Life In 19x19

mhlepore wrote:We should roll a die to determine what happens:

1 = Eric wins right now
2 = Mateusz wins right now
3 = Resume game
4 = Rematch
5 = both are eliminated and next players are up
6 = nothing - roll again until get 1-5

I'm not even joking. Making it random will take away people's ability to complain about the organizers' rationale when they reach their decision. This is a crappy situation, and there is no consensus, so roll a die/flip a coin and move on with life.

One more option occured to me:
Consider the move which the server decided to be too late as a pass and continue from it.

Matti wrote:...

One more option occurred to me:
Consider the move which the server decided to be too late as a pass and continue from it.

Since Lui's last recorded move was defensive, this has virtually the same effect as resuming with Surma to move. ( The score might be a point or two different, but still not enough to alter the inevitable result. )

mhlepore wrote: But some for some people, the best solution is trying to find the most proper application of the tournament rules, and for other people, the best solution is to compromise.

I think the dynamic here is rather different to a typical tournament with a referee making decisions. In your typical real-life individual tournament you have an organiser and referee who make an event, players register and turn up and are supplicant to the organisers. It's reasonable for the referee to issue rulings with an air of "My tournament, my rules, if you don't like it, don't play in it".

This event is rather different. For starters there don't seem to be clear rules (at least publicly, and presumably privately or else the decision would be easy/obvious and accepted). But also it's not like this event existed without the two teams registering as one among dozens or hundreds like an EGC. It exists only as a collaboration between the 2 teams/organisations. So a dictator referee saying "Sod off if you don't like it" makes the event cease to exist, not carry on with 99 instead of 100 players. So finding an outcome that both sides can accept and preserving goodwill is relatively more important.

P.S Joaz it is Lui and Surma.

Uberdude wrote:I think the dynamic here is rather different to a typical tournament with a referee making decisions. In your typical real-life individual tournament you have an organiser and referee who make an event, players register and turn up and are supplicant to the organisers. It's reasonable for the referee to issue rulings with an air of "My tournament, my rules, if you don't like it, don't play in it".

Well, no. Speaking as a former "dictator" the right attitude is to issue rulings from the book, i.e., from what rules you have that the players could have consulted and which you can use to justify your ruling. If there is no book, we are back to pre-1949 days. Having given a ruling, you then inform the players of their right to appeal your ruling, which they can do immediately by presenting an argument to you or even by a simple request to reconsider. If you then reconsider and either repeat your ruling or change it, they can appeal that ruling to higher authorities. Maybe not that exact procedure, but some reasonable procedure allowing for appeals, unless the rules explicitly state that the referee's ruling is final.

It appears that the issue is being appealed and there will be a final ruling in time.

Rules about appealing must also depend on the rules :-p

I'm going to invoke Joaz's Uncertainty Principle ( which, very conveniently, I just invented )

It says that whenever you have an unresolvable uncertainty, you fork the world. You create one world in which the cat lived, and one in which it did not. In this particular application, we create one world in Lui wins on time, and another in which Surma wins over the board. That should satisfy everybody.

In practice, world-creating is not necessary - and the AGA probably doesn't have the budget for it. So we'll just fork the tournament.

Instead of one game, we'll have two. On one board, Lui plays the next EGF member in line. On the other board, Surma plays the next American.

The games can probably be concurrent. On occasion, one player may find himself on both boards simultaneously. We'll allow him to delay one game until he finishes the other. ( And if he wins both, he can collapse to one board )

This produces a result that everybody can live with.

Just for compare and contrast, an experience of lag woe in an official competition. Nothing could have been done there, for the server could certainly not have been changed here. Perhaps there were some protests, but they were ignored.

I think the situation here is rather different. It's not that KGS is better or worse at handling lag, it is that here you have the option to make an intelligent decision. You have 1 game to manage, and referees you are supposed to be trusting, well okay you decided to call them protors. Of course, it depends on your moral compass, so to speak. What would your judgement be in this dispute?
Which do you respect, the game or the rules?

Javaness2 wrote:Just for compare and contrast, an experience of lag woe in an official competition. Nothing could have been done there, for the server could certainly not have been changed here. Perhaps there were some protests, but they were ignored.

The site states that the player lost anyway, therefore the match had been ruled to continue already, no?

Javaness2 wrote:I think the situation here is rather different. It's not that KGS is better or worse at handling lag, it is that here you have the option to make an intelligent decision. You have 1 game to manage, and referees you are supposed to be trusting, well okay you decided to call them protors. Of course, it depends on your moral compass, so to speak. What would your judgement be in this dispute?
Which do you respect, the game or the rules?

