Well, if we consider that cheating, then playing against any AI is playing against a cheater, they fully analyze the position before any move to find the one who gets the absolute best material advantage (fortunately, this can be easily exploited with the queen gambit and similar openings).
I fully support a reputation system. However, filtering out people without rep means new players will never get to play anyone, so I'm not sure I'd support filtering by rep. Same with not allowing them to play in tournaments. However, someone posted an idea in another forum about not allowing players with virtually no games (i.e. a high variance value in their rating) not being allowed into the more significant tournaments (daily, weekly, marathon, etc.) which I could also get down for.
As for assuming if we don't talk about the problem it will go away, that's absurd. No one is saying such things. But I don't see how talking about it in the manner you do helps either, Nada. The way I see it, pointing out the obvious, that there are cheaters in online chess, doesn't do a lot, even if you're pointing out ways to supposedly identify them, there is no need. The proper way to handle cheaters, as laid out in the FAQ, is to report suspicious activity if and only if you find it suspicious after a computer-assisted analysis. So basically, generate analyses and file reports. That's how any chess site finds cheaters, not by posting 30+ forum messages over a few months about how big a problem cheating is and discussing ways to "catch" cheaters. These are not beneficial to actually deterring cheating, in fact it's doing the opposite by pointing out those games, you're effectively giving them the attention they are seeking.
As for assuming if we don't talk about the problem it will go away, that's absurd. No one is saying such things. But I don't see how talking about it in the manner you do helps either, Nada. The way I see it, pointing out the obvious, that there are cheaters in online chess, doesn't do a lot, even if you're pointing out ways to supposedly identify them, there is no need. The proper way to handle cheaters, as laid out in the FAQ, is to report suspicious activity if and only if you find it suspicious after a computer-assisted analysis. So basically, generate analyses and file reports. That's how any chess site finds cheaters, not by posting 30+ forum messages over a few months about how big a problem cheating is and discussing ways to "catch" cheaters. These are not beneficial to actually deterring cheating, in fact it's doing the opposite by pointing out those games, you're effectively giving them the attention they are seeking.
The essence of experience and reputation system is that you have to EARN privileges.
It is not that - as opponents of the system say - newbies will come in and downvote decent players or no one will play provisional players.
You have to earn the right of voting, and yes, the right for filtering. When you are a provisional player you don't have the luxury to be choosy about your opponents. You have to play against anyone. There is no danger that nobody will play beginners.
A player with more than 5000 games played, with more than a year long membership, rated in all categories, a lichess donor, leader of a team with more than 200 members, should be allowed to filter their opponens. This would pose no danger to the liquidity of the site.
It is not that - as opponents of the system say - newbies will come in and downvote decent players or no one will play provisional players.
You have to earn the right of voting, and yes, the right for filtering. When you are a provisional player you don't have the luxury to be choosy about your opponents. You have to play against anyone. There is no danger that nobody will play beginners.
A player with more than 5000 games played, with more than a year long membership, rated in all categories, a lichess donor, leader of a team with more than 200 members, should be allowed to filter their opponens. This would pose no danger to the liquidity of the site.
Well, what you say is a nice idea, however that would require a gratuitous amount of restructuring the current filter system, as it simply filters by flags. You'd now have to add flags to filter with these new rep ratings, and then now flags for which users have access? That's getting to be quite a bulk of added databasing just to keep track of who can and can't rep people and who can and can't filter certain rep levels out, and so on. I don't know a lot about coding, but I am pretty sure it would be quite a beast to develop. But I definitely like that you've thought of ways to challenge the usual opposition to this sort of idea. I'm still largely in support of it.
Look, what I say does not come from thin air, these systems are in place and are working elsewhere. Therefore it should be impossible to put in place such a system especially for a site that is obviously capable of coding complex problems requiring much databasing, just look at the new Insights feature.
I have cited several times a very early correspondence chess site by Artur Wachelka ( which has since then been destroyed by ads and trying to make money out of it ), which in many ways was revolutionary and in many ways gave me the fundamental ideas about how a chess site should look like. Even at this site the elements of an experience system were already in place. Players were marked with icons from pawn to king based on their experience and activity, and this was visible in their profiles and seeks. Also in seeks you could require several criteria like number of games played, how many times the player forfeits on time etc.
