On Tuesday, July 24, 2018, 10:21:40 PM PDT, juan <juan.g71@gmail.com> wrote: https://www.ccn.com/first-assassination-markets-appear-on-gambling-platform-... Well, you've got my attention! What can I say? Never expected that Betty White would be "on the list", so to speak. Jim Bell
On 07/24/2018 10:55 PM, jim bell wrote:
On Tuesday, July 24, 2018, 10:21:40 PM PDT, juan <juan.g71@gmail.com> wrote: https://www.ccn.com/first-assassination-markets-appear-on-gambling-platform-...
Well, you've got my attention! What can I say? Never expected that Betty White would be "on the list", so to speak. Jim Bell
Hey! Props to you, Jim! Very cool! But still, in a gambling context, this is more like typical dead pool betting, which has been a thing like forever.
On Wed, Jul 25, 2018 at 4:21 AM, Mirimir <mirimir@riseup.net> wrote:
On 07/24/2018 10:55 PM, jim bell wrote:
On Tuesday, July 24, 2018, 10:21:40 PM PDT, juan <juan.g71@gmail.com> wrote: https://www.ccn.com/first-assassination-markets-appear-on-gambling-platform-...
Well, you've got my attention! What can I say? Never expected that Betty White would be "on the list", so to speak. Jim Bell
Hey! Props to you, Jim! Very cool!
But still, in a gambling context, this is more like typical dead pool betting, which has been a thing like forever.
https://twitter.com/AugurProject https://kryptous.com/the-first-augur-assassination-markets-have-arrived/ https://twitter.com/search?q=augur%20assassination https://www.reddit.com/r/Augur/comments/91a54w/what_to_do_about_assassinatio... https://twitter.com/matt_odell/status/1017025681136455680 https://twitter.com/AugurProject/status/1021631296265768960 https://twitter.com/peterktodd/status/1021634814401687552 Note IPFS links vs censorship https://twitter.com/lawmaster/status/1021675034761486337 They're the first on Augur, and there have been about four short lived prior instances of such prediction markets (whether faux, in devel, or production) in other namespaces such as onion and i2p. "You’re not a crypto OG if you’re not using the assassination market in Augur for your life insurance needs."
On Wednesday, July 25, 2018, 1:22:09 AM PDT, Mirimir <mirimir@riseup.net> wrote: On 07/24/2018 10:55 PM, jim bell wrote:
On Tuesday, July 24, 2018, 10:21:40 PM PDT, juan <juan.g71@gmail.com> wrote: https://www.ccn.com/first-assassination-markets-appear-on-gambling-platform-...
Well, you've got my attention! What can I say? Never expected that Betty White would be "on the list", so to speak. Jim Bell
Hey! Props to you, Jim! Very cool!
But still, in a gambling context, this is more like typical dead pool betting, which has been a thing like forever.
Quite true. But as Nick Szabo said, http://unenumerated.blogspot.com/2015/05/small-game-fallacies.html "A sufficiently large market predicting an individual's death is also, necessarily, an assassination market, and similarly other "prediction" markets are also act markets, changing incentives to act outside that market to bring about the predicted events." And he was right. It is the size, or lack of it, which is the primary thing which has prevented 'dead pool' betting from turning into AP. The other thing necessary to turn such a prediction market into a classic AP system https://cryptome.org/ap.htm is good, i.e. perfect, anonymity of bets and payouts: So nobody else, other than than the bettor himself, knows who he is, and he knows he can collect the payout anonymously without risk. I don't know if this Ethereum/Augur system has such anonymity, but I suppose we'll find out soon enough. It wouldn't be particularly useful if it didn't.× The main advantage added by Ethereum/Augur, I assume, is that it acts as a distributed computer system that, being "everywhere", ends up being "nowhere": It cannot be shut down by government authority. Jim Bell×
Another practical requirement is leveraged betting. This enables those intending to act outside of these online venues but with limited financial means to wager in such ways as their payouts are greatly disproportionate to their wagers. On Wed, Jul 25, 2018, 9:46 AM jim bell <jdb10987@yahoo.com> wrote:
On Wednesday, July 25, 2018, 1:22:09 AM PDT, Mirimir <mirimir@riseup.net> wrote:
On 07/24/2018 10:55 PM, jim bell wrote:
On Tuesday, July 24, 2018, 10:21:40 PM PDT, juan <
juan.g71@gmail.com> wrote: https://www.ccn.com/first-assassination-markets-appear-on-gambling-platform-...
