CIVIL and the future of media?

David Siegel emails me:

CIVIL is a new start-up from Consensys, whose goal is to change journalism.

The Civil marketplace is built on a protocol that in turn is built on the Ethereum blockchain.

This ecosystem is built around a token-curated registry, using what we call a “skin-in-the-game coin,” the CVL. This is an application of mechanism design to blockchain-based tokens that can be acquired, exchanged, and go up in value, creating a new micro-economy for – in this case – truthy journalism. The basic unit of Civil is a newsroom. A newsroom is a person or group who can publish anything they like. They can charge readers using CVL tokens or credit cards or anything else. What makes Civil interesting is that anyone can challenge a story’s veracity.

To challenge a story, you send some CVL coins to a smart contract. The community then votes on the veracity of the story, or even the newsroom itself. Anyone who votes must stake coins. If the story is voted true, those who voted true take the pot – they win all the staked tokens. If the community finds it’s false, then those who voted for false share the purse. This skin-in-the-game mechanism is the next evolution of communities like Steem and is game-theoretically far more advanced than Reddit or Quora. It promises to eliminate fake ratings, reviews, and content farms pumping out propaganda. By creating token-based games that reward virtuous behavior – the first one of which was Bitcoin – today’s blockchain entrepreneurs promise to bring us a new era of less biased news, better blogging, more accurate ratings, and potentially better science.

Comments

If you ever find something that can be more easily gamed, let me know.

Yea, that was my immediate thought. Pick your bogeyman, Koch or Soros, how much do they have to spend to buy a result?

Key word there is 'spend'. A Russian troll farm only has to spend whatever it costs to have people come into the office, set up fake accounts and spend day in day out tweeting and sharing whatever fake memes you want.

Here if you want to 'buy the truth', well you could but you'll spend and possibly spend a lot and if you fail then everyone you're trying to take the truth from gets your money. Bitcoin is a bit like that. You could push an 'alternative ledger' on Bitcoin where everyone has gives you half of their bitcoins. To make that work, though, you need to buy enough computing power to challenge everyone else on the network.

You need to replace the word "truth" there with "popular" No one is required to vote on what is true, just what they like more. Hell, not even that, really.

Perhaps some outrageous fake stories will be voted away (like Canadian PM Trudeau has fake eyebrows), but for the most part this will simply be a popularity contest.

To be fair, the photos did look pretty convincing prima facie, and with the prior that Trudeau is a bit vain and over-manicured...

But no, he just has really unusual eyebrows. Happy to be wrong once in a while.

I know this website presents quality dependent articles and additional data, is there any
other web page which offers these kinds of data in quality?

Perhaps but then are you saying there is no demand for accurate news anywhere, just popular stories? Could CNN and the NYT save money by closing all field reporting and just have some creative writing students make stories up in the basement?

I suspect there is a market that wants accurate facts. It may not be the editorial page but it is out there. Consider FiveThirtyEight amassing data on all the available polls. Political junkies aren't seeking that type of data out there?

There is demand for accurate news, im just not sure it beats out everything else. Fox news is the highest rated news program in America.

If there is money on it, ill bet everyone votes to confirm their bias.

Fox is mostly content conservatives like, so it has a right-skew. So too, leftwing news sources. Some websites attract lots of different people and, even though there may be skew, it seems less overtly partisan. So you solve the "large skew" problem by attracting a little of everyone; then you add the skin-in-the-game quality to keep people honest.

Not everyone has to be honest. If you have enough dishonest actors on both sides to cancel one another out, the honest actors will all break one direction and give you a clear signal. There are ways this can break: some issues won't have enough supporters on both sides for a cancel-out effect. Not sure if there's a democratic system that solves that problem, though.

"If you have enough dishonest actors on both sides to cancel one another out, the honest actors will all break one direction and give you a clear signal. There are ways this can break: some issues won't have enough supporters on both sides for a cancel-out effect."

When it does work, this is how Wikipedia operates. The critical component is having enough users of all sides.

'If you ever find something that can be more easily gamed, let me know.'

Because it sounds like an excellent investment opportunity? Even the buzz word index seems reasonable in comparison to many such recent announcements.