I have cited several times a very early correspondence chess site by Artur Wachelka ( which has since then been destroyed by ads and trying to make money out of it ), which in many ways was revolutionary and in many ways gave me the fundamental ideas about how a chess site should look like. Even at this site the elements of an experience system were already in place. Players were marked with icons from pawn to king based on their experience and activity, and this was visible in their profiles and seeks. Also in seeks you could require several criteria like number of games played, how many times the player forfeits on time etc.
@Fenris1066:
At one hand you say that I should not point out ways to identify cheaters with the help of computer analysis (by suspicious move times for example) and on the other hand you say that everyone should report cheaters only based on such analysis. Dont get it.
I had no idea how to make a good report until I found CIA forum and other players gave me some advice. Basically you are saying that this kind of communication is more or less pointless and only giving cheaters some attention.
My experience is contrary to that: Cheaters dont like any kind of attention if this attention is only based on the fact that their cheating will be detected sooner or later.
But as I already said - I will be quiet from now on.
People obviously dont like talking about cheating and I will respect this.
Best wishes
Nada
At one hand you say that I should not point out ways to identify cheaters with the help of computer analysis (by suspicious move times for example) and on the other hand you say that everyone should report cheaters only based on such analysis. Dont get it.
I had no idea how to make a good report until I found CIA forum and other players gave me some advice. Basically you are saying that this kind of communication is more or less pointless and only giving cheaters some attention.
My experience is contrary to that: Cheaters dont like any kind of attention if this attention is only based on the fact that their cheating will be detected sooner or later.
But as I already said - I will be quiet from now on.
People obviously dont like talking about cheating and I will respect this.
Best wishes
Nada
Which for a correspondence site, where you have plenty of time to research your opponents before deciding to play, that's great.
But for the average lichess player that won't help. It would have to be pretty much in the filter system, which as I said would likely require a lot of restructuring of the entire filtering system to do what you are suggesting in terms of limiting filters to people of a certain rep and so forth. I'm not saying it's impossible, I'm saying for a group of developers who have real jobs and don't get paid for their time, it would be something that is probably going to need a lot of convincing that it's worth the effort to go through and create such a system and implement it.
In other words, you're going to need to get a lot of community support behind it or code it yourself, basically. The database is already there for the Insights stuff, they just had to implement an algorithm to pull from it. You're asking for an entire new layered database structure to be added. Not to mention one that can be interacted with on a minute-by-minute basis. I'd love to hear a developer opinion on how daunting a task such a thing would be.
But for the average lichess player that won't help. It would have to be pretty much in the filter system, which as I said would likely require a lot of restructuring of the entire filtering system to do what you are suggesting in terms of limiting filters to people of a certain rep and so forth. I'm not saying it's impossible, I'm saying for a group of developers who have real jobs and don't get paid for their time, it would be something that is probably going to need a lot of convincing that it's worth the effort to go through and create such a system and implement it.
In other words, you're going to need to get a lot of community support behind it or code it yourself, basically. The database is already there for the Insights stuff, they just had to implement an algorithm to pull from it. You're asking for an entire new layered database structure to be added. Not to mention one that can be interacted with on a minute-by-minute basis. I'd love to hear a developer opinion on how daunting a task such a thing would be.
#108 was in response to #106. Nada posted his passive-aggressive whiny repetition of the same thing he's said 6 other times while I was typing it.
The problem with cheating is that if you have to prove it has to be very obvious. There are players who I'm not comfortable playing with, and can be never be proven to be cheaters. But they could my get my downvote. And I could downvote them with a straight face. Like a 2300 correspondence player playing as a godlike tactician to punish me when I make a mistake in a correspondence game, but rated 1600 in tactics after solving 50 puzzles, which is a joke, not played a single live game. And making all the suboptimal moves which score inaccuracy in engine analysis, till I make a mistake. Impossible to prove cheating. But possible to downvote. Or even simple cases of playing three moves and then giving up the queen for a bishop. Clicks on the next seek, again loses in 3 moves. Rated 1900+. Sandbagging? Joking? Bad mood? I don't care, I blocked this player, but may be my downvote would have helped other players to avoid them. Reputation is a very valuable piece of information, and as I'm convinced that the vast majority of voters have good intentions, a reliable one also.
@Fenris1066:
You are obviously aggressive enough as well so you should have no problem with the aggressive (or "passive-aggressive", whatever this should be) behaviour of others.
Nada
You are obviously aggressive enough as well so you should have no problem with the aggressive (or "passive-aggressive", whatever this should be) behaviour of others.
Nada
This topic has been archived and can no longer be replied to.