Bill Spight wrote:BTW, is it necessary to go through KGS or any other public go server? A dedicated site could give the tournament organizers better control and might even be more efficient.
I've been advocating for a private, dedicated tournament server for years.
This sounds like a good use of AGA funds.
As Oberlappen noted earlier, the AGA players do not have the same access to face-to-face tournaments that their EGF peers do. So it seems a natural goal for the AGA to ensure that there are online opportunities for their top players.
Ok. I'm reading a lot about the circumstances regarding this situation and how that affects the decision that should be made here, and I think that's really approaching this situation from the wrong perspective.
Tournament policy is founded on tournament philosophy, namely, the idea that the integrity of the tournament must be protected above all else.
This situation should be looked at like this: In a game that was not yet completed, one player ran out of time on the clock, and as a result, the other player was declared the winner by the server. The player who ran out of time claims that they made their move in time, but the server lag prevented the move from being made.
Who each player is and how trustworthy they are is completely irrelevant. The board state of the game is completely irrelevant. Nothing else matters here, because we are trying to determine tournament policy, and basing policy around given circumstances unless they're truly unusual or unique is a bad way of creating policy.
The first thing to do is to try and find proof of the player's claim. If there is evidence that the server that lag did occur and a win was falsely given, then that is certainly a reason to deviate from the given result.
If that is not the case, then in lieu of better options, you must accept the timeout result. If you want to implement ways to check or deal with lag, that is something that must be done in future tournaments. With the policy you have, you have to use what the clock has given you, or else the clock is meaningless, and the integrity of the event is compromised.
If Mateusz was not himself but another player with a history of cheating or underhanded play, you must be able to make the same ruling.
If the game was not almost decided but instead still very much in question, you must be able to make the same ruling.
You cannot resume play from the same position here because you wouldn't have resumed play from the same position had this happened from a board state that was still very much in question, the player would be compromised and would both be able to consult outside assistance for the position, something that compromises the integrity of the game and the event.
You cannot rematch this game, because at any other point in time, for any other player who claims to be experiencing lag and times out as a result, you cannot rematch all of their games, for logistical reasons and common sense.
Regardless of how trustworthy a player is or how good their track record of integrity is, you cannot as a tournament organizer take them at their word if you aren't ready to take every players' statements as fact when you're issuing a ruling like this, it would be blatant favoritism if you treat any player differently, and so when you make these rulings you have to discard what you know about them and what you know about the game state. Those things are completely irrelevant here.
Unless there is evidence that supports Mateusz's claim that him running out of time was due to KGS and not him, you have to accept the KGS result as the result.
tpgtc wrote:A tournament team that includes officials from the EGF, the AGA, KGS, and the tournament referee (Myungwan Kim 9P) is still reviewing the case of the 4th match. We will announce the official decision as soon as it's confirmed. Thank you for your patience.
Violence wrote:
The first thing to do is to try and find proof of the player's claim. If there is evidence that the server that lag did occur and a win was falsely given, then that is certainly a reason to deviate from the given result.
...
Unless there is evidence that supports Mateusz's claim that him running out of time was due to KGS and not him, you have to accept the KGS result as the result.
What do you propose should happen in cases where there is evidence of timeout due to KGS lag?
What do you propose should happen in cases where there is evidence of timeout due to KGS lag?
That I'm not so sure about, and I think that's far more open to discussion. I think things should be decided immediately, and because we're here a week later, it may just be that a full rematch is the only recourse. In the future, if evidence of lag can be ascertained immediately, I think resuming the game on the spot so long as there hasn't been any lapse of attention from the proctors would be a reasonable thing to do by default, with the full rematch only being done in situations where integrity of the game has been potentially compromised.
I still don't understand the reference to "Kazan"...
Most likely it is related to some event in European Go history.
I find it poor taste to perpetuate such a stigma, though, regardless of what may have happened in the past.
Violence wrote:Ok. I'm reading a lot about the circumstances regarding this situation and how that affects the decision that should be made here, and I think that's really approaching this situation from the wrong perspective.
Tournament policy is founded on tournament philosophy, namely, the idea that the integrity of the tournament must be protected above all else.
Tournament policy is beyond the scope of the referees' authority, as referees. I agree that it's important, but that's not the question they face.
This situation should be looked at like this: In a game that was not yet completed, one player ran out of time on the clock,
Given the lack of a ruling as yet, I doubt that any clock was specified as the official clock.
and as a result, the other player was declared the winner by the server.
As a tournament director, I doubt if the server was given the authority to declare the winner.
Who each player is and how trustworthy they are is completely irrelevant.
That's why there were monitors.
The board state of the game is completely irrelevant.
I agree.
Nothing else matters here, because we are trying to determine tournament policy,
Tournament policy is beyond the scope of any inquiry.
The Adkins Principle: At some point, doesn't thinking have to go on?
— Winona Adkins
Violence wrote:...
Tournament policy is founded on tournament philosophy, namely, the idea that the integrity of the tournament must be protected above all else...
Embedded in this is the idea that policy trumps everything else, an idea with which I do not agree.
In a prefect world, following policy is a great idea - especially a policy written by someone with godlike wisdom. But experience has show that we mortals are imperfect, and that no policy will always cover every event.
Every system of rules has to have some final arbiter(s) who can make wise decisions by the seat of their pants when policy fails.
The integrity of the tournament is a laudable goal, but its last line of defense must be people, not policy.
In other words, I disagree with Violence's premise, and thus, with his conclusion.
We had a failure of policy. It is time for the judges to fill in the gap.
I still don't understand the reference to "Kazan"...
Most likely it is related to some event in European Go history.
I find it poor taste to perpetuate such a stigma, though, regardless of what may have happened in the past.
I genuinely cannot tell clever humour from serious discourse in most circumstances, so just in case: a) this article is satire, b) Mateusz currently resides in Kazan, which is also brought up as the reason behind the lag and a potential future excuse in similar situations.
Sorry if I got whooshed.
https://www.facebook.com/groups/go.igo.weiqi.baduk/permalink/10157361922046514/ wrote:Update on the transatlantic situation from a source close to the organizing team: Still not resolved because both teams/federations have differing views. A decision had been made but has been appealed since. They will discuss it next week in person at the IMSA tourney in China, hence the delay in a decision.
I read today that some people will meet during the IMSA tournament to try and agree on what to do. Internet tournaments do have these problems built in. As I recall it, there was a UK player in the Pandanet Team event who was losing over 10 seconds per move just due to lag. It's a pity that this event, which is probably supposed to be something of a celebration of the young professional structures, has to deal directly with such base issues.
I suppose that there are 4 main options to pick.
* Win due to timeout
* Restart the game due to faulty clock
* New Game
* Double Loss
Whichever one is chosen, at least we can complain that it is the wrong one.
The proctor is a neutral person obeserving the player. One technical solution would be to use client side timing. The EGF and AGA could provide a laptop to its player pre installed and given to the player by the proctor who would after the game take the computer back. The player would have no chance to install cheating clients etc.
Matti wrote:One technical solution would be to use client side timing.
Correctly, I mentioned it earlier.
When one side receives opponent's move, counting starts and is executed locally.
When move is played, move and time stamps are sent and compared with server time.
Also, it would be good to be able to customize byoyomi reading.
Matti wrote:The EGF and AGA could provide a laptop to its player pre installed and given to the player by the proctor who would after the game take the computer back. The player would have no chance to install cheating clients etc.
If a proctor is present, even with some kind of cheating client, it would not be so easy to activate cheat.