Based on the wisdom of crowds, who are often not wise, just remember Hitler and Trump were elected

... but not by majorities in either case, so...... what's your point exactly?

So was Hugo Chavez.

But don't worry, the community will be composed of experts chosen from among the elite, the best and the brightest. What could possibly go wrong?

The two comments above missed my point - it costs money to vote. If you're in the minority, you lose your stake. This is analogous to having to pay 2 cents for any email message - a system that would eliminate spam immediately, because it negates the moral hazard. That's what makes it different from sites that are ALREADY the way you say - Quora, Steem, Reddit, etc.

But wouldn't this simply lead to very rapid and serious epistemic closure? There is no objective measure of "fake news" after all. It is just a measure of what people do not want to hear most of the time. So people who tell the group something they do not want to hear will be fined. It will be a strong pressure to conform.

We have this problem already. The Left and the Right don't really read the same newspapers any more or hear the same news. You see it here all the time. Surely this will just make it worse?

First, I don't really buy the claims that people who believe 'fake news' really believe it. For example, among global warming skeptics...I mean real skeptics as in "there is no warming" or "it's all to the good" not the "there is AGW but the costs of fixing it are bigger than the benefits in 90% of proposals" I suspect not even a quarter of them really believe it. They pay lip service to the positions because of tribalism, the opportunity to get positions and power, money etc. If you asked them to actually put their money down and lose it should their assertions be wrong....you'd discover the crowd will shrink very fast. I also don't think the reverse holds. If you asked global warming advocates to put $100 into an escrow account and lose it if, say, average temps 10 years from now are not higher...or not higher above some threshold, more often than not they'd do it.

So the way I understand this proposal: Newsroom publishes stories and readers pay for them by buying tokens. Paper keeps tokens *but* if someone challenges a story and the token holders agree it's bad the paper loses tokens plus. On the other hand if the community keeps agreeing with some other paper then fake news types will get driven to bankruptcy and such outlets would be bolstered.

I suspect this market would gravitate towards multiple possible equilibriums. Consider wikipedia vers conservapedia. The first is pretty much an example of the 'wisdom of crowds' by essentially making a market of reputation. The second is a niche market for propaganda and self-delusion (if you're a real popular contributor on the second all you've established is you can write stuff that people who agree with you 100% really like).

Like Bitcoin it only works for its purpose if you get a huge market but if you have a small niche one it will just create vanity presses for ideological ghettos.

f you asked global warming advocates to put $100 into an escrow account and lose it if, say, average temps 10 years from now are not higher...or not higher above some threshold, more often than not they'd do it.

Al Gore bought beach front property. He did in fact do what you asked. He put his money down - and bet against global warming.

Leonardo di Caprio flies like there is no tomorrow. He shows no signs of believing it. In fact the only Greenie who looks remotely serious was Ted Kaczynski.

"He did in fact do what you asked. He put his money down - and bet against global warming"

Actually that would be called a hedge. But according to Wikipedia Montecito is 180 ft above sea level so presumably it can manage even a dramatic sea rise of a few feet.

But then perhaps he has placed his bets on global warming. Wikipedia lists as notable residents not Al Gore but his ex-wife Tipper. So moving your ex-wife to ocean side property might just be a bet in favor of global warming!

Montecito is on average 180 feet above sea level. Gore's house is on the beach. The only thing between him and the sea is his lawn. I doubt it is 180 inches above sea level.

I think this would only work if you got buy in from fairly large number of actors from different intellectual perspectives.. A too small or not diverse enough crowd and you would simply have an enforced bubble where no one disagrees with the group think lest they lose their coins.

You kind of see this in Wikipedia, if enough editors are interested in writing a good encyclopedia article, or if there is a reasonable balance in perspectives, then the system generally works. If neither of those things is true, then you produce shitty articles and the system begins to break down under credibility problems.

Conservapedia is a key example. The only people who read it are like minded people so editors who put out a lot of crap are rewarded, provided it is crap in line with Conservapedia's readers.

But under this system Conservapedia's readers would be 'paying' for the outlet by giving them tokens, which cost them real money. As long as the editors and readers keep to a small pond, this will work nicely. But you could have 'short sellers' buy up tokens, then challenge some of the crap (i.e. 'earth is 6,000 years old') and start cashing in. Readers would have to decide if they love the crap so much that they pay even more to keep the short sellers at bay or if they give up.

