The Out-of-Order Corollary

[Zamiell]

This is not a new convention. It is the way that we have already been playing for the last 4 years.

The Out-of-Order Corollary

• First, see the section on the Out-of-Order Play Clue.
• As specified in the sections above, if Alice gives a clue to Cathy touching the next playable card, but the focus is wrong, Bob would normally give a OOO Fix Clue instead of blind-playing a card.
• Thus, if Alice gives a clue to Cathy, and Bob does blind-play a card, then by default, Cathy can mark all the cards that were touched with a note of not being the next playable card.
• For example, in a 3-player game:
  • Red 2 is played on the stacks.
  • Alice clues red to Cathy, touching two cards on slot 1 and slot 2.
  • Bob blind-plays a blue 1 and it successfully plays on the stacks.
  • Cathy marks her slot 1 card as the red 4. (Cathy knows that Alice performed a Bluff and slot 1 was the focus of the clue.)
  • Cathy also marks her slot 2 card as the red 5. (Cathy knows that if it was a red 3, then Bob would be forced to give a Fix Clue.)

[sjdrodge]

This is not a new convention. It is the way that we have already been playing for the last 4 years.

This may be the way you have been playing, but it is not the way everybody has been playing.

This strikes me as part of a series of changes designed to wring all of the context and judgement out of our convention system.

Currently, it's quite common that when r3 is on the stacks and Cathy holds r4 g3 y2 r5, Alice clues red to Cathy and Bob blind-plays (thinking he's providing r4). Bob makes this assumption because it would be extremely weird for Alice to clue red instead of 4 or 5 when both of those clues are unblocked. Furthermore, if Bob ends up playing something other than r4, Cathy should not go on to assume that slot 1 is trash. Cathy is still supposed to apply Good Touch Principle.

It is not possible to enumerate all the scenarios where a lie should be deemed "good enough," and we should not try to do so. Dictating that players should always assume a lie (complete w/ fix clue) in possible OOO scenarios is simple, but is seriously lacking in other merits.

[Zamiell]

as stated in the context section, the document just specifies the default or most common case. it doesn't dictate to players to not use context. if players think that a lie is good enough, then they should disregard the corollary, just like they would any other section in the document

[sjdrodge]

as stated in the context section, the document just specifies the default or most common case. it doesn't dictate to players to not use context. if players think that a lie is good enough, then they should disregard the corollary, just like they would any other section in the document

You misunderstand. The corollary you've written instructs players to assume that a lie has occurred that they must now fix even when they can see that the truth would accomplish the same task. This runs counter to the way we actually play, where Bob instead only assumes a lie if the same thing couldn't have been accomplished with the truth.

[Zamiell]

even when they can see that the truth would accomplish the same task

i've written no such thing

This runs counter to the way we actually play

ime, the way we play is that no-one would bluff this way, because it is undefined and confusing

[sjdrodge]

i've written no such thing

In the example given in my first message in this thread. 4 or 5 followed by reds plainly works, and the corollary that you have written instructs players that as Bob, if Alice gives reds instead of either of those two alternatives, they should give a fix clue instead of blind-playing.

In other words, even though Bob can see that the truth works w/ the same efficiency as the OOO lie, you've told players As specified in the sections above, if Alice gives a clue to Cathy touching the next playable card, but the focus is wrong, Bob must give a Fix Clue instead of blind-playing a card.

I am eager to hear how I've misrepresented what you've written.

[Zamiell]

if alice has a free choice between a red color clue and a number 4 clue, and chooses to clue red instead of a number 4 clue, then that is an undefined move, and bob is free to interpret that however he wishes. the document doesn't care and neither do i. if you want to propose that as a move then create a new issue.

the words "bob must give a fix clue" do not apply to undefined situations, because bob can do whatever he wants in undefined situations, its all up to bob

[sjdrodge]

If this change to the document is made as-is. The situation becomes defined. It becomes defined as Bob must give a Fix Clue and should not blind-play. That is a bad definition. Hence the problem!

[Zamiell]

It becomes defined as Bob must give Fix Clue and should not blind-play.

More specifically, Bob must give a fix clue and should not blind-play, barring any exceptional circumstances. (Just like every other section in the document.)

