Can You Write a Sure-fire Algorithm to Stop Bots?

[Tatarana]

Hey JJ, u dont need to be rude, I'm just putting some ideas on the table.I'm not advocating any one these ideas. Running bad lately ?

Some other ideas :

7 - Segregated Fish Tanks : Players are automaticaly seated at tables where the others players are at same skill level. Bots will eventually play each other.
8 - new poker variants
9 - No screennames and moving tables: You will be moved to other table after, lets say, 20 hands.You will be always playing against someone u dont have enough stats. I think, but I'm not sure , that a bot cant beat a good human player without a profile. A tie is the most probable outcome.

What we do understand today as multiplayer games may be obsolete in 5 years and new enviroments and devices could open new opprtunities to poker.

I think Public profiles may help to preserve fish from the big grinder bot. The fish must know that they are playing against someone who plays X hours/day and sees % of flops etc ...

About the item #4, Bots are VERY sensitive to collusion. Either explicit or implicit. We just need to say : " Hey guys player X is a bot" to make them a losing proposition.

I do agree that the grinder friendly online poker will be dead in a very near future.

[Josem]

it seems to me that bots (ie, computers) are very good at numerical calculations and algorithms.

thus, if i was going to come up with an bot detection system, it'd need to be outside the realm of calculations and algorithms and the like.

[jjshabado]

[ QUOTE ]
Hey JJ, u dont need to be rude, I'm just putting some ideas on the table.I'm not advocating any one these ideas. Running bad lately ?

Some other ideas :

7 - Segregated Fish Tanks : Players are automaticaly seated at tables where the others players are at same skill level. Bots will eventually play each other.
8 - new poker variants
9 - No screennames and moving tables: You will be moved to other table after, lets say, 20 hands.You will be always playing against someone u dont have enough stats. I think, but I'm not sure , that a bot cant beat a good human player without a profile. A tie is the most probable outcome.

What we do understand today as multiplayer games may be obsolete in 5 years and new enviroments and devices could open new opprtunities to poker.

I think Public profiles may help to preserve fish from the big grinder bot. The fish must know that they are playing against someone who plays X hours/day and sees % of flops etc ...

About the item #4, Bots are VERY sensitive to collusion. Either explicit or implicit. We just need to say : " Hey guys player X is a bot" to make them a losing proposition.

I do agree that the grinder friendly online poker will be dead in a very near future.

[/ QUOTE ]


Sorry if I came across as rude. I think this is a good debate, and I'm not trying to be insulting. And interestingly enough I don't really play online. Basically because I know I'm not good enough to be any more than break even. I find live play more fun, and a lot easier to make money.

7. This would kill online poker. Poker is a giant pyramid scheme. There's a few great players who make money off of everybody. Then the next level makes money off of lots of people but loses it to the people above them, and down we go. At the bottom are the fish. The people that either know they lose money and don't care because they enjoy it, or the people that convince themselves that they're winning players but they aren't. (Note: This is another major threat to online poker. Its hard to convince yourself you're a winning player when you're not online. In real life its way easier to lose track of your actual wins/losses). On top of that, you could easily make your bot play bad enough that it would be the best of the bad players. If you can't tell the difference between humans and bots, you can't separate them into separate tables.

8. You'd have to convince players to play these new games. There's no reason to think that people could learn these games faster than bots (and plenty of reasons to believe bots could adjust faster).

9. Bots will definitely become good enough to adjust to the optimal strategy against unknown players.

Again, I'm not trying to be rude, this is just my writing style when I really believe what I'm writing. I could be wrong, but I see almost no situation where online poker can stay ahead of bots indefinitely.

[jjshabado]

[ QUOTE ]
[ QUOTE ]

All of your solutions rely on Bots not becoming really good at mimicking people, or on technology that would be "tamper proof". Its more than naive (its stupid) too make either of those assumptions. The reality is in time AI will be advanced enough that bots will be able to beat most human players. Thats all it will take to kill online poker.