Well, you've got my attention! What can I say? Never expected that
Betty White would be "on the list", so to speak.
Jim Bell
Hey! Props to you, Jim! Very cool!
But still, in a gambling context, this is more like typical dead pool betting, which has been a thing like forever.
Quite true. But as Nick Szabo said, http://unenumerated.blogspot.com/2015/05/small-game-fallacies.html
"A sufficiently large market predicting an individual's death is also, necessarily, an assassination market, and similarly other "prediction" markets are also *act *markets, changing incentives to act outside that market to bring about the predicted events."
And he was right. It is the size, or lack of it, which is the primary thing which has prevented 'dead pool' betting from turning into AP.
The other thing necessary to turn such a prediction market into a classic AP system https://cryptome.org/ap.htm is good, i.e. perfect, anonymity of bets and payouts: So nobody else, other than than the bettor himself, knows who he is, and he knows he can collect the payout anonymously without risk. I don't know if this Ethereum/Augur system has such anonymity, but I suppose we'll find out soon enough. It wouldn't be particularly useful if it didn't. ×
The main advantage added by Ethereum/Augur, I assume, is that it acts as a distributed computer system that, being "everywhere", ends up being "nowhere": It cannot be shut down by government authority.
Jim Bell ×
Ah yes, excellent point, and quite true. I'd forgotten... Which is one reason this new system (I haven't looked at it yet) might not "work", at least it won't bring us an AP-type system. If the question is, "Will [fill in the blank with a name] be alive at the end of the year 2018?", then everybody who bets "No" is essentially on the same footing. They'd all share, proportionately, in the reward. A person who predicted death before 2019 is on the same footing as everyone else who made the same prediction. Thus, there is little incentive to "adjust the odds"acting outside of the game. If, instead, the system is designed to allow bettors to 'predict' the date and perhaps even the time of the death, that system carefully excludes most other 'competing' bettors: If one bettor predicted December 1, and another predicted December 3, if the actual death occurred on December 1 all other bettors would lose the 'pot', and only those who predicted December 1 would share the 'pot'. This greatly leverages the betting. Which is why death-on-a-certain-date was my intended model when I wrote AP. Jim Bell On Wednesday, July 25, 2018, 9:54:13 AM PDT, Steven Schear <schear.steve@gmail.com> wrote: Another practical requirement is leveraged betting. This enables those intending to act outside of these online venues but with limited financial means to wager in such ways as their payouts are greatly disproportionate to their wagers. On Wed, Jul 25, 2018, 9:46 AM jim bell <jdb10987@yahoo.com> wrote: On Wednesday, July 25, 2018, 1:22:09 AM PDT, Mirimir <mirimir@riseup.net> wrote: On 07/24/2018 10:55 PM, jim bell wrote:
On Tuesday, July 24, 2018, 10:21:40 PM PDT, juan <juan.g71@gmail.com> wrote: https://www.ccn.com/first-assassination-markets-appear-on-gambling-platform-...
Well, you've got my attention! What can I say? Never expected that Betty White would be "on the list", so to speak. Jim Bell
Hey! Props to you, Jim! Very cool!
But still, in a gambling context, this is more like typical dead pool betting, which has been a thing like forever.
Quite true. But as Nick Szabo said, http://unenumerated.blogspot.com/2015/05/small-game-fallacies.html "A sufficiently large market predicting an individual's death is also, necessarily, an assassination market, and similarly other "prediction" markets are also act markets, changing incentives to act outside that market to bring about the predicted events." And he was right. It is the size, or lack of it, which is the primary thing which has prevented 'dead pool' betting from turning into AP. The other thing necessary to turn such a prediction market into a classic AP system https://cryptome.org/ap.htm is good, i.e. perfect, anonymity of bets and payouts: So nobody else, other than than the bettor himself, knows who he is, and he knows he can collect the payout anonymously without risk. I don't know if this Ethereum/Augur system has such anonymity, but I suppose we'll find out soon enough. It wouldn't be particularly useful if it didn't.× The main advantage added by Ethereum/Augur, I assume, is that it acts as a distributed computer system that, being "everywhere", ends up being "nowhere": It cannot be shut down by government authority. Jim Bell×
My concept is similar to current financial futures except, for example, the "put" buyer only generically specifies day, week or month within the full wager period. The exact day, week or month is encrypted and can only be revealed by the buyer should their prediction prove true in order to collect. The leverage on the wager will vary by 365, 52 or 12, for example, for bets based on a day, week or month, respectively. On Wed, Jul 25, 2018, 10:35 AM jim bell <jdb10987@yahoo.com> wrote:
Ah yes, excellent point, and quite true. I'd forgotten... Which is one reason this new system (I haven't looked at it yet) might not "work", at least it won't bring us an AP-type system. If the question is, "Will [fill in the blank with a name] be alive at the end of the year 2018?", then everybody who bets "No" is essentially on the same footing. They'd all share, proportionately, in the reward. A person who predicted death before 2019 is on the same footing as everyone else who made the same prediction. Thus, there is little incentive to "adjust the odds"acting outside of the game.
