I am wondering what the most popular option is.

_________________
“Discussion is an exchange of knowledge; argument an exchange of ignorance.” ― Robert Quillen

After reading Ing's rules, I voted superko. And among superkos, my prefered one is the positional superko, because it is the easiest one to define, understand and apply.

While I kind've like the special cases of "no result", positional superko just always seemed more elegant.

_________________
Tactics yes, Tact no...

I prefer either a draw or no result if neither player will back down. It seems too easy to get lost in the ko rotation and play a ko out of sequence in byo-yomi when playing over the board, losing the game through an illegal move. If that's going to be the case, you may as well flip a coin or something.

I agree it's not ideal for tournaments, though. In that case, I almost prefer "both players lose", but that too creates havoc with direct elimination. It comes up infrequently enough, though, that I think it's manageable.

Of course, I also dislike the idea of a pass as a valid move, so take my words with a grain of salt.

I agree, with a superko rule, an illegal move should be taken back, and not lead to loose the game.

As far as I know every major ruleset with superko regards an illegal move as a pass. So if there is a superko situation, the person that makes a mistake loses the superko but can still catch up on the rest of the board, which seems fair.

_________________
“Discussion is an exchange of knowledge; argument an exchange of ignorance.” ― Robert Quillen

Well, superko is only really relevant if the ko is game-deciding. If it's not, why wouldn't the winning player ignore it to claim victory, assuming decently skilled players?

I wouldn't say that only game-deciding kos are relevant. Large kos can appear early in the game and the losing player can still go on to win if they do better elsewhere.

Let's suppose there is a 15 stone black group living/dying in a superko and there are local threats. Also, no tenuki is worth more than 10 points. The ko is significant, but even if the black group dies, black can still win if (s)he manages to kill a 30 stone group later.

_________________
“Discussion is an exchange of knowledge; argument an exchange of ignorance.” ― Robert Quillen

I voted for super ko, but the choice could be also left in the tournament rules. If the tournament schedule allows extra games, no result could be used. If the tournament is not knock out a draw could a viable option too.

I voted draw. I got a triple ko game with a friend, it was funny (it is a long history) and it is good result draw, that I think that it is the same that "no result" if it is a non-tournament game.

If it is not a tournament game then the choice is irrelevant. The game is over and there are no consequences.

If it is a tournament then they definitely are not the same, as you imply. If a play won all his other games in a 5 round tournament then a draw would give 4.5/5 wins = 90%. A no result would give 4/4 wins = 100%.

_________________
Still officially AGA 5d but I play so irregularly these days that I am probably only 3d or 4d over the board (but hopefully still 5d in terms of knowledge, theory and the ability to contribute).

I recently discovered the abominable "molasses ko". I knew that it had the ability to slow down the speed of the game to 20 % of the normal speed, but I didn't know that under strict superko rules (PSK, SSK or NSSK) the first player to pass looses the group !

So, in AGA, french, british, neo-zelandese or simplified Ing rules, both players are forced to play stones in their own territories until one of them has no legal move left. Their strategy is then to build shapes that reduce the number of legal moves of their opponent. This concept doesn't exist in japanese rules (the molasses ko leads to infinite repetition, and the game is voided), or chinese rules (any repetition more complex than a ko or a sending 2 returning 1 can be voided by the referee).

From an academic point of view, if we want to respect asiatic traditions, it seems that having the game replayed is the wisest choice. But I can't see how this could be done in a real-life tournament where already normal rounds never finish in time.

In conclusion, leaving the decision to the referee, like in chinese rules, is probably the best thing to do. If there is enough time, the game may be replayed. Otherwise, it may be a draw

super ko is the most elegant, but really it should be draw.

Very subjective. I find super ko to be anything but elegant as it is contrived.

_________________
Still officially AGA 5d but I play so irregularly these days that I am probably only 3d or 4d over the board (but hopefully still 5d in terms of knowledge, theory and the ability to contribute).

not really. superko can be a rule as simple as you can't repeat any previous board state. making more than one repeat a draw is contrived. but only the most serious go players can really keep track of triple ko which is why i think it should be draw.

Since triple ko is an unbranched 6 move sequence, it is sufficient to recall the initial position of the 3 ko stones. This is MUCH easier than almost all go problems. Therefore one should not say that players would not be able to play triple ko correctly. It becomes difficult only if a player first plays a couple of moves WITHOUT THINKING and then starts to wonder which the initial position was. Stupidity!

If you need an argument about difficulty, use quadruple ko as an example.

I think
1) a game should be finite
2) in finite play, jigo should be possible
3) a game should only count as won if player A can force a result in which B ends up with fewer points

If neither player can force a winning position in finite play, the game should be a draw.
Here, it's irrelevant whether the two players formally agree to a draw - after all, if either could force a win "naturally" by playing until there are no legal moves left to play, they would. By refusing to play a finite game (continuing the multi-ko), the players already implicitly agree to a draw.

As to how that draw should be dealt with in a tournament setting, that's a different question. Obviously most organizers don't want to deal with draws at all.

Moreover, I would not want to have to deal with a rule that forces me to remember if a given position has occurred before and if it was my move or my opponent's... that's neither elegant nor logical, just convenient - but not for the players.