[/ QUOTE ]
This is 100% true... as long as you have users who are connecting to a poker client through a insecure medium it is going to be prone to attack.
You can only really attempt to make it tamper proof because bots will be eventually able to defeat all statistical analysis that you can throw at it through the poker client.
However there are a few steps that the poker websites can take... the gaming industry has long dealt with this issue of "bots" and "hackers" reaking havok in almost any major online game out there and there method to fight back is called punkbuster however poker websites could go even further than that.
1)All of there traffic is and should always be relayed over an encrypted connection and should be monitored for any tampering which is a realistic thing to do today.
2)The system themselves should be monitered. This is the biggie and it involves screen captures and hardware identification. Sure you could have all of the input coming from another computer that is connected to yours through fancy hookups but the computers should always have a program running in the background watching the user for various things such as:
-weird 12 hour, no break 8 tabling sessions
-consistently winning players should be monitored more than losing players(who cares about a losing bot, but should still be monitered because it could improve)
-consistent changes in hardware, or hardware going undetected regularly
-"blue pill" (I think that's the correct phrase) the systems.. monitor for rootkits by becoming a rootkit.

Anything out of the ordinary, such as a user changing mice before and after every session should be logged... eventually they will become perfect and honestly, that will be the day that online poker dies but untill then you should hopefully be able to get enough information off of their computer to detect something out of the ordinary. Poker is a fickle game that is a statistical and "gut felling" game which will take a few senior level programers years to get a computer to understand, if they ever attempt it, I'm sure BBV will hear about it one way or another.
But then again there is monetary incentive to keep quite if you do produce a bot but humans make mistakes so we'll hear about it eventually... welcome to the golden age of online poker, it's not going to be here for much longer.
Eventually everybody here will end up stacking tourists in B&M casino's I guess.

[/ QUOTE ]

I stated earlier that I think the bot-writers will win in the long run because of the very limited interface between the poker site and the player. Even in the best case for poker sites, I don't think they can overcome this problem. The best case is that they come up with a hardware/software solution that prevents any sort of access by bot software.

How can bot-writers overcome that? Set up a video camera to look at the screen and have it connected to a seperate machine (the bot) that connects to the machine with the poker software (poker host) only through the mouse interface ( whatever the future equivalent of USB will be). It would be easy to build such that the poker host can't distinguish it from any other mouse.

I'm quite confident that in 10 years we'll have image recognition advanced enough that any information displayed for a human user will be able to be processed by the bot machine. At this point it feeds its actions through its mouse interface and does what it needs to do.

Even if the "mouse interface" could be overcome by poker sites, it comes down to always being able to have a human being be the interface between the bot and the poker site. You pay someone $10/hour to do exactly what the bot tells you.

[kerowo]

The poker sites will always have enough money to hire better programmers or buy better technology than the bot builders. WOW does pretty good job against bots and there is much less money involved. When the poker sites feel they are losing money because people are afraid of bots they will find way to stop the current bots and get ahead of the curve. And really, no one is afraid of the guy who creates a Movie Plot bot to beat the game, that will never be popular enough to be a concern.

[indianaV8]

[ QUOTE ]
Meh. Leave the drama bomb in the zoo. Withholding "ideas" that aren't implemented, and probably won't be implemented is silly. If the sites cared about the bot problem, if there is a bot problem, they would do something.

Off the top of my head I would think some kind of question written with the letters scrambled would be enough to mess up a bot. Not a captcha but a real, short answer question.

[/ QUOTE ]

Not only bots are not an issue currently, but they actually are good for the games.

The ask a question approach unfortunately will not work. While it will ban some fishes with IQ < 20, a good bot will make google search with the question, then look of which of the a,b,c,d answers (or whaever) on the site match best toward what is found in google.

[indianaV8]

[ QUOTE ]
A captcha like phrase was developed for online forms etc and could be implemented in a software for a site, but it might get annoying. I saw a captcha on a site last night that if I were color blind, would have been impossible to determine.

