Author Topic: Expanding the use cases for DACs - Using DPoS delegates as trusted third parties  (Read 5364 times)

0 Members and 1 Guest are viewing this topic.

Offline monsterer

Take a glance at the thread I've started if you get a chance - I'd like to hear what you think about mitigating collusion in a trustless environment?
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline CoinHoarder

  • Hero Member
  • *****
  • Posts: 660
  • In Cryptocoins I Trust
    • View Profile
I've started a thread here https://bitsharestalk.org/index.php?topic=10071 about the existing issues with trustless, decentralised poker - I'd love to get your input :)

I think you are misunderstanding me. What I am proposing would be a DAC in which everything is automated and trustless whenever possible. I have always thought a decentralized poker DAC should implement a "mental poker" framework to automate most of its functions and provide trustless games as in the shuffle, deal, and integrity of game play. I have done a lot of research on it and I believe it is possible, if a system to stop collusion and the inefficiencies of the mental poker framework can be worked out well enough to provide real time gaming. Due to the heavy cryptography and multi party computations mental poker requires, it is challenging to make it fast enough to where the gaming experience would be fast enough so that it is similar to current centralized poker offerings.

However, a Mental Poker framework can only do so much as far as making decentralized poker possible, and for collusion to be policed effectively requires human interference and judgement. Although an algorithm can be made to make it easier to spot collusion, it cannot be relied upon blindly to do its job properly and it can only be used as a crutch for humans to make the final decisions. In some cases something that would look like collusion to the algorithm may be random and honest game play, therefore you need people whom can make the final decisions.
« Last Edit: October 16, 2014, 11:15:42 am by CoinHoarder »
https://www.decentralized.tech/ -> Market Data, Portfolios, Information, Links, Reviews, Forums, Blogs, Etc.
https://www.cryptohun.ch/ -> Tradable Blockchain Asset PvP Card Game

Offline monsterer

I've started a thread here https://bitsharestalk.org/index.php?topic=10071 about the existing issues with trustless, decentralised poker - I'd love to get your input :)
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline monsterer

I imagine it's something along the lines of providing low/no collusion game types like heads up poker, fast fold poker, and multi-table tournaments.

No it's not - it's much more fundamental than that - but lets discuss it in the different thread, as its particular to poker, or at least proof of human :)
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline CoinHoarder

  • Hero Member
  • *****
  • Posts: 660
  • In Cryptocoins I Trust
    • View Profile
Quote
Some people (me included) are not happy with the amount of rake EBay and centralized poker sites charge, it could be done much cheaper as a DAC. Using DAC technology allows us to compete in many industries on a financial level by trimming the fat and automating most tasks. Although this is only a couple ideas, they are both some of the more apparent use cases where arbitration is needed. We are talking about Billion dollar industries.. I don't see why you would not be interested in going after them simply based on the fact that you can "trust" known companies and are "safe" with them... which is not a fact as companies go busto all the time.

I think decentralized poker is something that might not be as impossible as you think. I've got some ideas - if you'd like to discuss it, we should probably do so in a different thread.

I imagine it's something along the lines of providing low/no collusion game types like heads up poker, fast fold poker, and multi-table tournaments. Any other forms of poker need some form of arbitrators and reversible transactions due to collusion and bots. I don't see why we need to move the discussion, this is one of the biggest use cases for utilizing trusted third parties, so if you can prove me wrong that would be interesting. That would only leave decentralized market places as the biggest use case. Also there is competition in the industry (Nxt is crowd funding as I type but they are mute on technical details as they realize there will likely be competition), so maybe our conversation will stay more easily in between us/Bitshares community in this thread. ;)

My idea was originally to use a "cheating market place" (I call it a collusion dashboard or something like that in my paper). It is basically a decentralized market where people are paid a percentage of the rake and other fees to check for collusion and act as arbitrators. Since writing the paper I have kind of shifted my view on that as there are a lot of dynamics that go along with it. I think possibly giving delegates this role is better as then we can vote in people we trust to do a good job and people that are knowledgeable on the subject. It would be less subject to random manipulation as we have control of who the arbitrators are, and can vote them out if they act up. Anyways.. here is an in depth review of my idea. I wrote this a few months back, but I just made a few edits of some things I learned or changed my mind about. It still has the "collusion marketplace" in it instead of delegate arbitrators though. https://www.dropbox.com/s/hcadwuj6ji633lf/v0.03%20-%20The%20Future%20of%20Decentralized%20Online%20Poker%20As%20I%20See%20It.docx?dl=0

If you had something else in mind then please do share it, I would be interested in hearing it. I think any way you go about it, arbitrators will be needed in the form of delegates or "collusion market" participants.. at least for all game types and variants of poker. By only doing heads up games, rush poker, or multi table tournaments you are only servicing a small portion of the industry. I am more interested in a poker DAC that could serve all portions of the industry, as I think it will be more popular and profitable.

