Well I don't need to bone up on the invention of negative numbers etc. since I've been aware of the arabic origin of that particularly valuable invention for sometime now.
However, you have to admit there were a number of centuries lasting up through the 20th century that found Islamic nations lagging for some reason.
Perhaps it wasn't the religion but it certainly does more than reflection than resting on relatively ancient accomplishments.
PS: Congratulations on your recent accomplishments and in particular the funding of the X-Prize!
You mean their position of being hostile to technology while at the same time developing nuclear programs (plants/bombs, who knows)?
Yeah and Iraq had weapons of mass destruction ready to be deployed against the US within weeks of the US's invasion of Iraq and the US provides so much support to Israel due to Israeli nukes being deployed in all major US cities except New York, LA, Miami and maybe Chicago.
I find Iran's position here interesting due to the fact that Islam has a reputation for being hostile to technology -- and you must admit that, regardless of the source of that reputation, it does have such a reputation at least among the US if not among all Western countries.
And I'm not hostile to Islam -- I once shouted "Allahu akbar!" during rush hour at the intersection of Lawrence and Homestead. (I admit, mainly because I thought it was a subversive act.)
It might make sense for past winners to host future competitions so that could explain the coincidence of Shanghai being both host and winner of the competition.
After clicking around for a few minutes, it stil l was not apparent where one could find the statement of the problems to be solved by the contestants.
The host of the competition just happened to be the top winner.
So well let's assume this is a fair test of programming skill, why is it that an Islamic state's team, Sharif University of Technology, beat out not only the top technical university of India (IIT) but all of the US's Ivy League schools -- not just MIT and CalTech?
Ockham's Razor as a criterion for intelligence
on
The Baby Bootstrap?
·
· Score: 1
I agree that comprehension has some relation to ability to compress, at least in that knowledge of the rules of a language improves ability to compress. But that's not the only factor (even beyond the fact that knowing the rules != comprehension).
Prediction is far more more than coming up with language rules.
For example, if I see some text attributed to "anthony_dipierro" and the subject is the relationship between AI and compression, I can predict that the text is more likely to say "compression isn't the measure for AI" than "compression is the measure for AI".
This goes far beyond simple bias estimates in speech act content as well.
As a part of the system, pruning based on the Minimum description length principle was developed that can handle also continuous variables. It turned out that MDL pruning helps to build more
comprehensible [emphasis JAB] models, while at the same time preserves model's performance in terms of its prediction power.
There is a school of thought in the philosophy of science that this is in fact the precise way of measuring the validity of a body of theory.
If you don't like Ockham, then how about Einstein summing up his focus on invariance thus:
Everything
should be made as simple as possible, but not simpler.
We _are_ talking about information content.
on
The Baby Bootstrap?
·
· Score: 1
Anyway, I think the 1.3 bits per character mark would be reached in a matter of weeks, given an input as large as the Library of Congress, stripped down to 27 characters, and assuming the payout was at least a few thousand dollars.
Maybe I'm way off, but I hope someone takes your suggestion, because I'd love to give it a try. I'd almost be willing to do it just to prove you wrong, but it'd take a lot more than a few weeks without any monetary incentive.
First off, "1.3" doesn't appear anywhere in the prize criterion, but proving Matthew Mahoney's figure of 1.3 bits per character as the threshold for AI "wrong" would be valuable for the simple reason that it would be a demonstration of the best current natural language model. (Mahoney's reference to Shannon, BTW, does give a range of.6 up to 1.3 bits per character for human intelligence so if you're convinced Mahoney was too lax in specifying Shannon's upper bounds then that's not saying much about the range provided by Shannon let alone the overall conception of information content as intelligence.)
Secondly, the M-Prize-like structure specified for the K-Prize is adequate to incentivize arbitrarily high degrees of intelligence and avoids the issue which threshold is the "right" one.
In any case when we're talking about irreducably compressing a natural language corpus as the criterion for AI we _are_ talking about information content.
As I said, obsessing about a particular threshold of 1.3 or even.6 isn't the point. The point isn't even that at _some_ threshold you have a critter that is indistinguishable from a literate human in its ability to comprehend a large body of literature.
The point is that at some point you have a critter that is able to comprehend a body of literature more accurately than anyone. That's when you might start getting a real alternative to existing "authorities".
What advantages would orbital settlements have
over a colony built on another planet?
