BitShares Forum
Main => General Discussion => Topic started by: luckybit on October 28, 2014, 08:29:00 am
-
http://www.slideshare.net/friendlyfruit3804/casinogaming-keystroke-lotteries-a-speculative-essay-part-i
Maybe this could be turned into a DAC?
-
I am sorry ,but i can't open to the links.
If something is wrong with it?
:(
-
Hello,
You can see the article as I originally posted it on Scribd.com. Just go to:
https://www.scribd.com/doc/207299903/Virtual-Currency-Keystroke-Informant-Lotteries (https://www.scribd.com/doc/207299903/Virtual-Currency-Keystroke-Informant-Lotteries)
It's much more readable there, and I've provided an abstract.
-
Hello,
You can see the article as I originally posted it on Scribd.com. I can't post external links here, so just Google:
Virtual-Currency Keystroke/Informant Lotteries - Scribd
It's much more readable there, and I've provided an abstract.
Hey since you're here maybe you can work with our community? If you can implement some of your ideas why not see if you can become a delegate so you can get paid to pursue it?
Also it might be better if you explain the idea yourself. It's not so easy to understand and most people probably want to hear it directly from you.
-
Can someone explain the idea in a few sentences?
-
In a few sentences:
Instead of buying lottery tickets, people would do small amounts of work to earn them. On a computer, the smallest amount of work is a keystroke or a mouseclick.
Someone needs something typed. He offers $100 to have it done by midnight tonight. Guaranteed winner. This offer attracts (let's say) 100 typists online. They start typing the document. Some type the whole thing, others just a few lines. But in the end, each sentence of the document ends up being typed 50 times. That redundancy ensures accuracy because any one typist's mistakes are not made by anyone else. No proofreader needed. The program selects a keystroke or word that has also been typed by everyone else. It awards one of those people the $100. That person may have done a lot of work, or very little.
The typing could also be a translation -- group validated translations.
Separately, someone who has witnessed a crime will anonymously submit the name of the perpetrator. A keystroke done by a parolee earning good-behavior points but not eligible to win the lottery ends up being won by the person who provided the information -- informants can put a crime in play by submitting information about that crime. And they might win an award. Or not. But if they don't win, they haven't risked anything, and under the current system of bounties they would not have provided information anyway. So they have nothing to lose by providing cops with dangerous information. Using bitcoins no personal information need by given.
You can see how the concept of smart contracts could work with this idea. The owner of the document funds the lottery by sending bitcoins to be rewarded to an address that is controlled by the contract. (The award could be priced in dollars). The owner/funder stipulates a certain level of keystroke redundancy (fifty in this example). Once that number is reached, the computer program randomly selects a typist and sends them the money to the public address they submitted when they logged on.
For another simple explanation, please go to the version I uploaded on Scribd and just read the first section. There is also an "Objection!" section that deals with various criticisms made to the idea.
-
Hello,
You can see the article as I originally posted it on Scribd.com. I can't post external links here, so just Google:
Virtual-Currency Keystroke/Informant Lotteries - Scribd
It's much more readable there, and I've provided an abstract.
Hey since you're here maybe you can work with our community? If you can implement some of your ideas why not see if you can become a delegate so you can get paid to pursue it?
Also it might be better if you explain the idea yourself. It's not so easy to understand and most people probably want to hear it directly from you.
I hope to get involved as I learn more about Bitshares. I own some, as well as bitcoins, ethers, ripples, and dogecoins, so I understand the rudiments of DACs, smart contracts, and the blockchain.
I will do my best to respond to questions about my keystroke lottery idea. It's pretty simple once you get the hang of it. I recommend just reading the Scribd version, especially the "Objection!" section.
-
In a few sentences:
Instead of buying lottery tickets, people would do small amounts of work to earn them. On a computer, the smallest amount of work is a keystroke or a mouseclick.
Someone needs something typed. He offers $100 to have it done by midnight tonight. Guaranteed winner. This offer attracts (let's say) 100 typists online. They start typing the document. Some type the whole thing, others just a few lines. But in the end, each sentence of the document ends up being typed 50 times. That redundancy ensures accuracy because any one typist's mistakes are not made by anyone else. No proofreader needed. The program selects a keystroke or word that has also been typed by everyone else. It awards one of those people the $100. That person may have done a lot of work, or very little.
The typing could also be a translation -- group validated translations.
Separately, someone who has witnessed a crime will anonymously submit the name of the perpetrator. A keystroke done by a parolee earning good-behavior points but not eligible to win the lottery ends up being won by the person who provided the information -- informants can put a crime in play by submitting information about that crime. And they might win an award. Or not. But if they don't win, they haven't risked anything, and under the current system of bounties they would not have provided information anyway. So they have nothing to lose by providing cops with dangerous information. Using bitcoins no personal information need by given.