Cathy should not go on to assume that slot 1 is trash. Cathy is still supposed to apply Good Touch Principle

from what I know, this seems wrong. by default, cathy should go on to assume this. cathy should only break from the default interpretation if there is some other piece of context.

it's quite common

it is not common, my evaluation is that it falls under "exceptional circumstances" territory. link 3 games (with hyphen-ated members, no outside people) where it has happened however many replays you find, i will find three times as many with OOO finesses where Bob immediately gives a fix clue, proving that the document should properly describe the default case from Cathy's perspective

[sjdrodge]

I invite you to post three replays where Alice gives a clue similar to the one in my example and intends it as OOO. OOO Finesses have very little to do with the problems in the proposed verbiage.

It may be true to you that it's a highly exceptional circumstance that need not be stated that OOO isn't intended when the truth would work just as well, but the average reader of this proposed section will not come away with that impression.

If it is not explained to the reader that they should always consider whether the truth would work just as well as the lie before assuming a lie, then they will simply take the statement if Alice gives a clue to Cathy touching the next playable card, but the focus is wrong, Bob must give a Fix Clue instead of blind-playing a card at face value.

Also, as an aside, you gotta stop with this whole gatekeeping thing. If people are reading/contributing to the document and playing with the conventions, their input/experience is valuable. It doesn't matter whether they play with you or your friends as often as you'd like.

[Zamiell]

it doesn't need to be explained to the reader because it isn't the default case. if it were frequent enough, then sure, we could throw in a bullet point saying "hey theres this other move that sometimes happens that you might want to also consider". but im not going to throw in a bullet point for a move that doesn't exist, that would be silly. if you want the move to exist, then propose it in a separate issue.

[pianoblook]

ime, the way we play is that no-one would bluff this way, because it is undefined and confusing

ime no one would give a play-clue this way because it's undefined, confusing, and terrible. At least if Bob blind-plays it becomes undefined, confusing, and efficient 🤷

[pianoblook]

I'm imaging the Venn Diagram of (A) Players who know about the existence and nuances of OOO and (B) Players who would ever in their right mind decide to give such a clue in such a situation, and the overlap is very clearly zero.

[jack67889]

Let's go back to the original question, because imo it's already defined in the doc:

Level 23. Red 3 on stack. Alice clues Red to Cathy touching Red 5 on chop and Red 4 on slot 1. Bob responds by playing slot 1 as a known bluff. Cathy plays Red 4 from slot 1 since Bob’s blind play let her know that her red cards are in different order than she thought.

The doc has a new section on OOO play clues, which says:

Sometimes, a player will have two playable cards in their hand of the same suit, but they will be "blocked", meaning that both a color clue and a number clue will not be able to focus the first playable card.

That situation doesn't apply in this case, so by default it's a finesse/known bluff over an OOO play clue according to the doc, imho.

Moreover I think the main confusion/disagreement arises because the (older) section on Finesses with a Lie Component says:

The previously mentioned Out-of-Order Finesse is a specific example of a Finesse with a Lie Component. (…) Importantly, Finesses with a Lie Component are only allowed if there is not an alternative line that would "get" the same number of cards without any lies. Thus, if you are looking to do a Finesse with a Lie Component, make sure you carefully consider the efficiency of all of the possible alternatives.

To reflect the way the group has been playing I think it makes most sense that this efficiency requirement extents to OOO play clues. So for the example it would makes sense (in case there are no blocking cards) that you are still allowed to give an OOO play clue if Bob can give an efficient fix.

[Zamiell]

To reflect the way the group has been playing I think it makes most sense that this efficiency requirement extents to OOO play clues.

you are prescribing behavior for bob. if you want to do that, i think you should open a separate issue, because my thread is about clarifying what cathy should think, not bob. in the case where cathy gets a OOO fix clue, cathy has no idea whether her hand has a blocking card or not. and by default, she should assume that she does have a blocking card.

[Zamiell]

At least if Bob blind-plays it becomes undefined, confusing, and efficient

it sounds like you are prescribing behavior for bob. if you want to do that, i think you should open a separate issue, because my thread is about clarifying what cathy should think, not bob.

[jakestiles]

I think this new corollary is bad and it is not the way I play. In the future Zamiel, I think you should avoid claiming that everyone plays the way you play. I agree with Most of Stephen’s and timotree’s and Valletta’s points in the Convention-questions thread.

[Zamiell]

I think you should avoid claiming that everyone plays the way you play.

i've not done so. perhaps you are mixing up "most people" with "everyone".

[jack67889]

you are prescribing behavior for bob. if you want to do that, i think you should open a separate issue, because my thread is about clarifying what cathy should think, not bob. in the case where cathy gets a OOO fix clue, cathy has no idea whether her hand has a blocking card or not. and by default, she should assume that she does have a blocking card.

