@duran_was_the_mvp said in #543:
> Nope, wrong again. Lichess takes away every option available for sandbagging and cheating, without hindering any of the fair users. They have done exactly that for color abusers. The option to use it remains, but only for fair use.
Removing the option to choose a color hinders fair users like me. This is a case of targeted measures versus broad measures. When dealing with cheaters and sandbaggers, you address the individual players who violate the rules. However, by removing the ability to choose a color, you punish the entire playerbase.
If the counterargument is "invitational challenges," nobody challenges random people they find online. First, it's unsettling to receive random notifications from strangers, and second, it's cumbersome at best.
I won't bother explaining why the sky is blue.
> Yes, you said it wouldn't be so much, just ballpark me the figure and I'll relay the actual cost
> It's not what drove the upgrade for fair play, but writing out specific features to accommodate these specific usecases is what Thibault has spoken about. It's a big factor over time, I would say that his opinion on the matter should carry some weight.
If cost isn't what drove the update, bringing it up in the first place was irrelevant, as I mentioned. The only reason you keep mentioning it is to muddy the waters. Thibault, in his post, was discussing why new features may not be integrated into Lichess, even when well implemented. Color selection was an already built feature and has nothing to do with the cost of introducing new stuff.
A very simple detection mechanism that tracks players with over a certain percentage of games played with a specific color and over a certain number of total games wouldn't require more than 500 lines of code, accounting for data retrieval, processing, banning, possible UI elements for moderators (though I don't know what tools moderators have), and debugging/testing. That would be, at most, one month of work (which I think is a stretch, but let's call that the absolute ceiling). Given that Lichess is based in France, I would estimate that a well-paid developer there would earn around 3500 euros per month, so the cost would be roughly that.
But please, go ahead and give me the "actual" cost that's so relevant to this conversation. I'm still waiting.
> It's an upgrade, just like anti-cheating measures are.
Removing a feature is, by definition, a downgrade. Chess.com has it as well, and it’s working exceptionally. Now, queue the irrelevant argument about maintenance cost and how chess com is a multimillion dollar company
> Or you, who is advocating for this change, could make the change yourself
And have it denied by Thibault who took the decision to remove it in the first place. Sounds smart.
> Forcing your opponent to play with black in all your matches is objectively unfair, just like sandbagging and cheating.
If it was truly unfair, you would deal with the people that abuse it. You wouldn't remove the entire feature.
> They're almost identical. The site takes any measure it can against the use of engines. It detects plugins, detects if you're using the site's analysis board during the game, detects (to some degree) if you have been using a local instance of an engine for your games. It detects whether you're sandbagging in the same way.
It doesn't have any detection mechanism for people playing exclusively with one color, so by your description, it's not identical at all. Methods for detecting cheating or sandbagging are irrelevant.
> Treating color picking like cheating would be fair, in which case the site should be banning it completely as much as it can. In the case of engine-use: You don't have players that put themselves in a lobby and say "I'll be using an engine in all my games against you if we're matched", nor do you have the option for players to say "I'll be sandbagging this match" in advance. If the option was there AND Lichess were to rely solely on detection mechanisms, I would think it's fair that Lichess removes this feature to start with and then detecs people who are still trying to abuse it (handling it on a case-by-case basis).
> And this is exactly what Lichess has done: They have removed the option to abuse color, and they are taking a case-by-case measure against people that are still trying to abuse it (e.g. only playing white by starting games and aborting all black games).
This doesn't make sense. Good job banning stubborn people who frequently abort games. Now you have a player base missing a feature. The "detection" method you described only catches people who abort, not those who play 90%+ of their games with a specific color.
> It's the exact same thing, carried out to the best possible success by removing the option for it, and then handling persistent abusers on a case by case basis
Punish everyone and solve an entirely different problem (frequent aborting). Sounds fair.
The bottom line is that, in reality, good arguments don't really matter here. If the developer wants something removed for any reason, even if that reason is logically flawed, it will be removed. A feature was abused by some users, and there were two possible solutions: the lazy one and the more difficult one. The lazy solution is to remove the feature entirely, eliminating the problem at its source (since there's no feature to exploit). The harder solution is to implement a detection mechanism.
The developer chose the first option, which is fine, it's his site, and he makes the rules. I would just appreciate it if the "ardent supporters" of this change were honest about that. If the feature was truly problematic, Chess com would have removed it as well.
I won't be making any further replies on this topic. I believe my position is clear.
@NotTakenUsername said in #571:
> Removing the option to choose a color hinders fair users like me. This is a case of targeted measures versus broad measures. When dealing with cheaters and sandbaggers, you address the individual players who violate the rules. However, by removing the ability to choose a color, you punish the entire playerbase.
Of course it's a hindrance. It's worth nothing that they did leave the option in, but just not for automatic matchmaking. So you're only hindered in forcing a color on your opponent but may still invite others to participate freely.
> If the counterargument is "invitational challenges," nobody challenges random people they find online. First, it's unsettling to receive random notifications from strangers, and second, it's cumbersome at best.
> I won't bother explaining why the sky is blue.