You can see how the concept of smart contracts could work with this idea. The owner of the document funds the lottery by sending bitcoins to be rewarded to an address that is controlled by the contract. (The award could be priced in dollars). The owner/funder stipulates a certain level of keystroke redundancy (fifty in this example). Once that number is reached, the computer program randomly selects a typist and sends them the money to the public address they submitted when they logged on.
For another simple explanation, please go to the version I uploaded on Scribd and just read the first section. There is also an "Objection!" section that deals with various criticisms made to the idea.
Reddit seems to be implementing something similar to a lottery distribution for Reddit Note. You might be able to contact them as well so they know your idea as well.
http://www.redditblog.com/2014/12/announcing-reddit-notes.html
-
It's thoughtful. I just wonder whether the kind of person who plays a game of chance (such as a lottery) wants to work or just sit on his/her ass. Maybe there are people willing to work who also would go for this kind of incentive, but maybe not.
-
In my Scribd.com posting, I quote Alexander Hamilton:
"Everybody, almost, can and will be willing to hazard a trifling sum for the chance of a considerable gain."
With internet access becoming ubiquitous, people who have the time to sit around can do literal keystrokes of work that are themselves a "trifling sum" for the chance of a [Lotto magnitude] "considerable gain."
In the Scribd paper, I describe how typists -- players -- need not be committed to working a set number of hours or keystrokes, or even start at the beginning of the document to be typed. Everyone -- anyone -- with web access can log on, submit a bitcoin address, do some typing -- really any amount of typing -- and earn the chance to win a payout funded by whomever needs the typing done. People on break at the office for a few seconds or minutes; on the subway; on the plane; in the airport; in the den; on the couch; in the company lounge; in the gym; in the lobby waiting for a job interview -- anywhere or everywhere you would just log on and do some typing. You follow copy as closely as you can because you know that only validated -- replicated -- keystrokes have a chance of winning. Beavis and Butthead inputting random keystrokes would just be wasting their time because those random keystrokes will not be replicated. Finally even they get the message: just follow copy and they might win.
So the need for word processing and even translations can be outsourced and proofreading eliminated. Closer deadlines with higher mandated levels of replication would require higher payouts, or perhaps smaller payouts would be pooled into many such smaller payouts to make one bigger one. Lotteries could be worldwide, like Powerball-level games.
The really good thing here is that sedentary people who buy their Lotto tickets at the liquor store could instead earn their tickets without paying cash for them. Just do a few keystrokes anywhere you are. People who do not gamble might even be tempted, but now the form of gambling that is tempting them is fundamentally virtuous. They do productive work rather than pay cash. What could be wrong with that? Especially since no matter how much typing you actually do, you will only rarely win even trifling payouts -- exactly like conventional lotteries, except for them you pay cash.
Anyone who has become disenchanted with keystroke lotteries would be unlikely to resume paying cash for the same chance to win. That would be a good thing too.
-
In a few sentences:
Instead of buying lottery tickets, people would do small amounts of work to earn them. On a computer, the smallest amount of work is a keystroke or a mouseclick.
Someone needs something typed. He offers $100 to have it done by midnight tonight. Guaranteed winner. This offer attracts (let's say) 100 typists online. They start typing the document. Some type the whole thing, others just a few lines. But in the end, each sentence of the document ends up being typed 50 times. That redundancy ensures accuracy because any one typist's mistakes are not made by anyone else. No proofreader needed. The program selects a keystroke or word that has also been typed by everyone else. It awards one of those people the $100. That person may have done a lot of work, or very little.
The typing could also be a translation -- group validated translations.
Separately, someone who has witnessed a crime will anonymously submit the name of the perpetrator. A keystroke done by a parolee earning good-behavior points but not eligible to win the lottery ends up being won by the person who provided the information -- informants can put a crime in play by submitting information about that crime. And they might win an award. Or not. But if they don't win, they haven't risked anything, and under the current system of bounties they would not have provided information anyway. So they have nothing to lose by providing cops with dangerous information. Using bitcoins no personal information need by given.
You can see how the concept of smart contracts could work with this idea. The owner of the document funds the lottery by sending bitcoins to be rewarded to an address that is controlled by the contract. (The award could be priced in dollars). The owner/funder stipulates a certain level of keystroke redundancy (fifty in this example). Once that number is reached, the computer program randomly selects a typist and sends them the money to the public address they submitted when they logged on.
For another simple explanation, please go to the version I uploaded on Scribd and just read the first section. There is also an "Objection!" section that deals with various criticisms made to the idea.
Reddit seems to be implementing something similar to a lottery distribution for Reddit Note. You might be able to contact them as well so they know your idea as well.
Thanks for the info. I will take a look.
