Most of players that playing in anonimous seems to play with less 40cpl precision's. This seems really unrealistic. I have seen that other people have had the same strange feeling. Why a good site like lichess does this kind of thing? Help me to understand...
lichess.org/forum/general-chess-discussion/playing-versus-people-with-bot-assist#3
"Its hard to stop anon from cheating as lichess needs time and multiple games to find cheaters and this is done manually as an automated system will have alot more false positives. Also then what does lichess ban their ip that can be changed without a vpn. anything on the computer can be removed and lichess does not want to collect or make any id system just for non users that will be easy doged"
"Its hard to stop anon from cheating as lichess needs time and multiple games to find cheaters and this is done manually as an automated system will have alot more false positives. Also then what does lichess ban their ip that can be changed without a vpn. anything on the computer can be removed and lichess does not want to collect or make any id system just for non users that will be easy doged"
Or, in other words: if you want to play sensible games, don't play anonymously.
There is simply no good way to punish anonymous players for anything.
One could say that lichess should make registering mandatory, but that's throwing the baby out with the bathwater...
There is simply no good way to punish anonymous players for anything.
One could say that lichess should make registering mandatory, but that's throwing the baby out with the bathwater...
@nadjarostowa said in #4:
> Or, in other words: if you want to play sensible games, don't play anonymously.
>
> There is simply no good way to punish anonymous players for anything.
>
> One could say that lichess should make registering mandatory, but that's throwing the baby out with the bathwater...
I wonder if there could be a situation where Lichess would be able to identify a user (name, surname, etc.), even if they prefer to use an anonymous/obfuscated handle. (Don't know anything about chessdotcom, but I believe they usually know who's who on their site.)
That would allow users some privacy, and it will also make it possible for Lichess to identify offenders.
(Of course, this doesn't even begin to solve the problem of who, how, and with what budget is going to do all that.)
> Or, in other words: if you want to play sensible games, don't play anonymously.
>
> There is simply no good way to punish anonymous players for anything.
>
> One could say that lichess should make registering mandatory, but that's throwing the baby out with the bathwater...
I wonder if there could be a situation where Lichess would be able to identify a user (name, surname, etc.), even if they prefer to use an anonymous/obfuscated handle. (Don't know anything about chessdotcom, but I believe they usually know who's who on their site.)
That would allow users some privacy, and it will also make it possible for Lichess to identify offenders.
(Of course, this doesn't even begin to solve the problem of who, how, and with what budget is going to do all that.)
@nadjarostowa said in #4:
> Or, in other words: if you want to play sensible games, don't play anonymously.
>
> There is simply no good way to punish anonymous players for anything.
>
> One could say that lichess should make registering mandatory, but that's throwing the baby out with the bathwater...
I agree with you, but to be fair shadow banning may work for also for anonymous users.
So maybe aggressive automatic cheating detection + shadow banning may work sufficiently well to disincentive cheater in anonymous game
> Or, in other words: if you want to play sensible games, don't play anonymously.
>
> There is simply no good way to punish anonymous players for anything.
>
> One could say that lichess should make registering mandatory, but that's throwing the baby out with the bathwater...
I agree with you, but to be fair shadow banning may work for also for anonymous users.
So maybe aggressive automatic cheating detection + shadow banning may work sufficiently well to disincentive cheater in anonymous game
@Ender88 said in #6:
> I agree with you, but to be fair shadow banning may work for also for anonymous users.
> So maybe aggressive automatic cheating detection + shadow banning may work sufficiently well to disincentive cheater in anonymous game
The more "aggressive" cheating detection is made, the more false positives it will produce (players wrongly accused of cheating). This is a terrible road to go down.
By shadow banning, do you mean banning on the basis of IP address? (If not, what do you mean?) This is a non-starter because many people are on mobile wifi networks, even at home some people have internet providers which allocate an address from a dynamic list, plus places like schools and libraries might find themselves "shadow banned" from Lichess.
> I agree with you, but to be fair shadow banning may work for also for anonymous users.
> So maybe aggressive automatic cheating detection + shadow banning may work sufficiently well to disincentive cheater in anonymous game
The more "aggressive" cheating detection is made, the more false positives it will produce (players wrongly accused of cheating). This is a terrible road to go down.
By shadow banning, do you mean banning on the basis of IP address? (If not, what do you mean?) This is a non-starter because many people are on mobile wifi networks, even at home some people have internet providers which allocate an address from a dynamic list, plus places like schools and libraries might find themselves "shadow banned" from Lichess.
@Brian-E said in #7:
> The more "aggressive" cheating detection is made, the more false positives it will produce (players wrongly accused of cheating). This is a terrible road to go down.
>
> By shadow banning, do you mean banning on the basis of IP address? (If not, what do you mean?) This is a non-starter because many people are on mobile wifi networks, even at home some people have internet providers which allocate an address from a dynamic list, plus places like schools and libraries might find themselves "shadow banned" from Lichess.
When you connect to a site (that need to somewhat remember you, eg: to know the game in which you are playing), a session is initiated, typically using a cookie to hold the session key. After each site interaction (data transfer) the key is sent back and forth, so the site can remember what are you doing and do not restart as if the first time you showed up.
This is particularly true in lichess in which the site have to remember in which game are you while you are playing, even in anon.