Only way I can think of that might be pretty close to untouchable would be to somehow determine if a keypress or mouse click came from a software interrupt or a hardware interrupt. If it was software it would simply be discarded. Without being a deep layer Windows API guru I can't tell you if it would be possible or not but if it were, that would solve it immediatley I would think. Don't believe you can fake a hardware interrupt from software without causing the kernel to go wacky.

I could be wrong.

[/ QUOTE ]

You can't do that. If the bots are run in VMWare, and the click is from the host PC you get 1:1 a user behaviour.

[indianaV8]

[ QUOTE ]
It is dreaming. Current OCR and image processing is so powerful the level of "smearing" necessary for a computer to mess up would make it totally unreadable for the human players.

Just look at how complex current CAPTCHAs have to be in order to fool the bots.

And while messing around with the card images and making them harder to modify would certainly be a good step to prevent casual bot makers, slightly more complex software would easily handle it.

dave's idea up there seems pretty bullet-proof, but of course nobody would play on a system like that.

[/ QUOTE ]

Which is actually quite bad idea to go for, do you agree (i.e. it is the script kiddies that bring masses of losing bots and by stopping them, the only bots left will be the real consistent winners, as if a programmer have huge winner, he will spend the neccessary time to avoid any anti-detection).

[indianaV8]

[ QUOTE ]
[ QUOTE ]

All of your solutions rely on Bots not becoming really good at mimicking people, or on technology that would be "tamper proof". Its more than naive (its stupid) too make either of those assumptions. The reality is in time AI will be advanced enough that bots will be able to beat most human players. Thats all it will take to kill online poker.

[/ QUOTE ]
This is 100% true... as long as you have users who are connecting to a poker client through a insecure medium it is going to be prone to attack.
You can only really attempt to make it tamper proof because bots will be eventually able to defeat all statistical analysis that you can throw at it through the poker client.
However there are a few steps that the poker websites can take... the gaming industry has long dealt with this issue of "bots" and "hackers" reaking havok in almost any major online game out there and there method to fight back is called punkbuster however poker websites could go even further than that.
1)All of there traffic is and should always be relayed over an encrypted connection and should be monitored for any tampering which is a realistic thing to do today.
2)The system themselves should be monitered. This is the biggie and it involves screen captures and hardware identification. Sure you could have all of the input coming from another computer that is connected to yours through fancy hookups but the computers should always have a program running in the background watching the user for various things such as:
-weird 12 hour, no break 8 tabling sessions
-consistently winning players should be monitored more than losing players(who cares about a losing bot, but should still be monitered because it could improve)
-consistent changes in hardware, or hardware going undetected regularly
-"blue pill" (I think that's the correct phrase) the systems.. monitor for rootkits by becoming a rootkit.

Anything out of the ordinary, such as a user changing mice before and after every session should be logged... eventually they will become perfect and honestly, that will be the day that online poker dies but untill then you should hopefully be able to get enough information off of their computer to detect something out of the ordinary. Poker is a fickle game that is a statistical and "gut felling" game which will take a few senior level programers years to get a computer to understand, if they ever attempt it, I'm sure BBV will hear about it one way or another.
But then again there is monetary incentive to keep quite if you do produce a bot but humans make mistakes so we'll hear about it eventually... welcome to the golden age of online poker, it's not going to be here for much longer.
Eventually everybody here will end up stacking tourists in B&M casino's I guess.

[/ QUOTE ]

Guys,
you need to understand that the only thing that all these measures do is to stop the mass framework, script kiddies, scam based and so on LOSING bots. All these things are very easy to counter, they are ignorably easy compared to the time a guy have already spend on building a profitable winning bot.

I don't want to go further illustrating this, but there are currently very good examples of this effect.

[jjshabado]

Its naive to assume that script kiddies will always have bad bots. At some point code for a bot AI will be released that is better than the vast majority of humans. At this point online poker dies.