Access to 24-hour-a-day sunlight. This makes solar power a consistent,
economical energy source. Photovoltaic panels can convert sunlight into
electrical current, and solar mirrors can concentrate it for process heat
in industrial operations (such as the smelting of ore). A space-based solar
concentrator the size of a football field (which could still weigh less than
a car) could provide process heat equivalent to the burning of 1 million
barrels of oil over 30 years.
Sunlight also drives the life-support system of the habitat, so the day/night
cycle can be set to whatever is convenient. Compare this to the moon, where
there is 14 days of continuous daylight, and then a 14-day-long night. Here,
some alternate energy source would probably have to be used half the time.
Access to zero gravity. This may have a number of industrial and
entertainment possibilities. Structures (such as the above-mentioned solar
mirrors) could be built many times larger and flimsier in space than on a
planet.
Zero G would be a liability if there were no alternative to it. Astronauts
experience loss of bone mass and muscle tone after prolonged exposure to
weightlessness. But most of a space habitat would be under Earth-normal gravity,
although there would be easy access to regions of reduced gravity and zero
G (perhaps for personal flight). With planets, on the other hand, you have
to take the gravity that's there, and it's often the wrong kind of gravity
to keep us healthy. Lunarians or Martians would probably not be able to visit
the Earth (nor accelerate at 1 G).
Location near the top of Earth's gravity well. We here on Earth are
the "gravitationally disadvantaged". We are at the bottom of a pit 6,400
km (4,000 miles) deep. This is what makes space launches from the surface
so difficult and expensive. Settlers near the top of the gravity well would
be ideally situated for departures to points beyond.
Control of the environment. The weather and other aspects of the
surroundings would be those of the inhabitants' choosing. Agriculture in
space will benefit from weather control (fresh fruits and vegetables year-round!)
and the absence of pests.
Mobile territories. Although the first generation of space habitats
will doubtless reside in High Earth Orbit, there's no reason why space settlers
couldn't attach engines to their habitats, and over the course of months
or years gradually change their orbit to whatever solar system location they
found preferable.
Long-term expansion of the land area available to the human race. Let's
be optimistic and assume that Mars could be made totally Earth-like in the
near-term. This would basically double the land area available to humanity,
meaning problem solved...until the population doubles again. Right now, that
is happening roughly every 40 years. By contrast, if we were to conservatively
limit ourselves to using only the resources of the asteroid belt, we could
build, in the form of space habitats, 3,000 times the livable surface area
of the Earth. This makes space settlement a long-term solution.
Think about it more like compressing the entire content of the library of Congress.
Since you're going to have to produce a bunch of dictionaries anyway, you may as well use them to compress the rest of the library.
That's more the way to think about Kolmogorov compression's relationship to the artificial intelligence criterion.
Don't get hung up on Shannon's figure. You'll just end up chasing your tail.
Most information theorists tehse days recognize that algorithmic complexity using Kolmogorov complexity is the ideal way to define information content.
That's why I use Kolmogorov compression instead.
on
The Baby Bootstrap?
·
· Score: 1
Ultimately, if you were free to do things the way Shannon did things -- and all you had to do was produce a major corpus without counting the dictionaries, etc. as part of the compressed corpus, then you simply have a bit, the value of which could be 1 or 0, where, say "1" means "the corpus" and 0 means "not the corpus" and you hand that bit to the "interpreter" which is not counted as part of the compressed representation.
You can achieve arbitrarily high "compression" ratios through such a fallacious system.
Kolmogorov compression ratios have to include all dictionaries, algorithms, etc. that you use to produce the uncompressed corpus. That's why I specified things the way I did for the K-prize and why I called it the "K" prize.
3. Human vs. Machine Prediction
We now compare text prediction in humans and
machines, with the reassuring result that humans are
superior. Shannon bounded the entropy of written English
by having people solve character-completion puzzles
(Shannon 1951). Text samples were taken from a variety
of sources, such as classic literature or technical manuals.
The text was reduced to a 27 character alphabet of
monocase letters and spaces. Subjects guessed at each
letter until correct, and were allowed to use references
such as dictionaries and character frequency tables.
Shannon estimated that the entropy or uncertainty of
written English is between 0.6 and 1.3 bits per character.
Larry Page Should Seed the K-Prize
on
The Baby Bootstrap?