Ebay on the other hand, you've got a massive problem due to escrow of physical goods. Even ebay doesn't have a solution to this problem, which is why paypal has reversible charges. Again, the best you can do is to emulate the service they provide for a cheaper cost - unless you have all the delegates open up their houses to accept deliveries!
Yes, this is the point.. replicate Ebay/Paypal's business for cheaper. A DAC could do way less than 10% to 15% fees. The only time arbitration is needed is when a transaction is disputed by the buyer or the seller, otherwise everything is automated. I imagine if you also use some form of reputation like Ebay the amount of disputed transactions pales in comparison to mostly happy parties. By the use of a multi signature address arbitrators will not be needed if both parties are happy.
« Last Edit: October 15, 2014, 02:46:19 pm by CoinHoarder »
https://www.decentralized.tech/ -> Market Data, Portfolios, Information, Links, Reviews, Forums, Blogs, Etc.
https://www.cryptohun.ch/ -> Tradable Blockchain Asset PvP Card Game

Offline monsterer

Quote
Some people (me included) are not happy with the amount of rake EBay and centralized poker sites charge, it could be done much cheaper as a DAC. Using DAC technology allows us to compete in many industries on a financial level by trimming the fat and automating most tasks. Although this is only a couple ideas, they are both some of the more apparent use cases where arbitration is needed. We are talking about Billion dollar industries.. I don't see why you would not be interested in going after them simply based on the fact that you can "trust" known companies and are "safe" with them... which is not a fact as companies go busto all the time.

I think decentralised poker is something that might not be as impossible as you think. I've got some ideas - if you'd like to discuss it, we should probably do so in a different thread.

Ebay on the other hand, you've got a massive problem due to escrow of physical goods. Even ebay doesn't have a solution to this problem, which is why paypal has reversible charges. Again, the best you can do is to emulate the service they provide for a cheaper cost - unless you have all the delegates open up their houses to accept deliveries!
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline donkeypong

  • Hero Member
  • *****
  • Posts: 2329
    • View Profile
Fascinating discussion. There are possibilities here.

Offline CoinHoarder

  • Hero Member
  • *****
  • Posts: 660
  • In Cryptocoins I Trust
    • View Profile
... at the benefit of being much cheaper to use. Cost is the main benefit in my reasoning as to why people would want to use such services.

I think cost isn't such a major issue in the kinds of things we have centralised companies controlling now. For example, exchanges for trading one currency into another have pretty low fees already. Could it be cheaper? Yes, but I personally would want that at the risk of my money going away.
The risk of your money disappearing is still there, as it is inevitable with anything having to do with cryptocurrencies that are not stored properly and services you have no control over. One hack and it could all be gone a la Mt. Gox.

Being cheaper is one of the biggest selling points for cryptocurrencies/cryptoequity. This benefit can be spread through other industries as Bitshares is already doing, trusted third parties or not. I admit the decentralized multi token exchange was the poorest of use cases out of the 3 I mentioned in the OP, yet it could still be a profitable DAC.

Quote
Decentralized poker sites and market places are impossible without arbitrators. I would love if it was possible to do away with them, but that impossible. A system needs to be implemented that is sufficiently trustable and yet has the ability to mediate disputes and police the games.

I think poker in particular is a very interesting point and one that is really crying out for a trustless system - no one wants spyware installed on their machines which regularly take screenshots of the players's desktop to make sure they're not running bots (as you probably know, this is the online poker industry's current 'solution' to the bot problem).

It is impossible to have trustless poker as collusion and bots would be rampant. Trusted third parties are necessary to police the games. This is why I am proposing a system that is as trustless as possible meanwhile requiring third party intervention.