-
Most people don't like doing work with some random chance of getting paid. Your economics/psychology don't line up very well IMO. Honestly the idea doesn't even make much sense but that may just be me.
-
You are right that most people wouldn't want to do work for some random chance of being paid. What makes this idea different is the *small* amount of work required to "play" -- like buying a $5 (or less) Lotto ticket -- and the ease of accessing that work. The typing would indeed be work, but you could literally log off in the middle of a word as soon as you got bored and still have a statistical chance of winning the pot. And remember the pot could grow *very* large -- Powerball size payouts.
So: lots of work available anywhere, lots of random amounts of time people have to log on and participate whenever they want to, and either small and fast payouts with a guaranteed winner or huge payouts with no guaranteed winner (but which someone, somewhere *will* eventually win, since funders of keystroke lotteries -- document owners who need typing/translating -- would not get their money back once they sent it to their lottery's public bitcoin address, controlled by a smart contract). The cumulative amount of work outsourced to the planet would be huge.
I don't see how anyone would prefer buying a Lotto ticket instead. Can you imagine someone who has won $20 in bitcoins in a keystroke lottery immediately converting those bitcoins to dollars and then walking to the nearest liquor store and spending it on a Lotto ticket? Why would they? It's not as if "playing" that kind of Lotto is "fun".
-
I love this idea.
You work for the dollars you spend on a lottery ticket anyhow, why not hit some keys in hopes of a chance of winning? The more work you do the better your chances too, correct? I'd rather spend 1 hour working on a computer for a chance at winning than spend 1 hour of pay from shoveling asphalt on a lottery ticket.
This forum and the ideas that come through it are a big part of why I hold my Bitshares close to me.
-
You've *almost* got it!
Remember that although you could type for an hour if you wanted (you could type all day if you wanted), most people would probably just log on and type as their day allows -- on break, waiting in a lobby, etc. Probably a minute or two once or twice a day would suffice for most. For every person who is quickly bored or runs out of time, there will always be at least one person to take their place. Think of stars winking on and off as you look in the sky. Those winks are people around the world logging on and off, yet working together to achieve the same goal: type-set consensus -- consensus alone determining accuracy. And only with sufficient consensus achieved will a lucky winner be selected by the smart contract following the terms established by the funder.
-
The problem is that someone could submit someone else's work. No actual typing need occur.
-
I'm not sure what you mean. There would be no reason why someone funding a keystroke lottery would submit someone else's document to be worked on, although if they did it would be irrelevant to the system itself. As for the typists, the system could easily detect copy pasting. It would only accept keystrokes made in real time. So work could not be "submitted." And software bots that produce keystrokes would not work either, because such keystrokes would not be validated by anyone else -- only keystrokes replicated by others following copy as they typed would be accepted. Think of keystrokes overlaying other keystrokes.
You would have to get a *very* large group of people together to try to swamp a given lottery. But who would participate and why? Why not just play along with everyone else around the world working on that document? Might as well, since you would have to pay your own pool of typists for their labor -- surely a money losing proposition!
It's a bit like bitcoin mining. Why try for a 51% share for the purpose of controlling the system? You would just destroy it and lose the money you invested in order to achieve that control.
-
I don't see how anyone would prefer buying a Lotto ticket instead. Can you imagine someone who has won $20 in bitcoins in a keystroke lottery immediately converting those bitcoins to dollars and then walking to the nearest liquor store and spending it on a Lotto ticket? Why would they? It's not as if "playing" that kind of Lotto is "fun".
I'm not trying to be rude, but I'm not sure who you are trying to convince in this paragraph. You should never try to convince yourself with your own hype that your own ideas are solid.
It is interesting theoretically .. I just can't wrap my head around it working. Perhaps the big payoff psychology.
-
No offense taken. I think the reason you can't visualize it working is that you are not focused on the money to be won. This is all about people doing minimum amounts of work for the chance of winning a jackpot. That's it. Greed wins. Or call it ambition since it requires work to play. And increasingly ubiquitous Web access facilitates that ambition.
Let's imagine that someone decides to try this system, and posts the first document to be typed or translated. Let's say the payoff is quite small -- $20. If you saw an unused Lotto ticket for that amount on the sidewalk, would you decline to pick it up? In a keystroke lottery, all you would have to do is spend a few seconds logging on and typing until bored, following the copy very closely and literally; or translating as you think others will translate it -- this is all about consensus, remember.
If anyone thinks this idea would not work, I would welcome your comments as to exactly why. A few years back a Microsoft project manager came up with a list of objections, but I was able to answer them to his satisfaction. (See the Objection! section of the Scribd paper.)
I've never bought a Lotto ticket, but the people I have watched buy them don't particularly seem to be enjoying themselves. Nor do most of the gamblers you see in Las Vegas. An alien from another planet could be forgiven for assuming such people are actually hard at work (which goes double for the slot players).