You are prescribing behaviour for Bob and behaviour for Cathy based on the behaviour of Bob. I'm pointing out the doc currently prescribes different behaviour for both Bob and Cathy.

Furthermore if the focus of the clue is trash it would be an UTD/UTC according to the doc, so that can't be the reason for Bob to play finesse.

For an Unknown Trash Discharge:

• The empathy on the clued card contains one or more possibilities that are useful and currently unplayable.
• Thus, something extra is needed to prove it is trash. Other than this, don't use more complicated factors to prefer a Trash Bluff over an Unknown Trash Discharge (like Bob having to wait on some other card).

[jakestiles]

• “that is the way that we play...we = the hyphenated group”
• “This is not a new convention. It is the way that we have already been playing for the last 4 years.”

These are the comments I take issue with. So far you are outnumbered on this topic. This topic is either solidly disputed or more people disagree than agree with you.

[pianoblook]

it sounds like you are prescribing behavior for bob. if you want to do that, i think you should open a separate issue, because my thread is about clarifying what cathy should think, not bob.

I'm confused. This post explicitly says that in situations like this, "Bob must give a Fix Clue instead of blind-playing a card." That sounds like you're prescribing behavior for Bob, no? And my post, inversely, is disagreeing that such a situation should be necessarily prescribed.

As myself and others have argued, it seems extremely reasonable for Bob - at least in a large portion of possible scenarios - to blind-play instead of giving a Fix clue.

Therefore, if Cathy is a good Hanabi player and realizes that there are many realistic scenarios in which Bob would be blind-playing instead of Fixing, it would be preposterous to assume she should automatically mark her other card(s) as Trash. That sounds like you're prescribing bad behavior to Cathy

[Zamiell]

jake, in this context, the word "we" does not mean every single individual person inside of the hyphen-ated group, of which there are hundreds of people. i can't claim to read everyone's minds or know exactly how everyone individual is playing. however, i am in a unique position to have played with a greater portion of these people than anybody else. so when I say "we" in this context I mean "most people". hopefully that makes sense

[Zamiell]

i updated the wording slightly:

- As specified in the sections above, if Alice gives a clue to Cathy touching the next playable card, but the focus is wrong, Bob would normally give a OOO *Fix Clue* instead of blind-playing a card.
- Thus, if Alice gives a clue to Cathy, and Bob does blind-play a card, then by default, Cathy can mark all the cards that were touched with a note of not being the next playable card.

You are prescribing behaviour for Bob and behaviour for Cathy based on the behaviour of Bob.

based on this new edit, no, i am not prescribing behavior for bob. rather, i am making a descriptive claim about what bob does >50% of the time when this situation occurs, from an entirely non-normative standpoint, from cathy's perspective.

I'm pointing out the doc currently prescribes different behaviour for both Bob and Cathy.

i do not think that the doc prescribes different behavior. for discussion purposes, let's call the move that Stephen is potentially proposing as a Faulty Bluff. the doc only defines what OOO, it does not define what a Faulty Bluff is. therefore from Cathy's perspective, she should only assume that defined moves have happened in her game.

[Zamiell]

This post explicitly says that in situations like this, "Bob must give a Fix Clue instead of blind-playing a card." That sounds like you're prescribing behavior for Bob, no?

no, i am not prescribing behavior for bob. see my edit

[pianoblook]

Here's what I'd expect to see if I came across this example in the doc:

For example, in a 3-player game:

• Red 2 is played on the stacks.
• Alice clues red to Cathy, touching two cards on slot 1 and slot 2.
• Bob blind-plays an unrelated card.
• Cathy marks her slot 1 card as the red 4. (Cathy knows that Alice performed a Bluff.)
• Cathy also marks her slot 2 card as red 3 or red 5. (Cathy in general should know that she shouldn't be prescribed to mark her card as not being red 3, since level 19 Hanabi players are well-versed in conventions such as Context and Finesses with a Lie Component.

I really don't understand why on earth Cathy should always mark slot 2 as never being red 3 - again, that feels like a bad prescription + convention.

[Zamiell]

I really don't understand why on earth Cathy should always mark slot 2 as never being red 3

cathy doesn't always mark her slot 2 as never being red 3. however, by default she does (in the absence of any context), because OOO exists, and Faulty Bluffs don't exist.