Moral compass doesn't come into it, really. Technical issues should not call games. As for Jasiek's 'dilemma':

The game was played. Shortly before the end of the game, Robert was estimated to be about 30 points behind. Two successive passes occurred (first by Robert, second by Csaba). Then Robert Jasiek proceeded with board-plays to remove all those stones that Csaba Mero could not save (i.e., from a strategic point of view, dead stones inside his own territory). While he was doing this, Csaba continued to pass every move. When Robert had removed all stones he could, he also passed. Csaba then passed (3rd pass in a row) and Robert also passed again (4th pass in a row). Robert made the claim that under the 1991 Ing rules, all stones still on the board after four consecutive passes are considered alive according to the rules, and the position should thus be scored as such.

This is a well-known and well-researched case where Jasiek tried to abuse the rules and win by "HA! HA!" I would be more sympathetic if he had done it in a game where he was winning just to get the stupidity of the Ing rules out in the open. There are a lot of reasons why EGC uses modified Ing rules, this is just one of them. In addition, the case as quoted in EGF referee training also included the little detail after the first 2 passes, Jasiek disputed the game's status as being ended, then after two more passes he claims the All-Are-Alive rule comes into play.

The stated reason for the rule is, I believe, in the case of disputes it might make sense to have a different set of rules; complex seki or 10,000-year ko or other complex situations arising in the board can lead to a situation where one player will be forced to defend their claim that a group is alive or dead, or accept that EVERY STONE in the board is alive. Just for reference, a 20-point territory with a single opposing stone would be considered seki as it is technically space touched by both sides' living groups.

Thankfully the judges ruled Jasiek's "dilemma" as "technical bullshit" and told them to get on with the game and count normally. This is, again, equivalent to the situation at hand; don't call matches based on technical bullshit, play go and not "who has better internet connection" or "Who knows the rules of excited ko minus one better."

Disclaimer: I am not actually sure if this rule still exists in the modified Ing ruleset. I do remember on one occasion warning players who had a board state dispute that if they continued the game they would have to remove all opposing stones from the board as they would be accounted alive. If I recall correctly they stared at me for a while and asked if I could please forget that they had both passed once already. Miraculously, they counted points normally after that.

Anyway, in all of the cases above, it was decided that the game should continue normally as if the technical disturbance had not occurred in the first place. As it should be, and as it should be in this situation as well.

Javaness2 wrote:What would your judgement be in this dispute?
Which do you respect, the game or the rules?

I have expressed myself on that question. IMHO, there is no need to choose between the game and the rules. The rules define the game, after all. The guiding principle of the TD's rulings is to restore equity, if possible, not to punish transgressions. A ruling to restore equity was not only possible, but easy to accomplish, and the game could have been ended with the normally expected result. Most players do not understand the reason for the Ing four pass rule, treating it as just two extra passes for no reason. I do not want to get side-tracked into a discussion of Ing rules. The matter has been discussed to exhaustion on SL.

The problem is that we players do not always agree on what is fair. Over the board this was an obvious case for resumption. Over the internet (culture) that obviousness is lost. A rematch would be the next diplomatic attempt to solve the problem. If a rematch isn't agreeable, you would have to give a double default and cancel any future event in 2020.

Javaness2 wrote:The problem is that we players do not always agree on what is fair. Over the board this was an obvious case for resumption. Over the internet (culture) that obviousness is lost. A rematch would be the next diplomatic attempt to solve the problem. If a rematch isn't agreeable, you would have to give a double default and cancel any future event in 2020.

Resumption would not be equitable because the property of being in byo-yomi has been lost. OTOH, if the parties agreed to a resumption, that would resolve the issue. A rematch would not be equitable, either, because the state of the game has been lost entirely. But again, agreement by the parties would resolve the issue. But apparently the parties have not reached an agreement. Neither player is at fault, and neither should be penalized. Whatever happens, the problem of netlag needs to be addressed before any more games are played.

Joaz's parallel universes idea is very creative.

A couple of years ago I lost a PGETC match on Pandanet because of a timely with my opponent's move. I thought he/she was still thinking, but the server was counting down my time.

Eventually, my client clock started, but with insufficient time for me to make my move, so I lost.

We did enter an appeal, but this was rejected.

The reality is that there can always be glitches on the Internet that cause problems and we will always have unfortunate results in these cases. It's just a feature of this real world...

In this case the result should be that Mateusz has lost. However, if I were the AGA I would then declare that the result of the game should be void. (I'd hope then that the Referees would then declare that the match should then move on to the next players rather than replay this game.)

It is not impossible to resume the game in byoyomi. It is just really unusual because so much time has now passed.

Javaness2 wrote:It is not impossible to resume the game in byoyomi. It is just really unusual because so much time has now passed.

Well, IIUC, it is so near the end that at this point the players could finish the game in a minute or two.

mumps wrote:The reality is that there can always be glitches on the Internet that cause problems and we will always have unfortunate results in these cases. It's just a feature of this real world...

No, it's not just a feature of the real world. Having the server keep the official time is a choice.