On the other hand the reverse wouldn't work against Wikipedia. Conservapedia readers who try to buy tokens, challenge articles would be like trying to empty the ocean with a bucket. A 'hostile takeover' is possible either way but the larger, more diverse, your audience is the more expensive such a takeover would be. Might they go to the Koch Brothers and ask for a half a billion to take over Wikipedia? Sure but if that doesn't work all that was accomplished was transferring half a billion out of their hands and into Wikipedia's sane readers.

The way i understand the system, the tokens go to whomever wins a majority vote, truth or falsehood is assumed to be the deciding factor in a vote, but why should it be?

In your example, if you asked a thousand conservopedia readers which was true, a 6000 year old earth article or a several billion y.o. earth article, why would they vote for the latter? Why would they even be in conservopedia if they believed the latter? The only thing this system does is monetize bubble making. You win or lose based not on truth, but on popularity, whether you are a reader 'betting' on which article will win or a writer doing the same.

Yes but you don't just get to vote, in order to vote you have to pay some coins/tokens. You only get them back plus a share of the other side's tokens if you win.

The bubble making may be monetized but the useful thing about a bubble is it forces those participating in it to keep reaffirming their stake in the bubble. Every time real estate or Enron or tulips hit a new high are you going to stay in the market or cash out? As a player on Conservapedia you keep paying and voting the earth is 6000 years old but do you trust your fellow readers to keep pumping into that market? Maybe they consider "Hillary was a traitor" or "Pizzagate was true" more important articles for them to defend with their coins and leave your wallet empty!

Same is true in reverse. Are you going to keep throwing your money away in hopes that this time conservatives are going to change their minds about pizzagate? Why should they? Why would they even go to conservipedia if not to have their bias' confirmed?

When i say bubble, i mean intellectual bubble, not asset bubble.

As I said I think there's multiple equilibrium points here. A niche point would be that conservapedia becomes an intellectual ghetto and the token system isn't about confirming a consensus view of what's true but just a consensus of what 'orthodox right wing nutcasery' is. The other is a broader market where the 'short sellers' and 'long buyers' make bubbles difficult to form.

In theory Wikipedia could be bombarded tomorrow by a gang of new editors that try to force a "Earth is 6K years old" view onto it. Why do the Conservapedia editors not do that and instead work on curating the articles in their little niche ghetto that few bother to read as anything but a joke?

Because we already have this 'token' system in the sense of time. Attacking wikipedia would require a huge time investment and odds are the larger wikipedia community would shut down the attack making the entire effort a waste. For the same reason thousands of wikipedia editors see no point going over to conservapedia and trying to bring them to the light (plus I understand its founder rules his tiny pond like a tyrant, booting editors who he feels dis him or the party line).

"Why do the Conservapedia editors not do that and instead work on curating the articles in their little niche ghetto that few bother to read as anything but a joke?"

Couple of reasons. 1) Wikipedia has a system for determining truth outside of itself: 'reliable sources'. Ultimately Wikipedia is a summation of what gets printed in newspapers and journals and so on and so editors can appeal to that to resolve disputes. The system described here has no such external authority. 2) There are enough (mostly) neutral editors who care about Wikipedia to stop rampant abuse and 3) Liberals already do that, its just harder for you to notice because you likely dont visit the Wikipedia articles that are the worst offenders and/or your biases align with the biases of other editors.

As you point out in #3, #2 only works to the degree the community of editors makes it work. If Wikipedia saw it's editors swamped with young earth creationist editors, they would declare the Bible a reliable source and vote to resolve disputes in favor of 6000 year old earth.

It's not impossible that some mechanism along these lines could be useful, but history is littered with examples of how crowds can make bad decisions about accuracy.

In order to be plausible, any new iteration has to take that history as a starting point and explain how the new mechanism helps. Ideally there would be meaningful peer-reviewed literature backing up the claims about the new mechanism.

A) assuming peer review works, which I don't think it does, B) yes, these economies should be more Bayesian, with people who have more conviction (hopefully that = evidence) betting more. Can it be gamed? Yes, but it will cost money to consistently be wrong.

