[bluejak]

 aguahombre, on 2013-January-30, 22:49, said:

Don't even suggest in humor that an agreement requiring the person without the Ace to give true count is illegal and an agreement that the person with the Ace can do whatever he wants (partner doesn't need the count) is illegal encryption.

Of course it is an encrypted signal, normal bridge or not, which is why the EBU Orange book treats it as an exception to its general ban on encrypted signals.

[aguahombre]

 bluejak, on 2013-February-02, 15:33, said:

Of course it is an encrypted signal, normal bridge or not, which is why the EBU Orange book treats it as an exception to its general ban on encrypted signals.

Hence, my careful word choice..it is not "illegal encryption" in any jurisdiction. I would go a step further, though; the person with the Ace is not signalling at all, so it certainly is not an encrypted signal.

[Cascade]

 bluejak, on 2013-February-02, 15:33, said:

Of course it is an encrypted signal, normal bridge or not, which is why the EBU Orange book treats it as an exception to its general ban on encrypted signals.

How is signaling the number of small cards you have in a suit encryption?

There is nothing fundamental to the game that one is only allowed to signal the total number of cards one has or had in a suit.

[bluejak]

Encryption is defined as a different signal based on a key known to the defence but not to declarer. This position is clearly covered by that.

[paulg]

 bluejak, on 2013-February-03, 12:03, said:

Encryption is defined as a different signal based on a key known to the defence but not to declarer. This position is clearly covered by that.

I agree that if you change your signalling as suggested by the OP, then it is encryption.

However if your standard signalling method is always to show the number of small cards in the suit, it is not an encrypted signal since there is no key.

[bluejak]

 paulg, on 2013-February-03, 13:31, said:

I agree that if you change your signalling as suggested by the OP, then it is encryption.

However if your standard signalling method is always to show the number of small cards in the suit, it is not an encrypted signal since there is no key.

If the one without the ace shows count, and the one with the ace does not, that is encrypted.

[gnasher]

Most, or possibly all, "encrypted signals" can be described in a way that makes them seem unencrypted. For example:
- If the defenders know each other's spade length but declarer doesn't, they play that "A high club shows an even number of black cards".
- If the defenders know which of them has ♠10, they play that "A high club shows ♠10 and good diamonds, or denies both."

Similarly, many standard signals are encrypted to some extent, because often you can't interpret them reliably without knowing your own holding in the suit. For example, playing standard attitude a 4 may be encouraging or discouraging, and the "key" is who holds the 2 and 3.

Given that the term "encrypted signals" is inherently ambiguous, we have to rely on the regulatory authority to tell us whether a particular method is permitted. In some jurisdictions that's a straightforward matter, but in the ACBL it's not.

[aguahombre]

 bluejak, on 2013-February-03, 12:03, said:

Encryption is defined as a different signal based on a key known to the defence but not to declarer. This position is clearly covered by that.

Again, encrypted signals are signals. When you are not signalling (message to partner), it is not a signal.

[campboy]

 gnasher, on 2013-February-03, 14:13, said:

Most, or possibly all, "encrypted signals" can be described in a way that makes them seem unencrypted.

Not only is this true, but also some "unencrypted" signals can be described in a way that makes them seem encrypted. "The defender with the ♣2 is playing that a high spade shows an odd number of cards which are either spades or the ♣2; the defender without it is playing the opposite."

[CSGibson]

I see this may be in the wrong forum after all.

[aguahombre]

♥

 CSGibson, on 2013-February-03, 16:22, said:

I see this may be in the wrong forum after all.

the word, "simple", can come back and bite us.

[nige1]

 mycroft, on 2013-January-31, 13:52, said:

In many cases, yes.
Choosing to do the normal, logical thing is one thing, bridge players including your partner can work it out; agreeing to do the same thing is an agreement.
Failing to state that you have *agreed to show standard count with the A* (playing upside-down "normally), as opposed to choosing to falsecard, is an undisclosed agreement. [Edit to add: basically, if partner plays you to hold the wrong number of cards not because he can count it out, but because you have signalled "correctly", then if declarer can't work out that "the one that lied has the A" because that's "normal, logical", it's both an SPU and not GBK.]
Failing to push on this leads to "we signal rarely, but when we do, we tell partner what he needs to know." And I have seen that, and it is absolutely normal and logical (taken to extremes), and is also prima facie not full disclosure, even to opponents of said disclosure player's level.
Yes, it's stupid. So's the regulation stating the one can't have agreements after opponents' infractions, or that one can't have an agreement to open 1NT with a singleton. So, don't talk about it, and don't pay attention to it, and hope "implied SPU" doesn't come up, can't be proven, or won't get pushed.