AFAIK : lichess use exactly one cookies just for this technical issue
So if you connect here, anonymous or not, untill you reset the session client-side (aka clear cookies and/or refresh cache) you can be targeted regardless of transport protocol you use (may be pigeon if like :) ).
So IP or other technical detail you described of transport protocols does not matter anymore, what matters are the rule of the application protocol (HTTPS)
Only for anonymous if the automatic system believes you are cheating, for now on you will be paired only to other believed cheater. No warning no blaming.
So the cheater didn't suspect anything and have no reason to try escape shadow banning, and regular players have fun.
Is false positive likely? Of course cab happen, but due to the temporary nature (after each disconnect site lose track on you and so you evade shadow banning) it will be by far less that keep the section a far west where cheating is de facto allowed.
And again I believe that automatic aggressive cheating detection is good only for this kind of volatile situation, where users are ephemeral.
Not for the rest of the site.
> The more "aggressive" cheating detection is made, the more false positives it will produce (players wrongly accused of cheating). This is a terrible road to go down.
>
> By shadow banning, do you mean banning on the basis of IP address? (If not, what do you mean?) This is a non-starter because many people are on mobile wifi networks, even at home some people have internet providers which allocate an address from a dynamic list, plus places like schools and libraries might find themselves "shadow banned" from Lichess.
When you connect to a site (that need to somewhat remember you, eg: to know the game in which you are playing), a session is initiated, typically using a cookie to hold the session key. After each site interaction (data transfer) the key is sent back and forth, so the site can remember what are you doing and do not restart as if the first time you showed up.
This is particularly true in lichess in which the site have to remember in which game are you while you are playing, even in anon.
AFAIK : lichess use exactly one cookies just for this technical issue
So if you connect here, anonymous or not, untill you reset the session client-side (aka clear cookies and/or refresh cache) you can be targeted regardless of transport protocol you use (may be pigeon if like :) ).
So IP or other technical detail you described of transport protocols does not matter anymore, what matters are the rule of the application protocol (HTTPS)
Only for anonymous if the automatic system believes you are cheating, for now on you will be paired only to other believed cheater. No warning no blaming.
So the cheater didn't suspect anything and have no reason to try escape shadow banning, and regular players have fun.
Is false positive likely? Of course cab happen, but due to the temporary nature (after each disconnect site lose track on you and so you evade shadow banning) it will be by far less that keep the section a far west where cheating is de facto allowed.
And again I believe that automatic aggressive cheating detection is good only for this kind of volatile situation, where users are ephemeral.
Not for the rest of the site.
Yes, it could be done with cookies. The cheating user would have to be savvy enough to realise what is going on and clear them.
Perhaps this is already in fact done? We don't know precisely what measures are in place.
I take issue with your suggestion @Ender88 that false positives don't matter in "ephemeral" situations. Players who are playing anonymously (and that in itself may be a temporary situation - they may sometimes play on an account, or they may create an account in the future) are still players and should not be treated lightly.
Perhaps this is already in fact done? We don't know precisely what measures are in place.
I take issue with your suggestion @Ender88 that false positives don't matter in "ephemeral" situations. Players who are playing anonymously (and that in itself may be a temporary situation - they may sometimes play on an account, or they may create an account in the future) are still players and should not be treated lightly.
@Brian-E said in #9:
> Yes, it could be done with cookies. The cheating user would have to be savvy enough to realise what is going on and clear them.
>
> Perhaps this is already in fact done? We don't know precisely what measures are in place.
>
> I take issue with your suggestion @Ender88 that false positives don't matter in "ephemeral" situations. Players who are playing anonymously (and that in itself may be a temporary situation - they may sometimes play on an account, or they may create an account in the future) are still players and should not be treated lightly.
AFAIK moderation is only manual and nothing of what I suggest is in place.
Just an educated guess, but based on the open source code I have readen (partially readen, so I have to admit, maybe I overlooked something).
Regards the pro/cons this is highly subjective, I know I just share my point of view about that. And in general I agree with you, but not for anon games (also take in account that aggresivity can be tuned).
All that said I think the biggest obstacle will be the huge computational cost (that will become a monetary cost) with no return in exchange.
After all those are not regular users yet, and investing resources on them may be not the best idea (especially when Lichess work on donations)
> Yes, it could be done with cookies. The cheating user would have to be savvy enough to realise what is going on and clear them.
>
> Perhaps this is already in fact done? We don't know precisely what measures are in place.
>
> I take issue with your suggestion @Ender88 that false positives don't matter in "ephemeral" situations. Players who are playing anonymously (and that in itself may be a temporary situation - they may sometimes play on an account, or they may create an account in the future) are still players and should not be treated lightly.
AFAIK moderation is only manual and nothing of what I suggest is in place.
Just an educated guess, but based on the open source code I have readen (partially readen, so I have to admit, maybe I overlooked something).
Regards the pro/cons this is highly subjective, I know I just share my point of view about that. And in general I agree with you, but not for anon games (also take in account that aggresivity can be tuned).
All that said I think the biggest obstacle will be the huge computational cost (that will become a monetary cost) with no return in exchange.
After all those are not regular users yet, and investing resources on them may be not the best idea (especially when Lichess work on donations)
This topic has been archived and can no longer be replied to.