But in spite of the fact that typists would indeed be working to gamble instead of paying cash, their sole motivation would be to earn the chance of a payout as quickly and/or easily as possible. And that could be achieved only by consensus. I point out in the Scribd paper that this idea could just as well be called Consensus Lotteries.
-
I don't see how anyone would prefer buying a Lotto ticket instead. Can you imagine someone who has won $20 in bitcoins in a keystroke lottery immediately converting those bitcoins to dollars and then walking to the nearest liquor store and spending it on a Lotto ticket? Why would they? It's not as if "playing" that kind of Lotto is "fun".
I'm not trying to be rude, but I'm not sure who you are trying to convince in this paragraph. You should never try to convince yourself with your own hype that your own ideas are solid.
It is interesting theoretically .. I just can't wrap my head around it working. Perhaps the big payoff psychology.
I think I can explain. It works by operant psychology. The variable interval schedule is the mechanism by which all lotteries work and this is proven by the success of the gambling industry. The science behind what he is doing is sound which is why I made the OP because it fit with my own research on similar subject matter.
The implementation is where it gets tricky. I think if Bitshares PLAY can work then so could this but it would require some sort of decentralized identity mechanism. I think for example if done through Bitshares then you could set up lotteries so only individuals with verified identities can take part.
As long as the verification process is anonymous and does not link to a direct person it can work fine. All you would need is some unique proof of human protocol which verifies that a unique human being is participating. I think there are technical challenges but it is doable.
-
First, I want to say thanks to Luckybit for posting this topic here. A while back I tried to start a discussion on ethereum.org's general discussion forum. It has so far generated all of 40 or so views and not a single comment.
Regarding your most recent reply, I take it that your reservations are two-fold:
1) how to establish whether actual people and not bots are participating as typists -- bots that would presumably try to swamp the system with falsely group-validated keystrokes.
2) fitting the lotteries into Bitshares PLAY in order to generate truly random winners without relying on a trusted central authority.
My comments:
1) Where the material uploaded to be typed/translated is handwritten, the work itself would prove that a human is participating and not a bot because software can't yet read handwriting nearly well enough. There is no way a machine could produce keystrokes that would be validated by the majority of (human) typists. As an additional form of protection, Captchas could be randomly required to continue typing, each Captcha being unique to each typist. And only keystrokes in real time would be acceptable.
In the case of a Word file uploaded for, say, group translation, Captchas could be randomly required here too. (The entire document might even be converted into a Captcha-looking format.)
[Edit: I'm not really sure all those Capcha tricks would really be necessary. If there were software programs out there that could fake keystrokes (or even roughly mimic handwritten text) for the purpose of swamping the system, there could also be software programs out there that could catch such programs, and there is no way such programs could be validated by typists doing real work. The fake stuff would stand out. And why would anyone use such a program anyway? Remember that chance would still rule. Easier to just type right along with everyone else for the purpose of achieving consensus.
[As for Word files uploaded for translations, any one or group of people who tried to copy/paste the text into a Web translator would still have to type the translated result in manually. But even so, the system could in theory be swamped by billions of Beavis and Buttheads using the same online translator. The solution would probably be to just to automatically run such a file through an identified online translator as a matter of course before submitting it to the typists -- and then have those typists correct the machine-made result. Here the work of real people working together to achieve consensus would defeat the copy/pasters.]
2) For those willing to pre-register with their real names and submit SSNs, etc. to firmly establish their identities (and pay taxes on any substantial winnings) I think it would be routine to establish a trustworthy way of randomly finding a qualified winner. These lotteries might even be government supervised, like a state Lotto. Libraries and archives seeking to digitize their handwritten material could fund such lotteries. Of course, such lotteries need not be bitcoin-based. Dollars and Paypal would do.
Even confidential material -- lawyers’ handwritten notes and diaries -- could be processed. It should be fairly easy to develop a smartphone app that would take a picture of a handwritten page (or of a computer screen of text) and then allow that picture to be parsed into separate files which would then be uploaded to separate lotteries for typing, thus avoiding any one virtual pool of typists from knowing too much about it. Looked at this way, it might not even be necessary after all to use smart contracts.