·
· Score: 2, Interesting
Since Larry Page is on the X-Prize Board of Trustees, and since Google is pushing the envelope of what is needed to index and compress the entire content of the Internet, Page should consider providing seed funds and then matching funds for any donations to a compression prize with the following criterion:
Let anyone submit a program that produces, with no inputs, one of the major natural language corpuses as output.
S = size of uncompressed corpus
P = size of program outputting the uncompressed corpus
R = S/P ... or the Kolmogorov-like compression ratio.
Previous record ratio: R0
New record ratio: R1=R0+X
Fund contains: $Z at noon GMT on day of new record
Winner receives: $Z * (X/(R0+X))
Compression program and decompression program are made open source.
If Larry has any questions about the wisdom of this prize he should talk to Craig Nevill-Manning.
"The Turing test for artificial intelligence is widely accepted, but is subjective, qualitative, non-repeatable, and difficult to implement. An alternative test without these drawbacks is to insert a machine's language model into a predictive encoder and compress a corpus of natural language text. A ratio of 1.3 bits per character or less indicates that the machine has AI."
This "K-Prize" will bootstrap AI.
OK, so he can christen it the "Page K-Prize" if he wants.
There is no particular reason you have to put structure in the address in order to have hierarchical routing.
You can run with the equivalent of name servers that tell you the current best (QOS) routes given a particular system address.
The big failure of the DARPA guys was to presume that the hardware capabilities of the routers they were running with during the early days of IP was not going to scale along with the network. It did but by then we were stuck with their god-damned network class structures.
My experience of DARPA's funding of "CS" is putting all 64 bit alternatives to the 32-bit (segmented) IP address plus 16-bit port out of business.
Aside from Xerox's 64 bit MAC address which was shelved as the basis for IP addresses, there was another standard promoted by a group of companies from Apple to Atari to Western Electric/Bell Labs to Packet Cable to Knight-Ridder circa 1982 which consisted of an unsegmented system identifier and object identifier combined in an 64 bit address -- the SID growing from the LSB up and the OID growing from the MSB down.
It would have been a very different and far superior world if they had not been stopped by DARPA's idiots.
Would you want to add an additional condition that all awardees would have to release the source code to their work
There are actually two pieces to this question:
Compression.
Decompression.
Kolmogorov compression is defined in terms of the minimal bit string (usually as program running on a Turing-complete machine) it takes to describe another bit string. In order verify a contestant has won some increment of the K-Prize, it would be necessary to make his minimal bit string public so that anyone could "execute" the description to produce the original corpus.
This would produce the valuable result that the language model he is using would be published. The language model itself is very valuable since it contains the model, not just of the language, but of the conceptual world represented by the corpus. This model would be very valuable -- particularly as it began to exceed normal human capabilities to compress the corpus -- for it would contain many novel and quantifiably valuable philosophical, scientific and practical concepts.
The program used to compress the corpus might be very valuable as well or it might be just the "dictionary" of concepts used to compress the corpus. An example of the former is given by some work with advanced inductive logic programming systems that invent predicates. An example of the latter is given by the approach Cycorp has taken for 2 decades (without much success yet) where they hire a bunch of philosophers to describe, in a variety of "microtheories", various aspects of "common sense".
Lately Cycorp has, very wisely, given up the ghost with the idea of having really smart humans try to produce all the concepts and has started using some predicate invention logic.
So very soon now I predict there will be some very profound breakthroughs -- if for no other reason than the long dark ages of AI, where high priests are relied upon for all wisdom, are terminally ill and recognized as such by the high-priests of the "church" of AI.
For that reason I think it wise to demand that the compression algorithm be made public.
Think about the Rosetta Stone in terms of compression. You have a single message represented in 3 different languages. Now imagine a huge Rosetta Library -- the same library represented in 3 different languages. Clearly, optimal compression of this Rosetta Library is going entail precisely the "conceptual processing" you desire.
The most important practical application is information quality. Information quality is a profound advance over simple data quality. Basically, information quality is about the quality of ideas and abstractions used in descriptions of phenomena or data. Think Ockham's Razor writ-large.
This penetrates all aspects of knowledge and society.
Quality information helps us simplify our world view without making it less accurate -- or conversely -- make our world view more accurate without making it more complicated.
"Publish or perish" produces precisely the explosion of words that creates the lack of understanding between people/disciplines. Academia just doesn't have the right incentives.
The real problem with intellectual property is that no one but the acquisitive can afford the lawyers that it takes to defend the royalty stream that should arise from it -- so we're beig inundated by pseudo-"inventors" who are really just tax collectors --thus destroying technological civilization's foundation.
