within a few decades the effort to keep patent rights going collapses of its own weight, much like the movie & music wars on P2P downloaders are collapsing.

So the templates are easily and rapidly dispersed to a wide audience, and no one can derive a realistic profit from the sale of the physical durable goods derived from those templates. How do you adapt?

The music & movie industries can adapt as I outlined previously. They substitute one type of human capital (access to concerts, back stage passes) for the older human capital (work on creating movies & songs), and then use those music & songs as loss leaders for the new profit engine. But fabbers of increasing capability make it pointless to substitute one form of durable good for another; all durable goods (including the most advanced electronic components) are now economic fungibles. Could the physical goods be used as loss leaders for some sort of social or human capital reward, as used by the (redefined) IP industry? Perhaps, but that would seem to undermine the motive to develop new science & technologies. There are easier ways to access a profit = popularity economy than 10+ years of esoteric education.

Bottom line, I think post scarcity manufacturing has a real possibility of radically SLOWING the rate of innovation once its achieved. It will nuke the profit motive for new advancements in physical technological durables - bluetooth, iPod, etc. We will still see dedicated academics and interested amateurs like the Royal Society of the 17th-19th centuries, or the Open Source movement of today, but the drive to newer and newer technologies is swamped as the only remaining profit sector is interpersonal.
Vae Victis

Luzr
Posts: 269
Joined: Sun Nov 22, 2009 8:23 pm

Post by Luzr »

djolds1 wrote: Bottom line, I think post scarcity manufacturing has a real possibility of radically SLOWING the rate of innovation once its achieved. It will nuke the profit motive for new advancements in physical technological durables - bluetooth, iPod, etc. We will still see dedicated academics and interested amateurs like the Royal Society of the 17th-19th centuries, or the Open Source movement of today, but the drive to newer and newer technologies is swamped as the only remaining profit sector is interpersonal.
I believe that the real problem there is how to make binary data free (because they are easily stolen) while still getting innovators rich. It simply requires a new system of capital redistribution.

What about this: There is a special "internet access tax". It gets collected in global fund. (Alternatively, this fund is created from different resources).

Anybody connected has a right to "vote" for or "send kudos" related to object (software, movie, fabber plan) freely available on internet. Global fund then simply redistributes collected tax based on on these votes.

Please fill in details like limited number of votes or maybe rather possibility to cast only singe vote per producer per year or something like that.

I believe there are little possibilities to really cheat in such system, but I might be wrong. In any case, it is the system that keeps stuff "free" but still makes innovators rich.

And the best part: Any country can join this system or decide not to. If not, digital producers in that country will be cut off from the system...

Betruger
Posts: 2321
Joined: Tue May 06, 2008 11:54 am

Post by Betruger »

That sounds a bit like the Whuffie scheme as in "Down & out in the magic kingdom" by Doctorow.

taniwha
Posts: 102
Joined: Thu Oct 29, 2009 9:51 am

Post by taniwha »

IntLibber wrote:You need to do a lot more than that. Comet water will be salty, dirty, with lots of other chemicals in it. You need to distill all that out of there.
Considering power will probably be the least of a ship's (or colony's) concerns, distilling water won't be a problem. Cooling it might. Getting it impure enough to drink might also be an issue.

No, water won't be a problem in space. However, the other chemicals humans need might be, though that might be solvable with creative algae and bacteria farms. Many of those chemicals might be mixed in with that comet water.

TallDave
Posts: 3141
Joined: Wed Jul 25, 2007 7:12 pm
Contact:

Post by TallDave »

The same could be said of running a hello world program written in C through a compiler and on down into machine code. The computer may be doing most of the work, but the thinking is still coming from humans.
Not quite the same thing. My little program writes new code, which does something completely new. It's the difference between executing instructions and writing new instructions. The latter sometimes gets treated as some strange and mysterious ability, but it's actually fairly common.
See, I do not dispute that computers are useful. But to say that silicon compilers represents strong AI is really far-fetched.
Sure, I wouldn't try to argue that, just that AI already does many of the things necessary for strong AI, perhaps more than is commonly appreciated.

AI is already doing perhaps 99% of the work. The thinking tasks left to humans are those that require generalized parallel processing sorting through massive amounts of data, because AI is not very good at that.

I think we can all agree a program that can build a new and better generation of processors from just the command "Design a new and better generation of processors" would be strong AI. It might even be possible to build such a program today, but the programming/data-gathering effort would be prohibitive, and it might take 1000 years to finish, so there's no reason for a free market to produce such a thing.
I believe you are missing the main point.
I didn't have a point with the link, just thought it was interesting.
Last edited by TallDave on Thu Dec 17, 2009 2:11 pm, edited 4 times in total.

TallDave
Posts: 3141
Joined: Wed Jul 25, 2007 7:12 pm
Contact:

Post by TallDave »

Even if the First & Second Worlds were to enforce patent law like the US does its current war on drugs, the Third World would jump patents on all consumer goods, medicines, etc. like there was no tomorrow. Within a few decades the effort to keep patent rights going collapses of its own weight, much like the movie & music wars on P2P downloaders are collapsing.