Is history really littered with such examples? Keep in mind it isn't just 'crowds' but the demand that each individual in the crowd have 'skin in the game'. I think the only real examples you have under that constraint would be bubbles where the 'crowd' is investors who not only believe the hype about a product/company/industry/etc. but risk their funds betting on it. But while I think bubbles are real they do tend to be rare.

What about the “truth” of non-binary claims, i.e. probabilistic claims based on degrees of belief? Can this betting mechanism be adjusted for such probabilistic claims?

Yes! A Bayesian coin would help us learn more about things like global warming, politics, scientific research, and much more. Please learn about mechanism design and be one of the first to launch such a coin. Everything we do now is an experiment; we can all learn from the various projects coming online in the next years.

True or False: (1) Maybe you're right, maybe you're not; but (2) many people agree with you and (3) the experts I've contacted all disagree.

What about just having fully funded investigative journalism staffed by experts, academics; staffed by people who have read everything there is to read regarding their focus of inquiry. So much of this tech aspect of "the future" is just smoke and mirrors to me. More Theranos.

We have that, it doesn't work, they lie.

The problem is that the business model doesn't work and never did. Running an advertising business is contrary to quality news. That's what makes the Civil experiment and others like it promising. Maybe (big maybe) readers will someday actually pay for the news they consume. They do anyway by buying stuff. Now maybe we'll see a news-based ecosystem. Or not - time will tell.

"...the community then votes on the veracity..."

Aye, there's the rub ...

"You know, I think you’re taking a lot of things for granted. Just tell me where in the world you find these angels who are going to organize society for us ? Well, I don’t even trust you to do that."
- Milton Friedman

There's always a rub. Bitcoin is a virtuous game played by tens of thousands of people that results in virtuous behavior (safeguarding the network) even though many of them would be happy to subvert it. Can there be functioning token economies around health, journalism, science, education, etc.? We're about to find out!

How is this different than the subscription model? what is the blockchain technology adding?

It adds betting. If you want to give something a negative or positive rating, you have to put up your money and be prepared to lose it if you're in the minority. This is not the case on Tripadvisor, Amazon reviews, etc.

"be prepared to lose it if you're in the minority."

I'm all for experimentation with markets for news, especially since the the incentives in the current system are perverse (some balance of the website getting more clicks and the journalists getting more access to high status people). But the way you describe it, it seems that the main incentive for CIVIL will be to agree with the majority, so that one is optimizing not for accuracy but for consensus. Possibly this will be better than the current system, but seems more like investing in pop music than in commodities.

So, based on the excerpt... there's literally nothing to tie the votes to the *actual* truth or falsehood of the story.

The tactics of how to take the money from the Other Voters don't include "being correct about whether the story is true or false". It's things like getting together a big group of people that always vote together (to take money from the losing group), not voting against the crowd when one side is comfortably ahead, voting with just the right amount of people at the right time (enough to win the pot, but not so overwhelmingly that the other side looks hopeless and no more suckers contribute).

Change "vote for whether the story is true or false" to "vote on whether the story is yellow or green" and all the mechanics stay the same.

"Change "vote for whether the story is true or false" to "vote on whether the story is yellow or green" and all the mechanics stay the same."

+1, there's no method of determining veracity. It's purely jumping on the band wagon.

It could fail. I think even if it does, we'll learn something. I don't think you can predict the outcome of a skin-in-the-game coin designed for virtuous behavior. Example: you could add quadratic voting, and then it would cost more than all the money on earth to overwhelm the system with votes. Skepticism is always healthy, but economic incentives have been shown to be a good way to drive behavior, better than hoping the New York Times will get the story right. See Taleb's recent book.

Re: quadratic voting. Yeah I don't you're original voting thing is quite right (in which, if I'm understanding correctly, the spoils are divided in proportion to stake among the correct voters). I don't know if quadratic is
right either though. You may want to make it isomorophic to a prediction market for "This sentence is assigned p>0.5 at time t ~ Exponential; also, this sentence is ostensibly assents to newsroom N being trustworthy."

Everyone above, except for David, has declared that CVL is sure to fail.

Voila! "CVL is sure to fail" is therefore TRUE. David loses his stake ;-) .