Quote
People seem to see the value in decentralized marketplaces, in which arbitrators are necessary for them to function. It is not that big of a stretch to implement the same schema in other businesses or industries if the benefits are sizable enough.

My point is, this is a solved problem. Companies provide this service already. What you're proposing will decrease the security in an industry which is currently completely full of scams, bad practices and bad publicity.

Cheers, Paul.

Some people (me included) are not happy with the amount of rake EBay and centralized poker sites charge, it could be done much cheaper as a DAC. Using DAC technology allows us to compete in many industries on a financial level by trimming the fat and automating most tasks. Although this is only a couple ideas, they are both some of the more apparent use cases where arbitration is needed. We are talking about Billion dollar industries.. I don't see why you would not be interested in going after them simply based on the fact that you can "trust" known companies and are "safe" with them... which is not a fact as companies go busto all the time.
« Last Edit: October 15, 2014, 01:57:58 am by CoinHoarder »
https://www.decentralized.tech/ -> Market Data, Portfolios, Information, Links, Reviews, Forums, Blogs, Etc.
https://www.cryptohun.ch/ -> Tradable Blockchain Asset PvP Card Game

Offline monsterer

... at the benefit of being much cheaper to use. Cost is the main benefit in my reasoning as to why people would want to use such services.

I think cost isn't such a major issue in the kinds of things we have centralised companies controlling now. For example, exchanges for trading one currency into another have pretty low fees already. Could it be cheaper? Yes, but I personally would want that at the risk of my money going away.

Quote
Decentralized poker sites and market places are impossible without arbitrators. I would love if it was possible to do away with them, but that impossible. A system needs to be implemented that is sufficiently trustable and yet has the ability to mediate disputes and police the games.

I think poker in particular is a very interesting point and one that is really crying out for a trustless system - no one wants spyware installed on their machines which regularly take screenshots of the players's desktop to make sure they're not running bots (as you probably know, this is the online poker industry's current 'solution' to the bot problem).

Quote
People seem to see the value in decentralized marketplaces, in which arbitrators are necessary for them to function. It is not that big of a stretch to implement the same schema in other businesses or industries if the benefits are sizable enough.

My point is, this is a solved problem. Companies provide this service already. What you're proposing will decrease the security in an industry which is currently completely full of scams, bad practices and bad publicity.

Cheers, Paul.
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline CoinHoarder

  • Hero Member
  • *****
  • Posts: 660
  • In Cryptocoins I Trust
    • View Profile
Everything I said was applicable. Excuse my tone in the last post, but I am annoyed that all of my ideas being ignored or immediately shut down without any discussion... this is just the latest example. (https://bitsharestalk.org/index.php?topic=9876.0)

Whereas registered companies can be sued/jailed and victims can sometimes be recompensated with success, decentralized companies can trim the fat and provide a cheaper service. It is a give and take situation.. some people will prefer to pay much higher fees with the "security" a known company provides and some will be willing to take a risk and save money by using a cheaper service. It comes down to how risk adverse or risk prone someone is.

There is a big difference between trustless and trusted, as you know. These things are binary opposites. What you're proposing is a middle ground solution which will be more vulnerable to exploitation than the established method of handling trust.

.... at the benefit of being much cheaper to use. Cost is the main benefit in my reasoning as to why people would want to use such services. Decentralized poker sites and market places are impossible without arbitrators. I would love if it was possible to do away with them, but that impossible. A system needs to be implemented that is sufficiently trustable and yet has the ability to mediate disputes and police the games.

Yes, DPOS is this system, but the asking delegates to produce blocks and asking them to handle critically sensitive decisions and processes using their own judgement are two different things.
Yes they are different things, but trust is still involved in both. DPoS is not "trustless" either, it is trusted in that the stake holders trust the majority of delegates to not act in a nefarious manner. Handling critically sensitive decisions and processes will need to work in the same way.

By providing incentive for people to do the right thing and allowing the ability for stake holders to remove their power if they do not, risk is mitigated in the best way possible. Similar to how employees with jobs which require a company's trust are vetted via interviews (delegate campaigning), paid (delegate pay) to remain trustworthy, and can be fired if they are not (delegate voting.) That along with having a majority rules voting system is the key to being able to give psuedo-anonymous trusted decision making to people. You aren't assuming that all of them will be trustworthy, just that the majority of them will be.

