This alert appeared on the screen after the game (https://lichess.org/wKofR2DZ/black#60) even though my opponent did not let the timer run out. We were both running out of time, and my opponent was just trying to keep from making a blunder. Besides, he was obviously leading.
If resources allow, I would recommend fine-tuning the algorithm that determines when time is running out.
There will always be a few false positives. I've received the warning myself occasionally even though I would of course never stall the clock intentionally. As long as you don't see the warning consistently, you have nothing to worry about. In my experience, the system works very well, although it can potentially temporarily ban people with a bad connection even if they weren't stalling the clock.
I honestly dont undertand this, if its a 5 0 game why cant i choose to let my time run out?, how do you know for a fact i am not thinking and my flag jus fell?, I mean, it either is my time to use or it isnt, which one is it?
@luchachess said in #3:
> how do you know for a fact i am not thinking and my flag jus fell?
You don't, but Lichess has an algorithm that guesses, and it's usually pretty good.
> I mean, it either is my time to use or it isnt, which one is it?
Of course you can use your time, but not to stall purely out of spite. If you're not stalling, then don't worry about using your time because it is unlikely that Lichess will confuse your thinking with stalling, and even if it does occasionally, it's nothing to worry about like I said in #2.
This is funny, lets say i decide that I want to lose everygame on purpose so i start hanging queens, and allowing mate and losing games, thats all fine and dandy, but if I use my time as I see fit, i might get banned, that makes no sense.
@luchachess said in #5:
> lets say i decide that I want to lose everygame on purpose so i start hanging queens, and allowing mate and losing games, thats all fine and dandy
No, it's definitely "fine and dandy". It's called "sandbagging" and as a case of rating manipulation it's forbidden and repeating offenders may even end up with their account marked for ToS violation. In other words, worse offense than letting your time run out.
And you prove that how exactly? Again, this is all based on assumptions that can't be objectively proven. Its their platform, so they can put whatever rules they want. But lets not assume that any of this has any foundation on an objective evaluation of evidence on a case by case basis. Basically its whatever we want whenever we want it. Either that, or a generic algorthim can read peoples minds like a God.
@luchachess It can't read people's minds, but it works very well. On the rare occasion that it gives a false positive and bans an innocent person you can always appeal.
It works very well, at what exactly? Identifying human intent based on generalizations? It doesn't have the ability to adjudicate a specific case. Imagine that logic being used on a criminal court? We are going to give you a 100-year jail sentence because this algorithm tells us based on whatever data we fed it, that you probably had intention to kill. Again their platform their rules, but lets at least admite the whole thing is arbitrary and not based on any specific evidence.
@luchachess said in #9:
> We are going to give you a 100-year jail sentence because this algorithm tells us based on whatever data we fed it, that you probably had intention to kill.
This is basically, how real life courts work. The algorithm is buried in the minds of jurors and judges and nobody knows it.
This topic has been archived and can no longer be replied to.