Hey now, I called it an investment opportunity - seems like bias in how votes are evaluated could be a problem.

Requiring an arbitration panel, which costs a reasonable sum, right?

The money making possibilities are endless when you have a business model based on people paying to judge truthiness.

Doesn't this system just propagate beating on however is losing?

The story is published -> Someone challenges it -> A lot of "internet citizens" come and bet at whichever is winning at the time (if the poll result cannot be viewed for the duration of betting, wait less than a week for the masses to self-organize into "always bet fake") -> PROFIT -> one-two months go by and the system becomes completely unusable and left to rot (those tokens trade basically at 0).

For an example of how masses that can derive profit decide to game the system check out how first iteration of "EVE Online: Project Discovery" went. Online players were tasked to work as neural network and decide what photograph meant what (it was a cell research), while getting some reward. The problem was hard, therefore everyone decided "always cytoplasm". The system then got almost no other answers and therefore got useless. One of the first things they needed to do for the second iteration was add a lot of spiked (known results) pictures and give bigger credit to those people who had higher rating (better at those randomly shown spiked pictures).

Do you propose we specifically spike the system and sometimes publish fake news and then lower the rating of those, that voted true? Seems weird and not how you Americans do such stuff. Although as a Russian, I fully approve of my own proposition!

"Seems weird and not how you Americans do such stuff."

That's right. We reward those who buy into the approved fake news and ostracise those who don't.

Printer/web-published journalism is OK. Newspapers add corrections or retract articles. The problem is the news in sound and video format. Ever seen a correction on TV news?

Printed articles are discrete units focusing on one topic, but what is a "story" in the context of a radio show, podcast or youtube? A "story" could be 4 phrases or a 10 second joke on TV news.

Who has that much free time to care about that and CIVIL?

Props to those behind this. It's not clear if it will work, but it's definitely interesting.

It seems the big worry is that people will "bet" on what they think the largest crowd will bet on, not necessarily what is correct, and that these two features will often be in disagreement. It seems like adding some kind of gentle contrarian bias to the system might be of use, though I'm not sure quite how.

They've decided to define who is the prettiest girl by having a Keynesian beauty contest. (Basically, I think they're using Vitaly's notion of a "Schelling auction," where the assumption is that the truth is the most natural Schelling point for people to converge to. That's probably true when the question is "what was the high temperature in NYC today," but probably not when there's a question that's driven by ideology. It's easy to converge on popular factual claims like:

a. The universe was literally created in six days.

b. Human CO2 emissions has a negligible effect on climate.

c. Saddam is associated with Al Qaida and is close to having a nuclear arsenal.

d. Vaccines cause autism.

e. Immigrants commit more crime per capita than natives.

As best I can tell, all those statements are factually wrong (though I hold that belief with much less certainty for some statements than for others), but they are also all held to be true by a large fraction of the IS public, including many people with a lot of power and wealth.

The other problem here is that there are many people who are demonstrably willing to spend serious cash to try to influence media coverage via boycots, letter-writing campaigns, lobbying, PR, advertising, etc. Those same folks would be willing to lose a lot of money to occasionally shut down, say, a *really annoying* investigative reporter who keeps making them look bad, but whose readers don't have a ton of money to help him win the auction.

TL;DR:

Problem #1: Many of the participants will be coordinating on ideology rather than truth as the Schelling point.

Problem #2: Many of the participants will be motivated by something other than winning money in these Schelling auctions.

All claims held mostly by people with no skin in the game.

Imagine a Democratic Congress back in the days of George W Bush passed the following authorization of the Iraq war by a veto proof majority:

"An invasion of Iraq is approved. An income tax hike of 5% will also be put into effect on those in the upper 10% of income. Five years from now if WMDs are found, the proceeds of that tax increase shall be refunded back to the taxpayers plus interest plus a premium equal to 1.5 times whatever gains the stock market had above interest. If no WMDs are found the Treasury shall issue no such refunds"

I suspect support for the Gulf War would have plummeted even at the height of the war frenzy and support by those impacted by this tax increase would have gone down even more.

Once you make people put their own skin at stake, their ability to be realistic gets a lot more focused.