But competition for typists might induce lottery funders to go via Tor to the Dark Web, thus allowing tax-free payouts and no identity registration. Here is where smart contracts would be required. But how difficult would it really be to have transparency? We want typists with group-validated keystrokes to earn tickets proportional to the work they do. I’m no mathematician (which may be obvious), but let’s say every X number of each typists' keystrokes + a unique and viewable random series of digits/letters are hashed using Sha-1. The result will be a unique 40-digit hexadecimal number. That number could then be fed automatically into bitaddress.org's brainwallet function to produce a public and secret key. (They would be used only as a time/date stamp.) The hash-input and public key would constitute a lottery ticket. Each lottery in advance could state what its own "goal number" hash/public key will be. It might be the previous days’ Dow Jones closing fed into the hash-bitaddress generator, the point being that it is something transparent and anyone can confirm it. Now as people type, they could watch their own tickets being generated and consensus-validated in real time. The ticket-hex number closest to (or perhaps furthest from, see below) the goal-number (again, stated in advance) would win. In this way, typists could see as they type how close to or far away from they are from likely winning anything. Of course, mere submission of a "winning" hex number by anybody on the Web would not be enough. Only logged-on typists with the actual recorded keystrokes would qualify because only they would have the numbers that produced the hashed result.
(Presumably the calculations to produce the numbers would be strictly client-side, to prevent other typists knowing anyone else's tickets, although some lotteries might allow everyone to see others' tickets as a reflection of typists' ability to communicate with each other anyway. In that case, the winning ticket would probably have to be the one furthest from a lottery's goal-number announced in advance, otherwise premature public knowledge of a generated random number very close to the goal number would discourage further participation. Even with that arrangement, a given lottery might have to offer rounds of payouts as knowledge of a generated number vastly far from the goal-number would encourage typists to quit and move to a more favorable "game". In fact, tracking of the numbers might generate a kind of side-game in which non-typists bet on whether a second round of payouts will be required. After all, true gamblers will bet on anything -- think of the famous example of betting on which drop of rain will reach the window sill first.
(Note that the above describes a process not unlike bitcoin mining, in which people are generating random numbers in a kind of "proof of work".)
Admittedly this is all a crude outline, but my point is that it shouldn’t be too hard to create a smart contract that can automatically collect the needed information (Dow Jones closing figure, etc.); create typists’ lottery tickets and the winning ticket in a transparent manner; and then send the winner the bitcoin winnings.
Or maybe I just misunderstood your comments. If so . . . never mind . . .
-
You should create Keystroke as a DAC using the BitShares toolkit. This community likes experimentation. If you simply comply with the social consensus (provide at least a 20% sharedrop to the BitShares community), that still leaves you with up to 80% ownership of your idea. And you'll have a vibrant community of stakeholders behind you here that includes developers and marketers.
-
You're right -- I should move in that direction. The concept might well be implemented as a DAC. But although I've been buying virtuals since 2011, I'm totally new to Bitshares and still making my way through the Bitshares Wiki. I gather the Toolkit itself is still in the early stages.
My generic description has been out there in the public domain for quite a while. I first posted the bare-bones outline of it on Google's 10^100 idea project of 2008. After that closed down I expanded on the concept and posted it to Google's Knol website. It garnered a few comments (mostly from a Microsoft project manager), but I began to notice that bots were copying the text and translating and back-translating it and posting it on various dubious websites. (At least they kept my name and email address.) I also posted it to a couple legitimate websites myself. After Knol was shut down, I moved it to Scribd where it resides today, ignored.
-
You work for the dollars you spend on a lottery ticket anyhow, why not hit some keys in hopes of a chance of winning?
Because there are already a million* other sites where you earn actual fiat / crypto for doing mundane tasks like this?
*) actual number may be less than 1 million.
-
But with keystroke lotteries you are actually gambling in exchange for your keystrokes in hopes of a payoff. It's just like any other form of gambling except you aren't paying cash. Instead of working for the cash to buy a Lotto ticket or play the slots, you just work directly at it. This is critical in understanding this idea -- people would indeed be working as they type/translate, but they are also gambling, pure and simple, hooked on the hope.
-
I think this makes sense. Most people who buy lottery tickets are already doing mundane tasks at their jobs to earn money to buy them.
Also, a lot of lottery tickets make you do work, even after paying for them, i.e. scratch cards. This work is pleasurable for the people who buy them, however this may be strongly influenced by the possibility of winning.
Instead of marketing it as "do work, earn a chance to win" you could market it as "HUGE JACKPOT! FREE TO PLAY!" and make it seem like you're giving them something rather than them giving the system something.
Heck, if you're clever enough, you could even CHARGE them to do the work, just like the scratch cards.
Whether this is ethical or not is another question... humans are highly irrational, obsessive, compulsive beings, and these traits can lead them to be enslaved by other humans willing to exploit them on these traits, all while seeming to allow them to make their own free choices.
