Can any move be a ko threat
Bill: Several strong amateurs have expressed the opinion that any move can be a ko threat. Under certain circumstances, any move can act as a ko threat. I coined the term virtual ko threat to address that question.
However, I object to the statement that any move can be a ko threat, if made without qualification, on logical grounds. If any move is a ko threat, then the term, ko threat, has no meaning. What's the difference between having a ko threat and not having a ko threat?
I do not wish to misrepresent anyone's views, so I shall say no more for now. But I have started this page for discussion and clarification of the issue, and invite responses.
Malweth: To use an extreme example: It's unreasonable for the 1-1 in an empty corner to be considered a viable ko threat. A ko threat must be able to live, escape, or have significant aji (assuming it's played as a placement or otherwise deep in the enemy's sphere of influence) in order to be viable.
Malweth: At a certain point, wouldn't many smaller ko threats result in a big enough gain for one side or the other? Additionally, there must be points that are still invalid as ko threats because it would take more than 2-3 moves to finalize the position. In any case, the premise must still stand - the literal discussion of "Can any move be a ko threat" must be answered no - there are sometimes valid moves on the board that are ineffective as ko threats, no matter what type of ko is being played. The practical discussion of "Are most moves valid ko threats" is yes, but care must be taken such that they are worthwhile ko threats.
Bob McGuigan: Since there seems to be a certain logical absolutism to this discussion it is clear that every move is a ko threat. Not necessarily an effective ko threat, though. Even worse than the 1-1 point move mentioned above would be a move filling one of the two eyes giving life to a big group. The problem seems to me that general usage considers moves that are smaller than the ko, perhaps made by mistake, as ko threats, just not effective ones. But it is also true that when we count ko threats to decide which side wins the ko we only count moves that will be answered. The problem is complicated by the fact that if one side is ahead by an amount equal to the ko she may be willing to lose the ko in exchange for a relatively small threat.
Bill: To be fair, I do not think that people who say that any move can be a ko threat are talking about bad play.
tderz: I feel addressed and was not "talking about bad play", rather having the meaning of Bob McGuigan's last sentence. Bill also stresses in KoThreat/Discussion the point, that it's not so much about winning the ko, rather winning during the exchange. If a "normal move" does (and you have nothing better), so be it.
Bill has written in KoThreatFunctions and virtual ko threat something which I interprete as that what I wanted to express.
Taking a ko can even be a ko threat (not for the same ko of course). ~srn347
Moved from Molasses ko
What is a ko threat? Depending on your definition, every tenuki or pass could be called a ko threat. More generally, one speaks of restriction threats. If passes are involved or threatened to be involved, then one calls the particular restriction fights pass fights. --RobertJasiek
Bill Spight: I see Robert's point. If a play removes the restriction against taking a ko, then it carries the threat to take the ko. In that case, any board play, and, according to some rules, any pass is a ko threat. However, such a definition is so broad as to be meaningless. The threat in a ko threat is to make a profit somewhere other than the ko if the opponent does not respond, not the threat to take the ko.
Sometimes certain plays which are not, properly speaking, ko threats, perform a function of a ko threat. I call these virtual, or tertiary ko threats. For more, see Komonster.
RobertJasiek: As I am currently defining ko in general, not each play is a ko play. So you do not need to fear that any play or pass was a ko threat and that it was meaningless. Be patient; I will publish a general definition of ko and maybe even soon:)