AGA’s objective is to promote Go and participation. Adding anti-cheating measure is just too expensive and will certainly discourage participation. Probably true to any other Go organizations.
It’s a fun event and impractical to control cheat, then don’t get too serious in my view.
For many years, European football clubs refused to use go line technology to control cost. What can you do?
It is ridiculous to participate in any serious tournament without a thorough anti cheating measures in place.
It is simply too easy too cheat, way to common, and human nature cannot be trusted. That the AGA sponsored a tournament without measures in place is surprising and a large oversight.
Especially in light of the troubles with other cheating incidents in GO and other games. And in light of the last European vs AGA Professionals online tournament with the computer issue. While it is more work and definitely less conducive to organizing online tournaments anti cheating measures must be in place. Sadly it is almost impossible to prevent cheating if someone is intent on it, especially in online situations. But the measures can help keep the honest- honest.
Perhaps this will help for future tournaments.
It’s their tournament, hence they are responsible for making sure it is fairly arbitrated. Moreover, there’s price money involved as well, making it not “just for fun”. Finally there is reputation involved: other AGA tournaments that are more serious need to be undoubtedly fair. Cases like this make a dent in such trust.
Seems like you approach the game with a casual mindset, which is fine and I like that. But please understand that many people see Go as a competition. Players who aspire to become a professional dedicate a lot of time to their goal. How can they simply shrug when there’s cheating going on?
Well, hope covid crisis will be over soon. If you are serious, just play in person games. I just don’t think with my annual AGA fee, AGA has the power to control this thing online.
Btw, I am a serious player and have been watching lots of teaching videos. My problem is I spend too much time on this gossip thread instead of playing to improve. LOL
You are probably correct in regards to the AGA funding and ability to deal with these issues. Sadly even playing in person in tournaments has to be highly regulated to discourage cheating. Especially if they are monetized. We can learn from the Chess world.
Ok, what would you do if a 5D brought a game of him/her with me to your attention on OGS and claimed I cheated?
I’d probably delegate it to another moderator, since I don’t have a good enough computer to investigate the game myself. But if you had won against an appropriately ranked 5d as convincingly as this particular game, like I said before, I’d keep an eye on you for a while to see if it was a pattern, and ask the stronger dan mods to take a look.
That said, I assume this is just a normal OGS game, and not an official tournament setting. If things aren’t completely secure here, that is not as big a problem as when cheating goes unpunished in a tournament setting.
So how would you feel if the 5D, while you are investigating, publish this game in this forum and maybe disclose my true identity?
Personally, i am kinda surprised why they didnt require video recordings before this match. European tournaments which have actual cash as a prize have required video recordings or video chat between the players for some time now, at least within the top groups.
AGA city league has 14 000 usd prize pool this season, which is plenty enough reason to cheat. Or to assume that someone else might be cheating.
I would then give the 5d an official warning and remove the topic. Like I said, I’m not on board with the public shaming.
I would however expect that this 5d gets told that the case is (a) under investigation and that (b) the suspicion is legitimate. I would not think a case like this gets overturned by “it’s not cheating”. I was under the impression that the AGA was not ‘still investigating’ when this controversy started, and that the player was instead told that there were no reasons to suspect cheating.
But even if that were to happen, and the 5d would complain on the forum whether our ruling was fair, without naming the particular game, or only by showing the AI analysis, that’s totally fair in my book.
And once again: OGS is not a tournament setting, these things are different.
Get in the car, Kata! We’re taking a trip to the Pandanet!
Dinershteyn left a comment about video control. He as team captain in European Team Championship suggested to organizers to enforce video connection through skype but they didn’t answer anything. And he’s individually offering opponents to video-connect.
One thing I’d like to see, just to test a hypothesis, is games of 6 dans against »Bender« and »Prof. Farnsworth« on https://leela-one-playout.herokuapp.com/
Things that keep me up at night: Whatever happened to this?
But that’s not the reason I’m zombiethreading, it’s because I found this
https://min.news/en/sport/01131c081240cc3df644e564c4dac24f.html
A. Posting translations below a text that doesn’t need translation and having differences is hilarious. B. who is Jin Lunying.
Accused person didn’t play another game I see. Pandanet: Pandanet-AGA City League
(Seriously tho, I’ve been thinking about this tea gossip scandal, especially since AGA congress is around the corner now, any more information emerged?
If AGA cheating regulations were under discussion, it’s time they announced new policies or something, no?)