You're right that in the case of the X-Prize criterion, it was set up to favor simply throwing money at a reasonable technician -- but if the altitude goal had been set to 200km instead of 100km, John Carmack's team would likely have beat Paul Allen's team and done so on about 10% of Allen's investment.
Primitive LUIs exist today in interfaces like Google, but will become dramatically more powerful over the next few decades.
I am quite excited by the confluence of advances in prize awards for technology advancement, and advances with the theory of compression. I'm convinced that if a substantial prize award can be created for dramatic advances in natural language text compression, it will lead directly to a solution to the most critical aspect of the "AI problem" -- that being the problem of the explosion of words without concomitant understanding. I had high hopes for the Internet being the new Gutenberg press leading to a new enlightenment but I'm concerned that without dramatic advances in AI to correlate the huge corpus being generated, the benefits of the new enlightenment may be too long in coming to save us from ourselves.
"The Turing test for artificial intelligence is widely accepted, but is subjective, qualitative, non-repeatable, and difficult to implement. An alternative test without these drawbacks is to insert a machine's language model into a predictive encoder and compress a corpus of natural language text. A ratio of 1.3 bits per character or less indicates that the machine has AI."
A simple prize criterion would be for the first program producing a major natural language text corpus, with the size of the program being less than 1.3 bits per character of the produced corpus. Smaller intermediate prizes would help spur broader interest.
Computer assisted instruction has been rigorously shown to work in a wide array of educational settings. Similar claims can be made for television. The key is the content, not the medium.
Gangs operate as gangs because the individuals are weak on their own. They have a _lot_ of incentive to make sure they dominate any place integrity might show its face. When you have people of integrity they generally stand as individuals and simply have too much to deal with in a world that protects weak individuals.
Slashdot Mirror
The same thing happened with Enron and Worldcom without nearly so much outsourcing. They did however have a huge HUGE H-1b staff.
However, you have to admit there were a number of centuries lasting up through the 20th century that found Islamic nations lagging for some reason.
Perhaps it wasn't the religion but it certainly does more than reflection than resting on relatively ancient accomplishments.
PS: Congratulations on your recent accomplishments and in particular the funding of the X-Prize!
Yeah and Iraq had weapons of mass destruction ready to be deployed against the US within weeks of the US's invasion of Iraq and the US provides so much support to Israel due to Israeli nukes being deployed in all major US cities except New York, LA, Miami and maybe Chicago.
Know any other "antisemitic canards"?
And I'm not hostile to Islam -- I once shouted "Allahu akbar!" during rush hour at the intersection of Lawrence and Homestead. (I admit, mainly because I thought it was a subversive act.)
It might make sense for past winners to host future competitions so that could explain the coincidence of Shanghai being both host and winner of the competition.
Which ACM chapter are you talking about?
Prediction is far more more than coming up with language rules.
For example, if I see some text attributed to "anthony_dipierro" and the subject is the relationship between AI and compression, I can predict that the text is more likely to say "compression isn't the measure for AI" than "compression is the measure for AI".
This goes far beyond simple bias estimates in speech act content as well.
For example, inductive logic programming, itself based on Ockham's Razor, has been extended to use regression analysis to derive physical laws using minimum description length criterion:
There is a school of thought in the philosophy of science that this is in fact the precise way of measuring the validity of a body of theory.If you don't like Ockham, then how about Einstein summing up his focus on invariance thus:
Maybe I'm way off, but I hope someone takes your suggestion, because I'd love to give it a try. I'd almost be willing to do it just to prove you wrong, but it'd take a lot more than a few weeks without any monetary incentive.
First off, "1.3" doesn't appear anywhere in the prize criterion, but proving Matthew Mahoney's figure of 1.3 bits per character as the threshold for AI "wrong" would be valuable for the simple reason that it would be a demonstration of the best current natural language model. (Mahoney's reference to Shannon, BTW, does give a range of .6 up to 1.3 bits per character for human intelligence so if you're convinced Mahoney was too lax in specifying Shannon's upper bounds then that's not saying much about the range provided by Shannon let alone the overall conception of information content as intelligence.)
Secondly, the M-Prize-like structure specified for the K-Prize is adequate to incentivize arbitrarily high degrees of intelligence and avoids the issue which threshold is the "right" one.