If, instead, the system is designed to allow bettors to 'predict' the date and perhaps even the time of the death, that system carefully excludes most other 'competing' bettors: If one bettor predicted December 1, and another predicted December 3, if the actual death occurred on December 1 all other bettors would lose the 'pot', and only those who predicted December 1 would share the 'pot'. This greatly leverages the betting. Which is why death-on-a-certain-date was my intended model when I wrote AP.
Jim Bell
On Wednesday, July 25, 2018, 9:54:13 AM PDT, Steven Schear < schear.steve@gmail.com> wrote:
Another practical requirement is leveraged betting. This enables those intending to act outside of these online venues but with limited financial means to wager in such ways as their payouts are greatly disproportionate to their wagers.
On Wed, Jul 25, 2018, 9:46 AM jim bell <jdb10987@yahoo.com> wrote:
On Wednesday, July 25, 2018, 1:22:09 AM PDT, Mirimir <mirimir@riseup.net> wrote:
On 07/24/2018 10:55 PM, jim bell wrote:
On Tuesday, July 24, 2018, 10:21:40 PM PDT, juan <
juan.g71@gmail.com> wrote: https://www.ccn.com/first-assassination-markets-appear-on-gambling-platform-...
Well, you've got my attention! What can I say? Never expected that
Betty White would be "on the list", so to speak.
Jim Bell
Hey! Props to you, Jim! Very cool!
But still, in a gambling context, this is more like typical dead pool betting, which has been a thing like forever.
Quite true. But as Nick Szabo said, http://unenumerated.blogspot.com/2015/05/small-game-fallacies.html
"A sufficiently large market predicting an individual's death is also, necessarily, an assassination market, and similarly other "prediction" markets are also *act *markets, changing incentives to act outside that market to bring about the predicted events."
And he was right. It is the size, or lack of it, which is the primary thing which has prevented 'dead pool' betting from turning into AP.
The other thing necessary to turn such a prediction market into a classic AP system https://cryptome.org/ap.htm is good, i.e. perfect, anonymity of bets and payouts: So nobody else, other than than the bettor himself, knows who he is, and he knows he can collect the payout anonymously without risk. I don't know if this Ethereum/Augur system has such anonymity, but I suppose we'll find out soon enough. It wouldn't be particularly useful if it didn't. ×
The main advantage added by Ethereum/Augur, I assume, is that it acts as a distributed computer system that, being "everywhere", ends up being "nowhere": It cannot be shut down by government authority.
Jim Bell ×
On 07/25/2018 10:34 AM, jim bell wrote:
Ah yes, excellent point, and quite true. I'd forgotten... Which is one reason this new system (I haven't looked at it yet) might not "work", at least it won't bring us an AP-type system. If the question is, "Will [fill in the blank with a name] be alive at the end of the year 2018?", then everybody who bets "No" is essentially on the same footing. They'd all share, proportionately, in the reward. A person who predicted death before 2019 is on the same footing as everyone else who made the same prediction. Thus, there is little incentive to "adjust the odds"acting outside of the game.
OK, so the prediction is "dead by 2019-01-01 00:00:00.000000000-00:00". By late 2018, the odds against that are ramping up. And so is the security level, I suppose. But anyway, at some time, the assassin (or whoever's in control) places a huge bet. And watches the odds, and monitors security etc. Then the assassin completes the job, and wins big. Unless captured or killed, of course. Others who didn't fold also win big, but hey.