I think you just try to protect IP as best you can, knowing you can't do it perfectly. The amount of R&D you can drive is dependent on your confinement time. People are still spending $500M to make one movie.

As poor countries become rich, they will become better about IP. DeSoto's Mystery of Capital makes the point that wealth is extremely dependent on property rights.

Luzr
Posts: 269
Joined: Sun Nov 22, 2009 8:23 pm

Post by Luzr »

TallDave wrote:
The same could be said of running a hello world program written in C through a compiler and on down into machine code. The computer may be doing most of the work, but the thinking is still coming from humans.
Not quite the same thing. My little program writes new code, which does something completely new. It's the difference between executing instructions and writing new instructions.
Interpreter is executing instructions.

Compiler takes your code and creates its functional equivalent in assembler - quite often it even changes the algorithm during optimization process. I would say there is hardly any semantic difference between what you did and what compiler does. There is some code on input and some different code on output.
AI is already doing perhaps 99% of the work. The thinking tasks left to humans are those that require generalized parallel processing sorting through massive amounts of data, because AI is not very good at that.
Because it is not strong AI :) That is one difference.
I think we can all agree a program that can build a new and better generation of processors from just the command "Design a new and better generation of processors" would be strong AI.
I would say this would fail because "better" is very general term...

Anyway, strong AI should understand that...

bcglorf
Posts: 436
Joined: Mon Jul 23, 2007 2:58 pm

I'd go further

Post by bcglorf »

The thinking tasks left to humans are those that require generalized parallel processing sorting through massive amounts of data, because AI is not very good at that.
I'd go further and say that current AI is terrible at that. What is more, humans don't only deal with processing huge amounts of data in parallel, but they deal with handling incomplete data and fuzzy goals. Today's AI can barely even pretend to handle that kind of processing.
Compare simple games requiring very small sets of data. Chess is a game of complete information, where the computer needn't guess at anything, it can approach the game like a math problem and find the solution. Chess AI's haven't been getting 'smarter', they've just been solving the same problem by throwing more and more memory and cpu cycles at it. Poker on the other hand doesn't have complete information, leaving no 'right' solution. You find that the level of AI's are pretty much on par with or below the skill of the human that wrote the code initially.

djolds1
Posts: 1296
Joined: Fri Jul 13, 2007 8:03 am

Post by djolds1 »

Luzr wrote:I believe that the real problem there is how to make binary data free (because they are easily stolen) while still getting innovators rich. It simply requires a new system of capital redistribution.

What about this: There is a special "internet access tax". It gets collected in global fund. (Alternatively, this fund is created from different resources).
Nope. Fabbers can make wireless access/retransmission nodes. Regulating communications becomes much harder.
Luzr wrote:Anybody connected has a right to "vote" for or "send kudos" related to object (software, movie, fabber plan) freely available on internet. Global fund then simply redistributes collected tax based on on these votes.
Some sort of prize system is doable. But spending currency on consumer items becomes pointless, unless they're limited run/"artistic" items that can demand a premium for rarity. Ford disappears but Ferrari does not, etc.
taniwha wrote:No, water won't be a problem in space. However, the other chemicals humans need might be, though that might be solvable with creative algae and bacteria farms. Many of those chemicals might be mixed in with that comet water.
Nitrogen for breathing gas dillutant, for one. As well as phosphorous, sulfur and other trace elements. Forget comets and focus on C-type asteroids. Plenty of water and other volatiles, as well as metals and so on...
TallDave wrote:
djolds1 wrote: Even if the First & Second Worlds were to enforce patent law like the US does its current war on drugs, the Third World would jump patents on all consumer goods, medicines, etc. like there was no tomorrow. Within a few decades the effort to keep patent rights going collapses of its own weight, much like the movie & music wars on P2P downloaders are collapsing.
I think you just try to protect IP as best you can, knowing you can't do it perfectly. The amount of R&D you can drive is dependent on your confinement time. People are still spending $500M to make one movie.
As IT continues to improve the confinement time must commensurately lessen.
TallDave wrote:As poor countries become rich, they will become better about IP. DeSoto's Mystery of Capital makes the point that wealth is extremely dependent on property rights.
Also technology adoption.

https://netfiles.uiuc.edu/parente/The%2 ... Growth.pdf

Beyond that, I take the long view, and in the 50-100 year horizon attempts to enforce IP regimes must inevitably collapse given the potentials inherent in ubiquitous computing, communications, and fabrication. Indeed, if fabbing weapons is easy (possibly up to and including WMD) then the basis of the state and civilization as we know it is threatened; there would not be security in mass numbers.

The danger I see is an extreme slowing of innovation by 2050 or so. Devices able to deliver new functions simply will not be able to make a profit if that profit depends on them being physical items. Using the physical items as loss leaders for access to status is possible, but does not seem compelling. When fabbers are realized, the things that are obviously in the pipeline (even with 20-30 years development time) will be realized, but beyond that things could very well slow down.

And note that the elites trying to enforce your IP system above would not necessarily dislike that. Dynamic societies are ongoing threats to the status of elites. Static status quo societies are much more predictable, meaning the elites can stay on top much longer.
Vae Victis

Post Reply