[pianoblook]

cathy doesn't always mark her slot 2 as being red 3. however, by default she does (in the absence of any context), because OOO exists, and Faulty Bluffs don't exist

This post clearly says she should mark it as being red 5 exactly:

Cathy also marks her slot 2 card as the red 5. (Cathy knows that if it was a red 3, then Bob would be forced to give a Fix Clue.)

We're saying that this statement is false and should not be the general default

[Zamiell]

This post clearly says she should mark it as being red 5 exactly:

i meant that she doesn't always mark her slot 2 card as being red 5.

[Zamiell]

We're saying that this statement is false and should not be the general default

it should be the general default, because OOO exists, and Faulty Bluffs don't exist. if you want to claim otherwise, then i think that Faulty Bluffs should exist in the document! call me crazy but I think that a reasonable position

furthermore, as explained in the section on context, the doc only prescribes the default thing to do when there is no particular context involved

[jakestiles]

Sure OOO exists but it does not declare itself to be mandatory or the only way to get out of order cards. In fact, its 2nd bullet point says OOO is Just one such way to get the cards.

I disagree that the doc currently prescribes this interpretation and I disagree that it should.

In my opinion, Bob is currently allowed and should be allowed to rule out OOO if the fix isn’t good or he has no clues to fix with. In my opinion, Cathy shouldn’t assume good touch principle has been violated just because Bob blind plays.

[jack67889]

however, by default she does (in the absence of any context), because OOO exists, and Faulty Bluffs don't exist.

I can repeat myself, but see my first post. Doc explicitly defines this to be not an OOO play clue, so it's definitely a finesse/bluff by default.

Again, I think a new section to clarify OOO play clues shouldn't change how the group should treat OOO clues in general. The efficiency requirement has been there, so that's how the group has been playing. (Didn't see an argument yet why this should be changed.)

[timotree3]

I don't think "Faulty Bluffs" need to be defined separately to exist conventionally. The definition of a bluff is enough

[Zamiell]

In my opinion, Bob is currently allowed and should be allowed to rule out OOO if the fix isn’t good or he has no clues to fix with. I

that's fine. again, this thread isn't about bob, it is about cathy. if you want to prescribe something new for bob, please open a new issue!

Doc explicitly defines this to be not an OOO play clue, so it's definitely a finesse/bluff by default.

i dont think so. if it isn't OOO, then it is undefined. if bob wants to interpret it as a finesse, then that's certainly a reasonable interpretation, but that isn't defined. and you shouldn't claim that it is.

Again, I think a new section to clarify OOO play clues shouldn't change how the group should treat OOO clues in general.

this new section does not change how the group treats OOO clues. what it does do is affect situations where players give undefined moves. i am not concerned with undefined moves in this thread.

I don't think "Faulty Bluffs" need to be defined separately to exist conventionally. The definition of a bluff is enough

tim, the way that we do things is that we give everything a different name. this helps for both comprehension and for discussion. why do you think we distinguish a Hard 3 Bluff from a normal 3 Bluff, even when they are extremely similar?

i dont think that you should blind-play in an undefined way and claim that "it's just a bluff". if it is a common move, then come up with a name for it, define it, and put it in the document!

[pianoblook]

mandating something weird like this and telling us to open some other proposal just to conform with this weird mandate feels unnecessarily complicated and cluttered.

If you agree that Cathy shouldn't necessarily believe anything in specific about her slot 2 card - and instead should trust context etc - then why on earth do we need this extra Corollary?

[Zamiell]

If you agree that Cathy shouldn't necessarily believe anything in specific about her slot 2 card

i don't agree with that. my position is that by default, in most situations, cathy should write red 5 on her slot 2 card, but not in every situation. the job of the doc is to describe the most common situations that come up, so I have typed up this section.

[jack67889]

this new section does not change how the group treats OOO clues. what it does do is affect situations where players give undefined moves. i am not concerned with undefined moves in this thread.

I see this differently. To me the way the group has been playing is OOO clues/finesses with a lie component in general can be confusing and are not fully defined. So only do them if you are sure they meet the efficiency requirement. In the original example the only way the OOO clue can meet the efficiency requirement is if Bob blind-plays. In my eyes this new definition you are proposing removes the efficiency requirement for OOO play clues and forces Bob to fix instead (thus Bob should assume Alice gives a confusing clue for no reason), doesn’t seem valuable enough to me. (And yes, you are still prescribing behaviour for Bob, since he has to react to the clue.)