It's not ideal, but still a solution, we agreed on that.
> If cost isn't what drove the update, bringing it up in the first place was irrelevant, as I mentioned. The only reason you keep mentioning it is to muddy the waters. Thibault, in his post, was discussing why new features may not be integrated into Lichess, even when well implemented. Color selection was an already built feature and has nothing to do with the cost of introducing new stuff.
Not at all. The feature of picking colors was there in the context of matching, but not in the context of creating different matching ecosystems for fair players, and another one for color pickings.
The request to enable this, is indeed adding a new feature. Choosing a color in and of itself exists, but not the feature to offer this in the exact way that you're pondering. This would require cost, development, so it's extremely relevant as to why this may not be implemented. This has nothing to do with whether the fair-play upgrade was cost-driven.
> A very simple detection mechanism that tracks players with over a certain percentage of games played with a specific color and over a certain number of total games wouldn't require more than 500 lines of code, accounting for data retrieval, processing, banning, possible UI elements for moderators (though I don't know what tools moderators have), and debugging/testing. That would be, at most, one month of work (which I think is a stretch, but let's call that the absolute ceiling). Given that Lichess is based in France, I would estimate that a well-paid developer there would earn around 3500 euros per month, so the cost would be roughly that.
It would be more than 500 LOC, but probably under 5000. Some issues with this:
1. You now have people who abort games that are not with their preferred color, let time run out, etc. They could simply keep making new accounts to get their preferred white, thus exploiting it.
2. If you would allow this percentage-based, you would need a cut-off point for current accounts, and a buffer for new accounts for the percentages to be relevant. This means a tracking system of the stats for the sole purpose of this one 'feature'. If you don't, older accounts with a color debt are instantly banned if they get 1 game with their previously chosen color, or mods would have to go through thousands of case-by-case instances to check for this. Or a detection system would need to be built on top of this, just to detect these cases (which would need its own abuse prevention).
3. 2 is a lot of extra work and extra upkeep.
To mitigate all of these (you would still need a cut-off point), the percentages, or the color debt would still need to be accounted for. So you would have to keep these color streaks small enough (let's say 10 in a row). That means, that in their next games they would have to repay this debt and, if the moderators' comments are any indication, they'll start complaining how they're paired with black too often (assuming they picked white). The bigger the allowed streak, the bigger the debt. And after the streak, you would have to enforce they 'repay' this debt before starting another streak. This too means extra code, extra upkeep, extra cost.
> But please, go ahead and give me the "actual" cost that's so relevant to this conversation. I'm still waiting.
I will, I'll try to provide it this week as accurately as possible.
> Removing a feature is, by definition, a downgrade.
Nope, This one is a fair-play upgrade
> Chess.com has it as well, and it’s working exceptionally. Now, queue the irrelevant argument about maintenance cost and how chess com is a multimillion dollar company
Very relevant, but it's easy to say it isn't, if you;re not paying for it or doing the work
> And have it denied by Thibault who took the decision to remove it in the first place. Sounds smart.
He did not remove a option that is fair. He took abuse-prevention measures. If you can find an elegant solution that is low maintenance, you could propose it in concept and write it after he approves. Just complaining that he needs to do it instead is no way to achieve this either.
> If it was truly unfair, you would deal with the people that abuse it. You wouldn't remove the entire feature.
Exactly! They didn't remove the entire feature. They dealt with the abuse of it, and still allow it under fair circumstances. You act as if you're not able to pick colors at all, this is uncorrect.
> It doesn't have any detection mechanism for people playing exclusively with one color, so by your description, it's not identical at all. Methods for detecting cheating or sandbagging are irrelevant.
Nope, directly relevant. It doesn't need a detection mechanism, because you can shut down abuse by force. You can't do this with cheaters and sandbaggers.
> This doesn't make sense. Good job banning stubborn people who frequently abort games. Now you have a player base missing a feature. The "detection" method you described only catches people who abort, not those who play 90%+ of their games with a specific color.
This detection mechanism is not needed, because the fair play upgrade has solved this detection already. Now they just need to ban the people that abort too frequently.
> Punish everyone and solve an entirely different problem (frequent aborting). Sounds fair.
Yes, but it also solves people that are trying to get around this upgrade
> The bottom line is that, in reality, good arguments don't really matter here. If the developer wants something removed for any reason, even if that reason is logically flawed, it will be removed. A feature was abused by some users, and there were two possible solutions: the lazy one and the more difficult one. The lazy solution is to remove the feature entirely, eliminating the problem at its source (since there's no feature to exploit). The harder solution is to implement a detection mechanism.
Not a lazy solution, but a cost-effective one. As I explained, handling these things on a server this big is not as easy as you thought it to be. Every edge case needs to be accounted for, and nobody wants to pay for it.
They did not remove the entire feature though! You can still use it under fair circumstances.
> The developer chose the first option, which is fine, it's his site, and he makes the rules.
You can still make contributions though
> I would just appreciate it if the "ardent supporters" of this change were honest about that. If the feature was truly problematic, Chess.com would have removed it as well.