If, instead, the system is designed to allow bettors to 'predict' the date and perhaps even the time of the death, that system carefully excludes most other 'competing' bettors: If one bettor predicted December 1, and another predicted December 3, if the actual death occurred on December 1 all other bettors would lose the 'pot', and only those who predicted December 1 would share the 'pot'. This greatly leverages the betting. Which is why death-on-a-certain-date was my intended model when I wrote AP.
That also works. But much better is Steven Schear's proposal (next post in thread) to encrypt more-or-less exact predictions, with leverage depending on precision of the prediction. That mitigates against increasing security when nearing the end of the prediction period.
Jim Bell
On Wednesday, July 25, 2018, 9:54:13 AM PDT, Steven Schear <schear.steve@gmail.com> wrote:
Another practical requirement is leveraged betting. This enables those intending to act outside of these online venues but with limited financial means to wager in such ways as their payouts are greatly disproportionate to their wagers.
<SNIP>
On Wed, 25 Jul 2018 16:44:59 +0000 (UTC) jim bell <jdb10987@yahoo.com> wrote:
The main advantage added by Ethereum/Augur, I assume, is that it acts as a distributed computer system that, being "everywhere", ends up being "nowhere": It cannot be shut down by government authority.
yeah, that part of the AP system seems to be in place. Although from a quick look at a couple of articles, seems to me that augur can be (easily?) censored "Augur currently has four options to resolve a bet: 1. yes 2. no 3. indeterminate 4. unethical" “The Augur Reporter community has a powerful tool in their ability to mark a market as ‘invalid,'” (so there's a 'community' of scumbags who go around censoring stuff. Not sure if augur is a serious system like the permisionless, censorship-resistant bitcoin network, or some sort of toy owned by a bunch of oportunists - seems more like the later)
The other thing necessary to turn such a prediction market into a classic AP system https://cryptome.org/ap.htm is good, i.e. perfect, anonymity of bets and payouts: So nobody else, other than than the bettor himself, knows who he is, and he knows he can collect the payout anonymously without risk. I don't know if this Ethereum/Augur system has such anonymity, but I suppose we'll find out soon enough. It wouldn't be particularly useful if it didn't.×
well augur runs on top of ethereum(probably untrustworthy) which in turn uses the 'internet'(biggest spy system on the planet), and then we the people live in a global surveillance state, and surveillance is especially bad in the so called 'developed' world. so the anonimity required mighy be lacking somewhat...
On 07/25/2018 09:44 AM, jim bell wrote:
On Wednesday, July 25, 2018, 1:22:09 AM PDT, Mirimir <mirimir@riseup.net> wrote:
On 07/24/2018 10:55 PM, jim bell wrote:
On Tuesday, July 24, 2018, 10:21:40 PM PDT, juan <juan.g71@gmail.com> wrote: https://www.ccn.com/first-assassination-markets-appear-on-gambling-platform-...
Well, you've got my attention! What can I say? Never expected that Betty White would be "on the list", so to speak. Jim Bell
Hey! Props to you, Jim! Very cool!
But still, in a gambling context, this is more like typical dead pool betting, which has been a thing like forever.
Quite true. But as Nick Szabo said, http://unenumerated.blogspot.com/2015/05/small-game-fallacies.html
"A sufficiently large market predicting an individual's death is also, necessarily, an assassination market, and similarly other "prediction" markets are also act markets, changing incentives to act outside that market to bring about the predicted events."
And he was right. It is the size, or lack of it, which is the primary thing which has prevented 'dead pool' betting from turning into AP.
OK, so I'm not setup for Augur. But the example in the article was "Will Donald Trump (President of the USA) be killed at any point during 2018". So let's say that someone kills him on October 31, at 11:10 pm EST, wearing a Michael Myers mask. How does does the killer profit on Augur?
The other thing necessary to turn such a prediction market into a classic AP system https://cryptome.org/ap.htm is good, i.e. perfect, anonymity of bets and payouts: So nobody else, other than than the bettor himself, knows who he is, and he knows he can collect the payout anonymously without risk. I don't know if this Ethereum/Augur system has such anonymity, but I suppose we'll find out soon enough. It wouldn't be particularly useful if it didn't.×
The main advantage added by Ethereum/Augur, I assume, is that it acts as a distributed computer system that, being "everywhere", ends up being "nowhere": It cannot be shut down by government authority.