-
I was hoping someone would get around to the ethics of a keystroke-lottery system. I do get into that a little in my Scribd posting (at https://www.scribd.com/doc/207299903/Virtual-Currency-Keystroke-Informant-Lotteries). I will take the liberty of quoting from it below:
"Because all this [keystroke lottery] activity would be taking place in a digital environment, it might be possible to select projects that offer odds modeled on popular conventional lotteries. This would allow participants to learn the practical meaning of the odds that such lotteries offer, but without paying cash for the lesson. In any case, as many chances beyond their customary daily or weekly conventional-lottery ticket-purchases that they might pay out in the form of work done at their keyboards, most people would still never win a significant payout -- even as they watched a news-ticker roll across their monitors announcing the names of winners around the world who had. So a keystroke lottery could be educational enough — California Lotto revenues are supposed to help education — to make conventional cash-purchases of lottery tickets unappealing to larger and larger numbers of players. More and more of them might switch to keystroke lotteries, or even — the more perceptive among them — quit lottery-gambling entirely. That said, although few players ever win lotteries, there have always been enough players to re-generate interest in a future payout. A chance for the money, as always, is the attraction. Quitters return to the fold. And so they would for keystroke lotteries."
-
Below you write
". . . let’s say every X number of each typists' keystrokes + a unique and viewable random series of digits/letters are hashed using Sha-1. The result will be a unique 40-digit hexadecimal number. That number could then be fed automatically into bitaddress.org's brainwallet function to produce a public and secret key. (They would be used only as a time/date stamp.) . . ."
Generating an address alone won't generate a time/day stamp. You have to put bitcoins into the address. So the typists would themselves have to be governed by a smart contract that would move a minimum amount of their own coin into an address with every X number of keystrokes, right? And since others might see the addresses generated, the coin would have to be moved right back out again.
First, I want to say thanks to Luckybit for posting this topic here. A while back I tried to start a discussion on ethereum.org's general discussion forum. It has so far generated all of 40 or so views and not a single comment.
Regarding your most recent reply, I take it that your reservations are two-fold:
1) how to establish whether actual people and not bots are participating as typists -- bots that would presumably try to swamp the system with falsely group-validated keystrokes.
2) fitting the lotteries into Bitshares PLAY in order to generate truly random winners without relying on a trusted central authority.
My comments:
1) Where the material uploaded to be typed/translated is handwritten, the work itself would prove that a human is participating and not a bot because software can't yet read handwriting nearly well enough. There is no way a machine could produce keystrokes that would be validated by the majority of (human) typists. As an additional form of protection, Captchas could be randomly required to continue typing, each Captcha being unique to each typist. And only keystrokes in real time would be acceptable.
In the case of a Word file uploaded for, say, group translation, Captchas could be randomly required here too. (The entire document might even be converted into a Captcha-looking format.)
[Edit: I'm not really sure all those Capcha tricks would really be necessary. If there were software programs out there that could fake keystrokes (or even roughly mimic handwritten text) for the purpose of swamping the system, there could also be software programs out there that could catch such programs, and there is no way such programs could be validated by typists doing real work. The fake stuff would stand out. And why would anyone use such a program anyway? Remember that chance would still rule. Easier to just type right along with everyone else for the purpose of achieving consensus.
[As for Word files uploaded for translations, any one or group of people who tried to copy/paste the text into a Web translator would still have to type the translated result in manually. But even so, the system could in theory be swamped by billions of Beavis and Buttheads using the same online translator. The solution would probably be to just to automatically run such a file through an identified online translator as a matter of course before submitting it to the typists -- and then have those typists correct the machine-made result. Here the work of real people working together to achieve consensus would defeat the copy/pasters.]
2) For those willing to pre-register with their real names and submit SSNs, etc. to firmly establish their identities (and pay taxes on any substantial winnings) I think it would be routine to establish a trustworthy way of randomly finding a qualified winner. These lotteries might even be government supervised, like a state Lotto. Libraries and archives seeking to digitize their handwritten material could fund such lotteries. Of course, such lotteries need not be bitcoin-based. Dollars and Paypal would do.
Even confidential material -- lawyers’ handwritten notes and diaries -- could be processed. It should be fairly easy to develop a smartphone app that would take a picture of a handwritten page (or of a computer screen of text) and then allow that picture to be parsed into separate files which would then be uploaded to separate lotteries for typing, thus avoiding any one virtual pool of typists from knowing too much about it. Looked at this way, it might not even be necessary after all to use smart contracts.
But competition for typists might induce lottery funders to go via Tor to the Dark Web, thus allowing tax-free payouts and no identity registration. Here is where smart contracts would be required. But how difficult would it really be to have transparency? We want typists with group-validated keystrokes to earn tickets proportional to the work they do. I’m no mathematician (which may be obvious), but let’s say every X number of each typists' keystrokes + a unique and viewable random series of digits/letters are hashed using Sha-1. The result will be a unique 40-digit hexadecimal number. That number could then be fed automatically into bitaddress.org's brainwallet function to produce a public and secret key. (They would be used only as a time/date stamp.) The hash-input and public key would constitute a lottery ticket. Each lottery in advance could state what its own "goal number" hash/public key will be. It might be the previous days’ Dow Jones closing fed into the hash-bitaddress generator, the point being that it is something transparent and anyone can confirm it. Now as people type, they could watch their own tickets being generated and consensus-validated in real time. The ticket-hex number closest to (or perhaps furthest from, see below) the goal-number (again, stated in advance) would win. In this way, typists could see as they type how close to or far away from they are from likely winning anything. Of course, mere submission of a "winning" hex number by anybody on the Web would not be enough. Only logged-on typists with the actual recorded keystrokes would qualify because only they would have the numbers that produced the hashed result.