I think the AGA handled this case very poorly, here’s a copy of a comment I made on Facebook back then:
Now that I’ve got over my surprise. Let’s analyse the severity of the infractions:
- Remi made a public facebook post. He posed it as a leading question and didn’t directly say the other guy cheated, though it’s clear that’s what he thought. He did this before concluding some official process for dealing with disputes/suspicions of cheating, but this was we hear borne out of frustrations with that process doing nothing before: no one found to have cheated, no stronger anti cheating procedures. If there was a well-working process and he directly accused maybe a 3/10 badness, as is I’d call it a 1 or a 2.
- Tong maybe cheated with AI. Let’s say 50% chance for sake of argument (I think higher). Cheating with AI is one of worst things you can do in a tournament, I suppose poisoning opponent’s tea, shooting them in face when they beat you is worse, so let’s say 8/10 badness. Risk adjusted a 4.
- Yoonyoung and Manuel are friends with Remi and probably knew about / supported his post?
- Avery and Jianqiu also on the Montreal team, did they know?
The sanctions received:
- Tong, nothing (ditto his teammates).
- Remi, Yoonyoung, Manuel, Avery, Jianqui all kicked out of the league
Hmmmm…
Most of the early posts in this thread seem to think Remi was wrong to post publicly (focussing on simply the game result: 6d whines about losing to 1k, rather than looking at the moves) and not so focused on the possible cheating (because it can’t be proved?). Are my relative badness judgements of 3 for a public cheating accusation in a system with good anti cheating measures, 1 or 2 for system with poor anti cheating measures, and 8 for AI cheating substantially different from those posters? I think it’s beneficial to make those axioms explicit.
And here’s an email I sent to the AGA, to which I received no response.
To the AGA,
I am writing to express my shock and disappointment at the handling of the recent suspected cheating case in the AGA city league between Remi Campagnie and Tong Qiu and the decision of the organisers to eject the entire Montreal team as punishment and demand removal of Remi’s post. This significantly damages the reputation of the AGA to me. I have previously been an AGA member, not to play in tournaments as I’m in England, but to support Michael Redmond’s excellent AlphaGo videos. This incident ensures I will not renew my membership (the Transatlantic match incident had already reduced the chances).
My understanding of the situation is based on Remi and then Yoonyoung Kim’s posts on Facebook as I could find nothing official on the pandanet league or AGA websites.
That no definitive proof that Tong cheated was found is a vacuous conclusion: to prove AI cheating online just through game records without physical evidence is nigh on impossible. A judgement has to be made on the balance of probabilities. I think the finding he didn’t cheat so no punishment is applicable to him is likely wrong, but it’s at least understandable.
What I find inexplicable is the decision to punish the entire Montreal team with ejection from the league. Remi’s infraction was apparently to violate a code of conduct by posting publicly. Yet I hear the code of conduct does not actually prohibit this, just requires reporting to the TD which he did (but not only that). Even accepting the rashness to post publicly with the cheating suspicions (and he didn’t even directly accuse Tong of cheating though it was implied) before concluding the official process (which I hear had not previously found someone to cheat or lead to stricter anti cheating measures, hence the frustration and public post), this is a fairly mild infraction totally at odds with the harsh punishment applied to the entire team.
I look forward to a response.
Yours faithfully,
Andrew Simons
I don’t know how the AGA arrived to their decisions in this matter and we may never know. But I agree that the AGA appears to have handled this very poorly.
I understand that it’s very difficult to prove beyond any doubt that Tong cheated. But how can they deny that it was highly suspicious for a 1k to play a whole (rapid) game at world class level, except for 1 move?
The only reason I can think of is that they perhaps realized that they made a mistake by mostly ignoring the possibility of cheating, and then decided that admitting that mistake might damage their reputation. Ofcourse, that decision hurt their reputation even more.
To a degree, I understand their predicament. For this game, they could have decided to annull the game as a compromise, because cheating cannot be proven beyond any doubt. That might have been acceptable to most involved.
But that would still leave the question of how to continue the tournament. If they opted for such a settlement, disqualifying Tong would still be seen as a guilty verdict, so they could not do that. And letting the tournament continue without additional anti-cheating measures would also be a problem, because similar cases could come up.
Should they have just aborted the whole tournament? That would have disappointed many players and fans.
Perhaps they could have paused the tournament and improve the anti-cheating measures, such as requiring players to record themselves? But it’s quite likely that changeing the rules mid-way would also endanger the continuation of the tournament. Players might drop out, because they are not willing to record themselves (for various reasons).