Lots of parents refuse to vaccinate their kids. They have skin in the game, they genuinely want what's best for their kids. They just have an incorrect understanding of reality that's causing them to do something that makes their kids worse off instead of better off.

This scheme makes people effectively bet on which answer will win the auction. There are a lot of ways this differs from betting (and actually facing real personal consequences you care about) on which answer is right.

Some parents refuse to vaccinate their kids. Some parents might refuse medical attention for a compound fracture and try to set the broken limb themselves.

That odds that horror will hit any one particular unvaccinated kid are low so while the anti-vaxers have 'skin in the game' it's actually a very low amount of skin and they are actually unintentionally rent seeking because some of the risk they are increasing falls on other parents who don't buy into the vaccine hype. On the other hand the odds of bad things happening by self-treating broken limbs are very high and you rarely see parents who want to go there.

You are right the market here is veering towards a consensus view of reality and that's not the same thing as reality itself, but it's a reasonable bet there's a lot of overlap.

" Five years from now if WMDs are found, ..."

Of course WMD's were found in Iraq.....

Just a hypothetical, no doubt if anything like that is ever done in real life there will be a lot of fine print by what is meant by 'find WMD' so things like a handful of 20 year old mustard gas shells nearly falling apart with rust wouldn't count as vindicating a war sold as a stopping a breakaway sprint to a nuclear bomb.

It doesn't matter how much fine print you ladle on, because the final judges are not actually compelled by any mechanism whatsoever to adhere to the fine print, to reality, or even to their own view of reality.

The judges in the WMD law scenario have the actual ability to declare that anthrax spores in soil constitute WMDs, or that actual operational hydrogen bombs do not, depending on what outcome of the case they prefer. The only thing that can reverse their decision is if there is a higher level of judges to appeal to, judges with the same ability.

Similarly, the only "consensus reality" that the CIVIL mechanism will extract is the consensus of the voters on what way the voters will vote. The very fact that skin is in the game will cause voters irrational enough to vote on their beliefs about the underlying truth to be driven out of the game as they lose skin to players playing the actual game of predicting which way the vote will come out.

In my hypothetical scenario...you have something very much like a bond issue that has various covenants that set forth what a default is. In those cases you can get disputes over whether or not a company has violated them and then you have to go to court. There again you have the fine print as well as the quality of your legal system to determine things.

Yes, right, of course it would go to court, to be decided by judges.

The problem is that you seem to believe that a Federal judge's decision on a massive economic issue in a matter of major controversy over an Act of Congress would actually be governed by fine print.

I look for media sources who fire their employees for not being good journalists.

Has Breitbart ever fired anyone for making up stories?

Tyler,

Think bigger. Consider how the judicial system is a lot like a blockchain. The premise is cases are decided and they establish precedent that guide future cases. Precedents can be overturned but doing so is difficult and disruptive. They become even more difficult if the precedents are very old and were used as a building block of a lot of other precedents.

So now imagine "Common Law Chain". Judges can defy precedents but at a cost of some type of token. If other judges accept their reading they get the tokens back, otherwise they lose them and are 'overturned'. Instead of computing power, new tokens are randomly given to judges with the most 'credibility' built up in tokens.

Doesn't this seem insanely complicated?

Or am I being naive?

There is a perceived problem with the quality journalism. In the past, when there has been a problem with quality -- in almost every industry -- there has been a move to deploy quality management and quality control systems that have been developed, pilot tested, introduced and audited according to ISO standards. This has happened in food safety, accounting, teaching, etc.

As far as I can tell no one has attempted to develop a similar system for journalistic output. As far as I can tell, journalists seem very reluctant to have any sort of innovative management systems placed upon them at all.

Introducing a prediction market model seems unusual. Making stories 'truer' isn't going to mean more people will read them. Haidt's work convinced me that people assemble sets of information to accord with their moral reasoning. Adding a prediction model wouldn't change that.

Adding a QMS, however, might standardise processes that can be improved when mistakes are made.

And I understand these are quite different goals. But understanding how outlets make such a shoddy product in the first place might be a first step.

Introducing a prediction market model seems unusual. Making stories 'truer' isn't going to mean more people will read them. Haidt's work convinced me that people assemble sets of information to accord with their moral reasoning. Adding a prediction model wouldn't change that.