(Presumably the calculations to produce the numbers would be strictly client-side, to prevent other typists knowing anyone else's tickets, although some lotteries might allow everyone to see others' tickets as a reflection of typists' ability to communicate with each other anyway. In that case, the winning ticket would probably have to be the one furthest from a lottery's goal-number announced in advance, otherwise premature public knowledge of a generated random number very close to the goal number would discourage further participation. Even with that arrangement, a given lottery might have to offer rounds of payouts as knowledge of a generated number vastly far from the goal-number would encourage typists to quit and move to a more favorable "game". In fact, tracking of the numbers might generate a kind of side-game in which non-typists bet on whether a second round of payouts will be required. After all, true gamblers will bet on anything -- think of the famous example of betting on which drop of rain will reach the window sill first.
(Note that the above describes a process not unlike bitcoin mining, in which people are generating random numbers in a kind of "proof of work".)
Admittedly this is all a crude outline, but my point is that it shouldn’t be too hard to create a smart contract that can automatically collect the needed information (Dow Jones closing figure, etc.); create typists’ lottery tickets and the winning ticket in a transparent manner; and then send the winner the bitcoin winnings.
Or maybe I just misunderstood your comments. If so . . . never mind . . .
-
You are right, but the more I think about it the more I think that a time/day stamp per ticket would not necessarily be required.
But your point does suggest how many different keystroke-lottery implementations there could be. For example, based on a growing typesetter consensus on a given document, the smart contract could move the jackpot to and from a succession of wallets designated beforehand as belonging to each logged-on typist, with the momentarily latest wallet-location of the jackpot reflecting some metric of an ever-shifting keystroke consensus. (Of course, the typists would not have the password to those wallets.)
This brings up another point. Gamblers like ongoing feedback, whether it is a show of cards at a blackjack table or a tumble of fruit in a slot-machine. I think that a successful implementation would show anyone (even non-typists) everyone else's keystroke-generated lottery tickets as they are generated. The program would instantly show its momentary chances of winning. A ticket with a relatively high probability of winning could be offered for sale. Of course, in such a system each typist would need a secret randomizer to include with their keystrokes before being hashed. Should their ticket win the lotto, or if they agreed to sell it before the drawing (such bidding/buying/selling being done instantly by a smart contract), their "proof of work" would be their keystrokes+their secret randomizer, which together produced the hexnumber. I think this is why a time/day stamp would not be needed for each lottery ticket generated, although come to think of it, the game's "goal number" (which would be stated in advance) could also be included in the hash. That would effectively make at least a day stamp for each generated ticket, which could not have been generated before that day. I think.
-
This idea is not ready for prime time. For one thing, why would anyone bother to type the parts that were difficult to read or translate? You would end up with clumps of consensus, like lumpy soup. Only the easy parts would attract typists, so you need a way to get people to take the risk of attempting consensus on difficult parts of the supplied text. In other words, difficult text needs to be made a part of the game itself.
Another problem is that if each keystroke string generates a unique hash per typist, how will consensus be determined? It seems to me you'd need to have people generating the same hashes again and again to reflect the same words being typed.
-
The kind of consensus clumping you describe would disappear as document owners -- the people who would fund the lotteries with their own money -- realized that certain documents are simply not suitable. The process is not magic. It's just a bunch of people looking at a certain document at the same time and typing what they see in order to maximize their consensus-validated keystrokes. Probably over time, the funders (say, lawyers who can't or won't type up their yellow-pad scribblings) will write up their notes with an eye towards basic legibility. This would not be too hard for most people. For years I worked as a proofreader in the financial printing industry. Lawyers would arrive onsite and have meetings, and then pass their yellow-pad notes on to the composition department. It was pretty rare for us to have to ask them to clarify their handwriting, it being a minor point of pride to be able to decipher the hieroglyphics.
The mass collective of typists working a lottery would probably be able to decipher bad handwriting well enough -- the wisdom of crowds in action -- but your objection does raise one danger: that typists would just fake consensus on difficult parts, working together the way professional card players can silently identify each other during a game and work as a team by a kind of intuition. To prevent this, I think it would be essential for typists to be able to seamlessly flit from one document to document -- from lottery to lottery. So when they encounter handwriting or a translation problem that they know will resist consensus, they can just leave that project and go to another more promising one.