[Zamiell]

mandating something weird like this

piano, I would not frame it that way. the way that I would frame it is like this:

piano - "Hey Zamiel, I read your new issue thread and I am very surprised by it. In the past, I have performed one or more Faulty Bluffs, and they have worked great! If your new convention is accepted, then Faulty Bluffs would not work anymore. Thus, this convention seems problematic, so I think it should be modified and/or deleted as to not interfere with Faulty Bluffs."

Zamiel - "Um, piano, the Faulty Bluff doesn't actually exist. Show me the section in the document where it says that a Faulty Bluff legal."

piano - [looks through the document] "Oh, you're right. That's strange, Faulty Bluffs are not in the document. That's weird, as it seems like a simple enough move. As I've said, I've done them before and they work great! And if I've done them, probably other people have done them too. So for the purposes of this thread, we should probably consider them as existing."

Zamiel - "If Faulty Bluffs are a common move, then lets create an issue for them and create a section for them in the doc. When the new section gets added, I'll completely delete the section called The Out-of-Order Corollary, since it will no longer apply. Furthermore, I'll add a bullet point to the OOO section that says: Remember that sometimes, clues that look like OOO are not OOO at all and are instead *Faulty Bluffs*. Don't get them confused!

Zamiel - "However, as it stands right now, that section doesn't exist. So no, for the purposes of this thread, we should not consider them as existing. For the purposes of this thread, The Out-of-Order Corollary follows from the existing conventions as written in a purely logical manner. Meaning that this isn't "my" convention - it just logically follows from what we have already written. Furthermore, for the purposes of this thread, we must not consider what "rogue" players are deciding to do on their own separate from the official moves and the official framework."

[Zamiell]

So only do them if you are sure they meet the efficiency requirement.

no. we commonly do OOO as a 2-for-2, so your post does not apply. if the doc is unclear on this please submit a pr

[jack67889]

Please provide an explanation how this applies to the original example without blocking cards.

[Zamiell]

to clarify, it almost always applies when there are blocking cards. cathy can't tell the difference, so she will assume it is the "almost always" part.

[jack67889]

That's not an answer to my question. So I can't comprehend your statements.

[Zamiell]

what part do you not understand

[jack67889]

Please provide an explanation how this applies to the original example without blocking cards.

[Zamiell]

what original example?

[jack67889]

Level 23. Red 3 on stack. Alice clues Red to Cathy touching Red 5 on chop and Red 4 on slot 1. Bob responds by playing slot 1 as a known bluff. Cathy plays Red 4 from slot 1 since Bob’s blind play let her know that her red cards are in different order than she thought.

[Zamiell]

that example describes undefined behavior, which this thread isn't concerned with. making undefined moves defined should be done in a new thread

in other words, bob is not allowed to play slot 1 as a known bluff, so the example is nonsensical

[jack67889]

I see this differently. To me the way the group has been playing is OOO clues/finesses with a lie component in general can be confusing and are not fully defined. So only do them if you are sure they meet the efficiency requirement. In the original example the only way the OOO clue can meet the efficiency requirement is if Bob blind-plays. In my eyes this new definition you are proposing removes the efficiency requirement for OOO play clues and forces Bob to fix instead (thus Bob should assume Alice gives a confusing clue for no reason), doesn’t seem valuable enough to me. (And yes, you are still prescribing behaviour for Bob, since he has to react to the clue.)

You try to define it as far as I understand. Otherwise I don't understand the need for this section.

[Zamiell]

nope, i am not defining anything new, only clarifying a logical consequence of the existing conventions as written

[jack67889]

That's not true, you just defined what Cathy should write on her cards when Bob blind-plays. (Or shouldn't write on her cards, actually)

[Zamiell]

if Cathy reads the document very closely, and does not take into account undefined moves, then Cathy would write those notes on her cards regardless of whether or not the The Out-of-Order Corollary section exists in the document.

that's why the section is classified as a "corollary" instead of a brand new convention. normally i do not open github issue for things that are purely logical consequences, but stephen requested it

[florian-5f3759df]

in other words, bob is not allowed to play slot 1 as a known bluff, so the example is nonsensical

Why is it a known bluff? Bob is thinking that he's playing the other copy of red 4, it originally looks like a normal finesse I've seen before.

Only after he plays a card that's not red 4, it becomes a bluff (and then perhaps good touch applies to Cathy's slot 1 or not, I guess, depending on the result of this discussion).