Over time users will trust the network more and more as long as nothing bad happens, similar to how companies gain trust. This is a really stupid/random example, but if someone randomly stops you while you're walking down the street and asks you if you want to buy a hot dog, you will probably say no. However, over time you see that same person asking other people the same question and them taking him up on it (the risk prone folk). None of them die or get sick, the hot dog is cheaper than what you could get elsewhere, and others say they are delicious, then you may go out on a limb and try a hot dog one day. After all of that, given that you really like hot dogs and that you don't know his name, address, etc, would you still not trust the random hot dog guy??

I just don't see the benefits you list are worth the inherent risk, especially in a nascent and delicate ecosystem which aims to gain acceptance from the outside world.
People seem to see the value in decentralized marketplaces, in which arbitrators are necessary for them to function. It is not that big of a stretch to implement the same schema in other businesses or industries if the benefits are sizable enough. I believe if a service is much cheaper than its competitors it will be able to gain a sizable market share... if the service they provide is good and they are proven trustworthy over time. I don't expect any service like this to be released and immediately crush the competition (Ebay/PokerStars), but over time I think it could gain a sizable market share if everything worked as advertised and could possibly be very lucrative for shareholders.
« Last Edit: October 14, 2014, 04:01:52 pm by CoinHoarder »
https://www.decentralized.tech/ -> Market Data, Portfolios, Information, Links, Reviews, Forums, Blogs, Etc.
https://www.cryptohun.ch/ -> Tradable Blockchain Asset PvP Card Game

Offline CoinHoarder

  • Hero Member
  • *****
  • Posts: 660
  • In Cryptocoins I Trust
    • View Profile
+5% Great post. I'm really liking the idea of a poker DAC. I think it's a natural fit.

How does PokerStars handle collusion?

1. They can see all hole cards to analyze hands. Being able to see hole cards is necessary to be able to combat collusion effectively in game types where collusion is more profitable/likely. I am suggesting the idea that all hole cards be publicized on a delay, similar to how live poker tournament streaming and TV shows do it. This is different to the current online poker offerings, but it is the only way I could think of that would be in a "trustless" manner.

1a. Otherwise if only delegates could view hole cards they could gain an unfair advantage while playing or relay that information to other players whom would gain an unfair advantage against players that didn't have that information. It will take some getting used to for poker players, as in most non-televised poker you can only see what your opponent had if it goes to showdown or they show you their cards. However, I think this is something that poker players could deal with if the rake is much cheaper than current online poker sites.. which I think it very well could be. It puts more of an emphasis on being game theory optimal, which is where most poker players think poker strategy is headed in the near future anyways. You will still get dealt 2 cards every hand with a random deal, and have chances to outplay the other players as people will still make poor/exploitable decisions.. people are not perfect.

2. They have full control over the "poker chips" and cash in/cash out gateways, so they can recompensate people whom were cheated and confiscate funds from people whom did the cheating.

2a. I think we could give Delegates control over the "poker chips" via Delegate controlled multi-signature addresses with which they can reverse payments and control the cash in/cash out gateways. For someone to play poker, they would need to deposit to a multi-signature address and then they will be given "credit" with which they can play in games.. similar to a poker site balance where you play with PokerStarsDollars or FullTiltDollars.. not actual dollars. When someone requests a cash out their play history should be checked to see if there is evidence of collusion or foul play, thus the gateways are controlled.

3. They use identity verification to prevent users from registering multiple accounts.

3a. BM stated the FollowMyVote DAC will have this ability, so we at least know it is possible. We could tie in the identity verification feature from this DAC or implement our own for a Poker DAC.

4. With the ability to see hole cards, reverse transactions, control the payment gateways, and stop multi-accounting, you then have all the tools needed to combat collusion.

4a. Algorithms should be made to detect every kind of collusion that is possible to highlight suspicious players, along with a mechanism for players to report other players for suspected foul play. People then need to analyze the hands and act as arbitrators, which is where I was suggesting delegates be used in a majority rules type manner. In the paper I posted above, I came up with another way this can be done but the dynamics would need to be fine tuned. You could create a "collusion dashboard/marketplace" where suspicious players/hands as pointed out by the algorithm and reported players/hands show up and allow the players to police themselves. You can give players incentive to police the games by offering a percentage of the rake for their participation in doing so, and come up with some sort of voting mechanism which would determine if colluding took place or not.