Or chess.com has the resources and money because they charge their users money. Lichess doesn't, so unless you're suggesting we start paying for Lichess, comparing the two sites is irrelevant. It's easy to say those factor don't matter if you don't need to code up the solution, or consider it's implementation in every update you make.
> I won't be making any further replies on this topic. I believe my position is clear.
Some guys posted the were a patron of lichess, but stop paying in protest for having removed the white/black feature. Assuming this is only the tip of the iceberg, I could imagine quite a few quit giving financial support to lichess in protest of this change.
Thus, the downgrade (removing a feature) might be cost more than working on a better solution.
I think the easiest one that satisfies all players is: white/black players appear in the lobby only, but never get automatically paired if their time control happens to be one of the quickpairings.
If you look up BeDecentForAchange profile, you see, he posts more messages (almost all exclusively in this thread) than he played games:
https://lichess.org/@/BeDecentForAChange
he is "hijacking" this thread, and my impression is, this is not his only account, he knows lichess well, maybe he is also duran? But doesnt matter, eventually he will get tired, while there will every be new users who wish the feature white/black to be back. Acutally, he helps to push the number of messages higher, maybe that will have some impact to lichess Admins?
And cutting financial support in protest, as some messaged here, might help, too. especially if they write why the cut their financil support. But most wont write. They simply leave to chess.com, I guess, and once lost, they never return.
@Munich said in #573:
> Some guys posted the were a patron of lichess, but stop paying in protest for having removed the white/black feature. Assuming this is only the tip of the iceberg, I could imagine quite a few quit giving financial support to lichess in protest of this change.
Some other guys that are in favor of this have started being patron to Lichess to support this.
> Thus, the downgrade (removing a feature) might be cost more than working on a better solution.
It's a fair-play upgrade to stop abusers. What's your calculation to support this?
> I think the easiest one that satisfies all players is: white/black players appear in the lobby only, but never get automatically paired if their time control happens to be one of the quickpairings.
Then, you are free to go ahead and make that change!
> If you look up BeDecentForAchange profile, you see, he posts more messages (almost all exclusively in this thread) than he played games:
> @BeDecentForAChange
> he is "hijacking" this thread, and my impression is, this is not his only account, he knows lichess well, maybe he is also duran?
You also made the suggestion that I'm @NotTakenUsername . Just saying random things without a single thought given is more hijacking this thread than me celebrating fair play and protesting abusers. Also, I'm rated 200 points lower than @duran_was_the_mvp in every time control. But.... good try.
> But doesnt matter, eventually he will get tired, while there will every be new users who wish the feature white/black to be back. Acutally, he helps to push the number of messages higher, maybe that will have some impact to lichess Admins?
I won't get tired of advocating for fair play!
>
> And cutting financial support in protest, as some messaged here, might help, too. especially if they write why the cut their financil support. But most wont write. They simply leave to chess.com, I guess, and once lost, they never return.
If they cut financial support to try to force Lichess to support abusers, I guess them leaving is good thing.
Moderators seem dead, dont they see a forum with 580 posts in lichess feedback, why are they not reacting? do they not care about the site?
Regarding unrated games, I don't like this innovation.
In the rated ones I never chose the color and I can understand the motivations of those who made this decision, but in the unrated ones I don't understand why you shouldn't have this option. I've never had problems finding players willing to play (even seriously) when I chose white, or black.
@a_Tauri said in #576:
> Regarding unrated games, I don't like this innovation.
An argument could be made for keeping it on for casual games. One could also argue that it does not offer a fair playing field then, for casual players. Or, perhaps the cost of maintaining that feature in an encapsulated way is too costly for the benefit
> In the rated ones I never chose the color and I can understand the motivations of those who made this decision, but in the unrated ones I don't understand why you shouldn't have this option. I've never had problems finding players willing to play (even seriously) when I chose white, or black.
That's because games were auto-matched when you chose a color, against people who chose the opposite color - but also players who play random side.
In your case, obviously you used it legitimately as it was intended, and if people were honest and fair like yourself, there would be absolutely no need for these measures. Unfortunately, there have also been people with 90K games with white, 0 with black. People abuse what they can, that's why we can't have nice things
@BeDecentForAChange said in #574:
> If they cut financial support to try to force Lichess to support abusers, I guess them leaving is good thing.
Oh, then Lichess either has enough financial resources (to use for a better code) or there weren't that many players that caused a problem if donations are dispensable.
If you are a member of the Lichess team, you can tell it now.
I'll put this short:
1. From UI perspective, it's a 100% bug. At least don't show white/black in the menu anymore, but only random.
2. I would have allowed choosing color at least on casual games. If you don't, after years of great experience with lichess, I'll have to move to chess.com, and most probably so would many others.
@Sofia-Mary said in #578:
> Oh, then Lichess either has enough financial resources (to use for a better code) or there weren't that many players that caused a problem if donations are dispensable.
>
> If you are a member of the Lichess team, you can tell it now.
Nope, but if they try to use financial pressure to get an unfair advantage, it's not a donation
This topic has been archived and can no longer be replied to.