Order of play after passing

Let’s say that black passes, then white passes, and the game is finished.

It came to my attention that if black asks for the game to resume, then black will play first.

This seems contrary to the rules : Article 1. The game of go

If a player requests resumption of a stopped game, his opponent must oblige and has the right to play first.

Beyond the rules, it also just seems like a bad idea. If a player made a mistake on the status of a group and realizes it during scoring, they should not be allowed to request the game to resume and play first to fix their mistake.

4 Likes

Yes, it is a known bug that contradicts the Japanese rules.

However, under other rulesets, the game resumes this way to settle life and death disputes.

4 Likes

If the order of play matters, then even the outcome of the game could possibly depend on who gets to play first. Hence, neither player would like to ask for the game to be resumed, if an unsettled and game deciding situation might exist on the board.

Technically, under the formal application of the Japanese rules, both players lose. Unfortunately, OGS does not support that outcome either. However, as a workaround, we can annul the game (since neither player should deserve the win) and make a note in the chat clarifying that the game was technically a “both players lose” situation.

Thanks - is that one of those situations where it seems very easy to fix, but is actually pretty hard?

I have never been able to find a document stating who should resume play first under French rules.

I can only assume it should be the same as under British or AGA rules, but that’s more a guess than a fact.

My own understanding is that the game should resume where it was (not jumping someone turn) but now i read that is not what japanese rules stipulates.

Yes, it’s not that simple with true territory scoring, where a stopped position could have seen analysis or hypothetical play before resumption (new info for players). This may be less relevant on OGS though.

Otoh an autoscoring oracle can create a new situation here - something that area scoring rules (with their simpler resumption approach) were not designed for.

I think technically I should have said “most other” rulesets. I guess the Korean rules, which are functionally very similar to the Japanese rules, may have a similar clause about giving the first move to the opponent of whoever requested the resumption.

The AGA and BGA rules do explicitly say something like:

  1. DISPUTES
    If the players disagree on the status of one or more strings after they have both passed, then alternation is
    resumed according to the following prescription:
    i. The opponent of the last person to pass has the first move of the resumed game.

I’m not sure what the official French rules of Go are.

1 Like

This seems very relevant to OGS, no? When both players pass you see the autoscore, and this might indeed give you new information and prompt you to resume the game.

1 Like

This kind of makes sense, but requires clarification with respect to pass stones. White has to pass last, which means that if White passes first, then it goes:

  • Black plays on the board
  • White passes (and gives a stone)
  • Black passes (and gives a stone)
  • White passes a second time (and gives a stone)
  • Players discover that they disagree about the status of some stones
  • Play resumes with White playing first, which means White has played twice in a row, including a pass stone

I assume the BGA and AGA rules take that into account, to avoid stealing an extra point from White?

As for the French rules… I believe that there currently exists no full text of the French rules available online, at least not linked from the French federation’s website.

The closest thing to an official rule I can find is this page: Jeu de go - Règle française du jeu de go which is linked from the French federation’s website with a link named “Règle officielle” (official rule) but looks more like a rule introduction for beginners rather than a very precise rule for tournaments.

Yes, depending on autoscore behavior (I’m not sure how the server behaves atm).

But as I wrote this may be more of a new challenge for area scoring rules, since they were never designed for such external analysis and its consequences (assuming aggressive autoscoring is used there as well - is it?). For territory scoring it could suffice to follow rules as written (even if the server doesn’t atm).

That is not theoretical; I have seen it happen many times, primarily among DDK players.

1 Like

Yes, how this works with respect to “White passes last” needs to be clarified. Basically, that extra pass is only applied (once, if at all) after the players are done with resuming the game and are finally counting the score with agreement on the status of all of the stones. Further, “White passes last” is only needed if using territory counting. If using area counting, the accounting trick of pass stones can be omitted entirely. The pass stones and provision of “White passes last” is just a counting trick to make the territory count agree with the area count.