Jim Bell×
Digging a little, I see that Etherium wallets sync using UDP, so Tor is out. So it's nested VPN chains, plus maybe VPN via Tor. Or the old open WiFi thing. I suppose that one could use a hosted wallet, but that seems iffy for AP ;) So I'm not optimistic.
If, after operating for some time, a PM supporting AP is widely believed to be anonymous for the wagers and and payouts, those who decide to fulfill AP bets might not even consider survival after the act (especially if the target is political and they are terminally ill people with the required trade-craft and/or access to the targets). These "Zombie Patriots" could use the ease of the AP crypto asset payout ownership by transferring the BIP39 mnemonic to a relative or friend (perhaps using a distributed paper 2-of-3 recovery protocol, https://coinstop.io/products/safewords-by-coin-storage-guru). This is somewhat similar to how Palestinian authorities offered "insurance" to the families of suicide bombers. For a cypherpunk spy-thriller novel, dedicated to Jim Bell's AP vision (and yours truly), using Zombie Patriots as a major plot point, see Graynet by DS Kane, https://www.amazon.com/GrayNet-Book-Spies-DS-Kane/dp/0996059199/ On Wed, Jul 25, 2018 at 10:48 AM, Mirimir <mirimir@riseup.net> wrote:
And he was right. It is the size, or lack of it, which is the primary
thing which has prevented 'dead pool' betting from turning into AP.
OK, so I'm not setup for Augur. But the example in the article was "Will Donald Trump (President of the USA) be killed at any point during 2018". So let's say that someone kills him on October 31, at 11:10 pm EST, wearing a Michael Myers mask. How does does the killer profit on Augur?
While ZKP privacy coins like say Zcash are perhaps likely leading cryptocurrency space regarding strong cryptographic private anonymous coin functions. Essentially all coins don't do much in area of secure network comms, both their own network packets / protocols, and the nets they run over. Few advertise / support compatibility with anonymous overlay networks, use TLS, etc. Some like Zencash-ZEN, and onion oriented ones, do have some features there. And global adoption isn't there yet to spend payouts into without exchanges, which are also weak. And all of today's major in production anonymity networks aren't really accepted as doing well against global / state level top secret passive / active adversaries. For influencing that dog / owner across the street that barks all day... yeah maybe Bitcoin-BTC over some toynet. But stronger coins and nets still have yet to be proven out and developed before taking on big game. Yet there will always be those randoms who come not to collect money, but fame... or quietly, collecting the predictions themselves... as bloody QR skins hung on a line in some shack deep in the jungle...
On Sat, Jul 28, 2018 at 02:44:23AM -0400, grarpamp wrote:
While ZKP privacy coins like say Zcash are perhaps likely leading cryptocurrency space regarding strong cryptographic private anonymous coin functions.
Essentially all coins don't do much in area of secure network comms, both their own network packets / protocols, and the nets they run over.
DCs should not be providing their own anonymising network layers - that ought really be a separate layer, and separate dev project, as such separation encourages sane cross-project cooperation.
Few advertise / support compatibility with anonymous overlay networks, use TLS, etc. Some like Zencash-ZEN, and onion oriented ones, do have some features there.
Main thing at the moment is that TOR is very popular, and this CIA and NSA funded team staunchly 'fails' to be funded for: - UDP based transport - chaff fill module/ option
And global adoption isn't there yet to spend payouts into without exchanges, which are also weak.
And all of today's major in production anonymity networks aren't really accepted as doing well against global / state level top secret passive / active adversaries.
For someone with inspiration and Java programming knowledge, Azureus Vuze is quite a solid (and amazingly featureful) base of code from which to begin some work. A quick jewgle brings up only clients, proxies and a Tor process controller, all of which rely on the existing core Tor code, so a reimplementation for that language if that's your style. Otherwise go hardcore and delve straight into C, or reimplement in some other language. Either way, "it's just a little network protocol".
For influencing that dog / owner across the street that barks all day... yeah maybe Bitcoin-BTC over some toynet.
But stronger coins and nets still have yet to be proven out and developed before taking on big game.
Aye.
Yet there will always be those randoms who come not to collect money, but fame... or quietly, collecting the predictions themselves... as bloody QR skins hung on a line in some shack deep in the jungle...
participants (6)
-
grarpamp
-
jim bell
-
juan
-
Mirimir
-
Steven Schear
-
Zenaan Harkness