If you read the link of the paper I wrote up thread (which needs to be updated as I have some improvements in mind), you can get an idea of what the collusion algorithm will be looking for and how it can help in the detection of collusion. Poker sites don't release specific information on how they detect collusion and bots, but I am certain they have an algorithm similar to what I described in the paper to make it easier to do so (along with having a team that reviews hand histories and allowing players to report other players they suspect of colluding.)
« Last Edit: October 14, 2014, 04:06:37 pm by CoinHoarder »
https://www.decentralized.tech/ -> Market Data, Portfolios, Information, Links, Reviews, Forums, Blogs, Etc.
https://www.cryptohun.ch/ -> Tradable Blockchain Asset PvP Card Game

Offline monsterer

Everything I said was applicable. Excuse my tone in the last post, but I am annoyed that all of my ideas being ignored or immediately shut down without any discussion... this is just the latest example. (https://bitsharestalk.org/index.php?topic=9876.0)

Whereas registered companies can be sued/jailed and victims can sometimes be recompensated with success, decentralized companies can trim the fat and provide a cheaper service. It is a give and take situation.. some people will prefer to pay much higher fees with the "security" a known company provides and some will be willing to take a risk and save money by using a cheaper service. It comes down to how risk adverse or risk prone someone is.

There is a big difference between trustless and trusted, as you know. These things are binary opposites. What you're proposing is a middle ground solution which will be more vulnerable to exploitation than the established method of handling trust. Yes, DPOS is this system, but the asking delegates to produce blocks and asking them to handle critically sensitive decisions and processes using their own judgement are two different things.

I just don't see the benefits you list are worth the inherent risk, especially in a nascent and delicate ecosystem which aims to gain acceptance from the outside world.
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline luckybit

  • Hero Member
  • *****
  • Posts: 2921
    • View Profile
  • BitShares: Luckybit
It's funny how everyone trusts Satoshi even though he is anonymous.

People are buying Bitcoins right now and at any time Satoshi could crash the price to $0 just by selling his stash.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline CoinHoarder

  • Hero Member
  • *****
  • Posts: 660
  • In Cryptocoins I Trust
    • View Profile
Mmk, in that case why are you using cryptocurrencies? I guess you analyzed each line of BTSX and BTCs code to make sure you are not being scammed? If Bitcoin or BTSX fails there is no recourse. If delegates collude to double spend there is no recourse. If someone 51%s

Is this your argument for why delegates are better than registered companies for handling centralised trust, or are you changing the subject to avoid the obvious conclusion?

Everything I said was applicable. Excuse my tone in the last post, but I am annoyed that all of my ideas being ignored or immediately shut down without any discussion... this is just the latest example. (https://bitsharestalk.org/index.php?topic=9876.0)

Whereas registered companies can be sued/jailed and victims can sometimes be recompensated with success, decentralized companies can trim the fat and provide a cheaper service. It is a give and take situation.. some people will prefer to pay much higher fees with the "security" a known company provides and some will be willing to take a risk and save money by using a cheaper service. It comes down to how risk adverse or risk prone someone is.

Overtime if the delegates were proven trustworthy and the service got good reviews, the service could become popular and used for bigger dollar amounts or higher volume. Again, trust in a pseudo anonymous community is earned not given. Some people will be skeptical and stay away.. others will want to save money and use the service. Obviously if the delegates are scamming people then the service will become unpopular very quickly and they will lose the income stream that being a delegate would provide. Providing incentive for people to be honest is key.
« Last Edit: October 14, 2014, 02:26:27 pm by CoinHoarder »
https://www.decentralized.tech/ -> Market Data, Portfolios, Information, Links, Reviews, Forums, Blogs, Etc.
https://www.cryptohun.ch/ -> Tradable Blockchain Asset PvP Card Game

Offline Method-X

  • Hero Member
  • *****
  • Posts: 1131
  • VIRAL
    • View Profile
    • Learn to code
  • BitShares: methodx
+5% Great post. I'm really liking the idea of a poker DAC. I think it's a natural fit.

How does PokerStars handle collusion?