As for your second objection, I think it is essentially a technical point because all kinds of differently styled games are theoretically possible, but probably consensus would not necessarily be determined by examining the lottery tickets that the typists generate as they type. In one possible approach, such tickets -- hashes -- would be there to allow everyone to see exactly what keystrokes (plus the winning typist's publicly-known randomizing string, assuming one would be needed for security) generated the winning ticket -- which would also have to be either closest to or furthest from that lottery's stated-in-advance "goal number."
But the computer could just randomly select a string of keystrokes that meets the required level of redundancy that the funder paid for (the price probably being determined by auction), identify who made the keystrokes, and award the jackpot. Such approaches might be used by public institutions -- university libraries seeking to digitize their handwritten archives -- using tried-and-true state-lotto software. In such a case, typists presumably would no more need to check on the outcome than buyers of conventional lottery tickets do. But dark-web keystroke lotteries paying off in virtual currencies would have to be totally transparent. Again, a technical point that could be solved any number of ways.
-
i think the time for you keystroke lottery idea is over. Computers are more and more powerful and making intelligent decisions now. probably can already read lots of handwriting and will get better. Computerized translations also getting better and better. So why use all those people by turning typing into a lottery? It can only be transitional. Also, mechanical turk (amazon) already does a lot of typing. That will only get bigger. I'm not saying you lotteries can't work, just they will probably becoming uneconomical in the face of computers faster and cheaper.
-
i think the time for you keystroke lottery idea is over. Computers are more and more powerful and making intelligent decisions now. probably can already read lots of handwriting and will get better. Computerized translations also getting better and better. So why use all those people by turning typing into a lottery? It can only be transitional. Also, mechanical turk (amazon) already does a lot of typing. That will only get bigger. I'm not saying you lotteries can't work, just they will probably becoming uneconomical in the face of computers faster and cheaper.
Computers will definitely get more and more powerful. But how smart they will get depends on the problems they are given to solve. Consider self-driving cars. The more of them on the road the safer they and everyone else will be because they will communicate and coordinate with each other. In so doing they will achieve logical consensus on what to do in any given situation. Currently self-driving cars are so few in number they can only react, not coordinate. To coordinate requires a whole new and massive body of data to work with. So conventional cars must begin coordinating with each other too until the driver becomes an actual hindrance.
Developers of handwriting recognition and translation software will probably get around to developing KLs themselves as an economical data-gathering technique to improve their own product. They could scan a bunch of handwritten pages from a library archive (the way Google currently scans books); submit those pages to the world via a KL; then compare those results with the software they have already developed; and then tweak or butcher their algorithms to produce the KL result. KLs could turn out to be the essential tool in the development of genuinely effective handwriting recognition and translation: KLs as AI’s killer app.
If anything is transitional, it would be Amazon’s Mechanical Turk (and maybe Gigwalk.com). MT doesn’t invite workers by the millions as a KL would. It ruthlessly rejects them by the millions with its punishingly low pay for work done. Those who accept that pay must be hyper-alert to new postings, start on time and finish the task within deadline. In the case of typing, you will need a proofreader who will work for the same pay and deadline. You don’t get consensus that way. KLs offer massive redundancy to achieve the best consensus possible for a given document, page, paragraph, word, letter, symbol, or digit and (possibly) translation. With the lottery model, there is no deadline or schedule. Anyone could just log on and type as little or as much as they pleased, trying to achieve consensus with as many others as possible, because only a consensus-validated keystroke can win the payout. Conceivably a lot of work being done by MT would migrate to KLs. When you think about it, MT and Gigwalk (which probably could be turned into a lottery for its strolling data-collectors) are unimaginative, conventional attempts to crowdsource certain kinds of work. They would probably be a lot more successful if turned into a way for their participants to trade their time and bits of work for a Lotto-sized payoff, at least for certain kinds of tasks.
As far as I can see, without KLs the only way computerized handwriting recognition and translations could ever outperform people would be if handwriting were written out in the first place by another computer using a known algorithm; and if people spoke languages designed by computers too. Until then, or until KLs become a reality, the old printshop maxim will apple: You can have your job cheap, fast, and right — but only two out of three. Faster computers alone won’t change that; MT will not change that. But KLs probably could change that.
Worth a try anyway. Read my full posting on all this at: https://www.scribd.com/doc/254106778/Keystroke-Informant-Lotteries-A-Bitcoin-DAC-Killer-App (https://www.scribd.com/doc/254106778/Keystroke-Informant-Lotteries-A-Bitcoin-DAC-Killer-App)). It’s a lot more readable than the Slideshare version, which doesn’t include my description of a hypothetical Informant Lottery (currently on page 11).