In any case when we're talking about irreducably compressing a natural language corpus as the criterion for AI we _are_ talking about information content.
As I said, obsessing about a particular threshold of 1.3 or even .6 isn't the point. The point isn't even that at _some_ threshold you have a critter that is indistinguishable from a literate human in its ability to comprehend a large body of literature.
The point is that at some point you have a critter that is able to comprehend a body of literature more accurately than anyone. That's when you might start getting a real alternative to existing "authorities".
See Mike Combs' space settlement FAQ which says:
Sunlight also drives the life-support system of the habitat, so the day/night cycle can be set to whatever is convenient. Compare this to the moon, where there is 14 days of continuous daylight, and then a 14-day-long night. Here, some alternate energy source would probably have to be used half the time.
Zero G would be a liability if there were no alternative to it. Astronauts experience loss of bone mass and muscle tone after prolonged exposure to weightlessness. But most of a space habitat would be under Earth-normal gravity, although there would be easy access to regions of reduced gravity and zero G (perhaps for personal flight). With planets, on the other hand, you have to take the gravity that's there, and it's often the wrong kind of gravity to keep us healthy. Lunarians or Martians would probably not be able to visit the Earth (nor accelerate at 1 G).
Since you're going to have to produce a bunch of dictionaries anyway, you may as well use them to compress the rest of the library.
That's more the way to think about Kolmogorov compression's relationship to the artificial intelligence criterion.
Don't get hung up on Shannon's figure. You'll just end up chasing your tail.
Most information theorists tehse days recognize that algorithmic complexity using Kolmogorov complexity is the ideal way to define information content.
You can achieve arbitrarily high "compression" ratios through such a fallacious system.
Kolmogorov compression ratios have to include all dictionaries, algorithms, etc. that you use to produce the uncompressed corpus. That's why I specified things the way I did for the K-prize and why I called it the "K" prize.
Let anyone submit a program that produces, with no inputs, one of the major natural language corpuses as output.
S = size of uncompressed corpus
... or the Kolmogorov-like compression ratio.
P = size of program outputting the uncompressed corpus
R = S/P
Previous record ratio: R0
New record ratio: R1=R0+X
Fund contains: $Z at noon GMT on day of new record
Winner receives: $Z * (X/(R0+X))
Compression program and decompression program are made open source.
If Larry has any questions about the wisdom of this prize he should talk to Craig Nevill-Manning.
If, in the unlikely event, Craig Nevill-Manning has any questions about the wisdom of this prize, he should talk to Matthew Mahoney, author of "Text Compression as a Test for Artificial Intelligence"
"The Turing test for artificial intelligence is widely accepted, but is subjective, qualitative, non-repeatable, and difficult to implement. An alternative test without these drawbacks is to insert a machine's language model into a predictive encoder and compress a corpus of natural language text. A ratio of 1.3 bits per character or less indicates that the machine has AI."
This "K-Prize" will bootstrap AI.
OK, so he can christen it the "Page K-Prize" if he wants.
You can run with the equivalent of name servers that tell you the current best (QOS) routes given a particular system address.
The big failure of the DARPA guys was to presume that the hardware capabilities of the routers they were running with during the early days of IP was not going to scale along with the network. It did but by then we were stuck with their god-damned network class structures.
Aside from Xerox's 64 bit MAC address which was shelved as the basis for IP addresses, there was another standard promoted by a group of companies from Apple to Atari to Western Electric/Bell Labs to Packet Cable to Knight-Ridder circa 1982 which consisted of an unsegmented system identifier and object identifier combined in an 64 bit address -- the SID growing from the LSB up and the OID growing from the MSB down.
It would have been a very different and far superior world if they had not been stopped by DARPA's idiots.
Why is it that this ruling by a New York court reminds me that New York has the highest rate of HIV infection?
There are actually two pieces to this question:
Kolmogorov compression is defined in terms of the minimal bit string (usually as program running on a Turing-complete machine) it takes to describe another bit string. In order verify a contestant has won some increment of the K-Prize, it would be necessary to make his minimal bit string public so that anyone could "execute" the description to produce the original corpus.
This would produce the valuable result that the language model he is using would be published. The language model itself is very valuable since it contains the model, not just of the language, but of the conceptual world represented by the corpus. This model would be very valuable -- particularly as it began to exceed normal human capabilities to compress the corpus -- for it would contain many novel and quantifiably valuable philosophical, scientific and practical concepts.
