There is a very elegant and simple solution that can be used here which ICC implemented when I played there.
A no escape feature function. During game creation you enable a checkbox for "No Escape", and you game gets 20-30 seconds for disconnection, that's it. If you leave you lose rating points. This removes rage quitters from the equation completely.
The people with bad internet keep the option disabled and are automatically paired with the same, so you can't complain about unfairness.
I don't understand why this is such a magnificently complicated problem when the solution is so simple?
Which option is the default, the one without the "no escape"? If so, then the pools on the front page will be without this option and this will reduce the complaints only marginally. If the default is no-escape checked by default, then this punishes people with bad connections. I don't love this idea.
Especially since is splits the pools.
I think I'd much prefer to continue improving the detection and ensuring people get the option to finish the game (claim win/draw) when it happens. With the new changes that have recently happened to improve this, I wonder how many of the complaints in that thread are still valid
This is really no different than what we have now. If we can detect that someone is disconnected (which the above solution depends on), then the difference between the current solution and this proposed solution is only whether the win is automatically given or if the prompt is displayed. I _much_ prefer the prompt being displayed as it gives the human the opportunity to decide how much time they want to wait. We should be focused on improving the detection of disconnects and punishing those with a pattern of behavior.
OK, thanks for the thoughtful analysis and I agree that sounds like a more constructive direction. The new timer UI hopefully helps calm players as well.
People are outraged for a reason. The option is not a default setting. Once you click the checkbox during game creation you agree to a No Escape 30 seconds. Chess.com gives everyone 30 seconds no matter what. Seems to work miracles there. I don't understand why there is such stiff resistance to giving the community an option. I am willing to bet that after a few weeks not only will you have the majority of the people clicking the No Escape 30 second option, but the entire arguments, rage, complaints, and everything else lichess is experiencing will vanish. If you are not a believer why not try it as a trial balloon for one month.
What you have now is alienation. The community is frustrated, alienated, and honestly looking for other options as I have seen on the posts. I personally tried chess.com because they have a 30 second disconnection grace period. The only reason I came back to Lichess is because chess.com's other tools are absolutely inferior to lichess, not to mention their interface is an atrocity, impossible to navigate. If lichess is so concerned about protecting disconnecting players this still protects them, as the default setting still allows for it. This just gives a useful took for all the players that are experiencing bad sports on a daily basis the option to be paired with people that do not want to continue to be abused. If for instance we choose the option, and WE now must wait for a long time to be paired ? So What? It was our choice right? But now this gives Lichess perfect cover. Every complaint under the sun can be met with. You have the option to choose 30 seconds No Escape. This option will also protect you from people that are threatening to no longer donate, etc. etc. Your chess community is suffering, and this solves all your problems by simply giving people an option.
@nephilx we have a system that provides a dynamic amount of time to players. When someone disconnects we are now displaying a timer countdown that tells you how much time you have left before you can claim a win. Then you can claim a win. It's as simple as that. There is no need for another checkbox.
I will also note that there was a bug in it that was fixed a few days ago. If you are still experiencing bugs related to it, please report them. But the goal we're aiming for is the same. Everything you say about your proposed system is also true about our proposed system which is already implemented. If it's buggy, then we simply need help debugging it. If it's not buggy, then it works already (IMO) better than your proposed solution.
We are listening and we are improving things.
By "dynamic" I mean the amount of time you must wait depends on a few factors. Whether you're still connected to the site at all, what the time control is and some other factors that I don't remember off the top of my head.
Ok, So for Blitz set your dynamic to 30 seconds. I will stop posting about this nonsense. What I dislike is when people leave for 2 minutes in a losing position, start another game, or just leave the tab to try and catch you snoozing. If someone agrees to longer time controls its normal to leave, get tea, think, ponder. That makes sense, in Blitz it is frustrating that every other player leaves the game without resigning and just starts another game. It is infuriating actually. They know they lost but they want to now punish you and waste your time while they continue to play. I know you ban people after a cerrtain about of times they do this but it is not harsh enough. For blitz lower it to 30 seconds abandon you automatically forfeit the game. If you set it to 30 seconds No Escape Option will not be necessary. Thanks
I totally get your frustration. I don't know what the time is for blitz pools off the top of my head. But I'd be surprised to find out it's 2 minutes for a 3+2 game
@nephilx I just tried it on a testing site in 3+2 and I was able to claim a win within 10 seconds of the other player disconnecting.
Most helpful comment
I think I'd much prefer to continue improving the detection and ensuring people get the option to finish the game (claim win/draw) when it happens. With the new changes that have recently happened to improve this, I wonder how many of the complaints in that thread are still valid