I'm not so sure. Peddlers of crap do seem to virtually blush when Snopes or Wikipedia is thrown at them. Those are 'consensus engines' of a sort that are difficult to game.

While everyone isn't going to be convinced of true things, it's a pretty powerful argument to say to them "why don't you throw some of your precious tokens in and challenge what you say is an untrue article"?

Snopes or Wikipedia or good for blatant untruth. The problem comes in when something is untrue but widely believed. The response to "why don't you throw some of your precious tokens in and challenge what you say is an untrue article" is "the majority of the users of CIVIL believe X, so they will vote for articles that say X is true irrespective of the actual truth."

Worse still is the coin aspect introduces a betting aspect to it. What happens when the majority of users dont care about whether an article is true or not, but rather are just betting on the outcome?

If you asked a thousand reddit users what the 'citizens united' supreme court case changed, id bet cash money the majority would say that "it allows corporations to donate to politicans' or something similar, even though you can go to wikipedia and see the real answer. Hell, you could go to the case itself and see the real answer, but this hasnt stopped nearly every liberal news outlet in the US from claiming otherwise.

Keep in mind voting isn't free, it costs you tokens. If your feel for the law is very fuzzy would you want to vote on the nuances of SC decisions? Since voting can cost me money, I would want to do research before voting in favor of something that seems like common sense.

A majority of the country probably supports the views that Twitter and Facebook purge members for not supporting. That fact isn't even slight evidence in favor of those views being correct.

The winners who control CIVIL will be that same side -- because they have most of the money and have more existing forums with which to propagate their arguments.

Might does not make right.

So, how does this work, if we have a story come out, its gets voted down as "fake" but then more evidence comes to light? Or voted on as true but ends up false.

Its going to look very dumb.

Or they will have to have judges who technically adjudicate claims. Like fact checkers. don't worry, they won't all be partisan lackeys...

What happens if someone 'publishes' a ledger in Bitcoin saying everyone gives me all of their coins (when they really didn't)? If the majority of computing power confirms that as the 'true' ledger, then it becomes true for all our purposes. The guard against that is assembling all that computing power would be hugely expensive, esp. when you have huge server farms controlled by diverse interests (China, Putin's Russia, etc.) who would obviously want to add their say against such a change.

Basically this is an engine for determining consensus by requiring those contributing to the consensus (either by putting out articles or trying to vote on them) to put 'skin in the game'. It's a matter of faith that the consensus view of reality will converge over time on what reality really is.

OK developing this idea a bit more.

"The Common Law Ledger"

Basically same idea but judges get so many tokens. Higher up judges get more tokens per year. Judges can affirm precedent in the ledger or challenge precedent. Challenging a precedent entails submitting one's tokens. Those opposed submit their tokens, the most tokens wins and writes the new precedent to the 'ledger'.

In this case you can have the Supreme Court overruled by a consensus of low level judges tossing in their few tokens to reject an unworkable decision. Likewise a low level judge can demonstrate natural leadership by writing original opinions that get adopted by the larger judiciary (and he would amass more tokens than a top judge).

Can't we look at existing betting markets, for example PredictIt, to see whether people vote based on truth or something else?

With predictit, the things people are betting on either happen or don't, so the system actually does reward voting based on the users' perceptions of reality.

But lots of people on the site were betting based on their ideological convictions, even when they were losing money doing that.

https://www.theringer.com/2018/3/21/17130490/predictit-politics-elections-gambling

What would have happened to this system if there was no underlying reality to fact-check against? Even losing money didn't dissuade the true believers. With no underlying reality, whoever can create consensus can control the vote.

I suspect the difference here is it is not simply "pay money and vote". It is "get tokens". You can buy tokens from others that have them or you can win them by being on the right side of votes you participate in. Accumulating tokens makes you more powerful but that is a stock of capital you can deplete if you suddenly go on a quest to force public opinion to your will.

This article completely misses the point. It's all about trying to corroborate facts when the "problem" (hah!) is divergent interpretation.

Most MSM article aren't stuffed with outright lies. But they are very selective about the data they report, the order and context in which they present it and helpfully smother the facts under layers of "analysis" to tell you the One True Way to Think.

That's the problem, not the lack of "facts".

Comments for this post are closed