The program used to compress the corpus might be very valuable as well or it might be just the "dictionary" of concepts used to compress the corpus. An example of the former is given by some work with advanced inductive logic programming systems that invent predicates. An example of the latter is given by the approach Cycorp has taken for 2 decades (without much success yet) where they hire a bunch of philosophers to describe, in a variety of "microtheories", various aspects of "common sense".
Lately Cycorp has, very wisely, given up the ghost with the idea of having really smart humans try to produce all the concepts and has started using some predicate invention logic.
So very soon now I predict there will be some very profound breakthroughs -- if for no other reason than the long dark ages of AI, where high priests are relied upon for all wisdom, are terminally ill and recognized as such by the high-priests of the "church" of AI.
For that reason I think it wise to demand that the compression algorithm be made public.
Think about the Rosetta Stone in terms of compression. You have a single message represented in 3 different languages. Now imagine a huge Rosetta Library -- the same library represented in 3 different languages. Clearly, optimal compression of this Rosetta Library is going entail precisely the "conceptual processing" you desire.
This penetrates all aspects of knowledge and society.
Quality information helps us simplify our world view without making it less accurate -- or conversely -- make our world view more accurate without making it more complicated.
It is sort of a power tool for philosophers.
"Publish or perish" produces precisely the explosion of words that creates the lack of understanding between people/disciplines. Academia just doesn't have the right incentives.
The real problem with intellectual property is that no one but the acquisitive can afford the lawyers that it takes to defend the royalty stream that should arise from it -- so we're beig inundated by pseudo-"inventors" who are really just tax collectors --thus destroying technological civilization's foundation.
I had a legislative proposal to fix that problem back in 1992 when I was doing technology politics, but the lock-down by acquisitors on politics is so tremendous I gave up on a "free market" approach, as well as academic approach to these problems. The only way around this stuff is creating new tools -- prize awards for technology is a good approach.
You're right that in the case of the X-Prize criterion, it was set up to favor simply throwing money at a reasonable technician -- but if the altitude goal had been set to 200km instead of 100km, John Carmack's team would likely have beat Paul Allen's team and done so on about 10% of Allen's investment.
Abstracting their prize criterion:
Previous record: X
New record: X+Y
Fund contains: $Z at noon GMT on day of new record
Winner receives: $Z x (Y/(X+Y))
Applying this to Kolmogorov complexity (ignoring several technical details for the moment):
S = size of uncompressed corpus
P = size of program producing uncompressed corpus
M = S/P
Anytime someone demonstrates a larger M, they are awarded money according to the abstract prize criterion above.
The only problem with it is that it doesn't include Mahoney's threshold of 1.3 bits per character for artificial intelligence.
I am quite excited by the confluence of advances in prize awards for technology advancement, and advances with the theory of compression. I'm convinced that if a substantial prize award can be created for dramatic advances in natural language text compression, it will lead directly to a solution to the most critical aspect of the "AI problem" -- that being the problem of the explosion of words without concomitant understanding. I had high hopes for the Internet being the new Gutenberg press leading to a new enlightenment but I'm concerned that without dramatic advances in AI to correlate the huge corpus being generated, the benefits of the new enlightenment may be too long in coming to save us from ourselves.
My work on a legislative proposal for fusion technology prizes was picked up by one of the founders of the Tokamak program. The more recent X-Prize award has a renewed the popularity of such prizes.
As a consequence I've been suggesting the creation of a new prize based on Kolmogorov complexity. As argued by Mahoney in "Text Compression as a Test for Artificial Intelligence":
"The Turing test for artificial intelligence is widely accepted, but is subjective, qualitative, non-repeatable, and difficult to implement. An alternative test without these drawbacks is to insert a machine's language model into a predictive encoder and compress a corpus of natural language text. A ratio of 1.3 bits per character or less indicates that the machine has AI."
A simple prize criterion would be for the first program producing a major natural language text corpus, with the size of the program being less than 1.3 bits per character of the produced corpus. Smaller intermediate prizes would help spur broader interest.
Computer assisted instruction has been rigorously shown to work in a wide array of educational settings. Similar claims can be made for television. The key is the content, not the medium.
Gangs operate as gangs because the individuals are weak on their own. They have a _lot_ of incentive to make sure they dominate any place integrity might show its face. When you have people of integrity they generally stand as individuals and simply have too much to deal with in a world that protects weak individuals.