I agree with Mycroft that sophisticated, stupid, and unnecessary rules cause stupid and unnecessary problems. Some directors interpret such rules in inconsistent and bizarre ways, trying to mitigate these effects. Some players ignore them. Or worse invent their own like Bobby Wolff

[blackshoe]

Rulings on the question whether encrypted signals are legal are simple. Rulings on the question whether a particular signaling agreement is encrypted are not.

[aguahombre]

and then, we still ignore whether the play of a card is a signal or not.

[barmar]

 gnasher, on 2013-February-03, 14:13, said:

Similarly, many standard signals are encrypted to some extent, because often you can't interpret them reliably without knowing your own holding in the suit. For example, playing standard attitude a 4 may be encouraging or discouraging, and the "key" is who holds the 2 and 3.

I don't think this counts as "encrypted". It's just ambiguous, and partner's holding may or may not help him disambiguate. When you play the 4, you don't know whether partner has the card that disambiguates it.

Encrypted signalling refers to selecting the card to signal with, or changing the signalling method, depending on a key known to both defenders (but not declarer).

An example given in "Bridge at the Enigma Club" is opening leads against 3NT, after a Stayman sequence where declarer showed a major. If opening leader has an odd number of cards in that suit, their opening lead is 3rd/5th, otherwise it's 4th best. After dummy comes down, 3rd hand knows his partner's length in the key suit (they acknowledge that the message will be wrong if declarer has 5 cards in his major), and this lets him know whether the lead was 3/5 or 4, but declarer can't tell.

[helene_t]

David, we could play a semantic game that would make standard count signals "encrypted":
- if I hold an honour I play high/low to show an odd/even number of small cards
- if I don't, I play low/high to show an odd/even number of small cards

I am not convinced that the concept "encrypted signal" makes much sense, other than as "a method that sounds confusing when explained to someone familiar with std and/or udca only".

[mycroft]

...and if you have two honours? Is the ten an honour?

re: signal vs "playing a card": if you've agreed that the person with the A will show false count, you're still signalling. If you agree that the person with the A will play the lowest card they have irrespective of count, you still have an agreement, which is disclosable, but not a big deal. If you play "if partner can't care, we play randomly" or "if partner won't care, try to misinform declarer", you still have a disclosable agreement, and that agreement plays better (at least against non-experts) if it remains undisclosed, "it's not a signal, it's just a card" or not.

[blackshoe]

Yes, the ten is an honor.

Quote

Laws, Chapter 1, Definitions: Honor: Any ace, king, queen, jack or 10.

[bluejak]

 helene_t, on 2013-February-04, 03:21, said:

David, we could play a semantic game that would make standard count signals "encrypted":
- if I hold an honour I play high/low to show an odd/even number of small cards
- if I don't, I play low/high to show an odd/even number of small cards

I am not convinced that the concept "encrypted signal" makes much sense, other than as "a method that sounds confusing when explained to someone familiar with std and/or udca only".

I really think this just wrong. Sure, you give a meaningless example, but encrypted signals are clear enough.

if you agree that in any situation where you, the defence, are known to have one honour only, and declarer does not know who holds it, then you will play high-low to show an even number with that honour, the reverse without, then that is clearly encrypted, and is illegal in the EBU, WBU and ACBL, except in the case of the specific exception mentioned earlier in the thread.

As for whether it is a signal, if one of you is known to have the ace, the other not, and the one without the ace shows count, the other not, then the defenders are playing high-low to mean something different based on a key, and that is unambiguously an encrypted signal: of course it is a signal.

What I dislike about some of these posts is that they seem to be based on the concept of how to get around a regulation. If a pair actually do that then they are highly unethical.

Certainly the are some situations, not mentioned in this thread, where it is not clear whether a signal is encrypted or not. That is no reason to suggest that obvious encrypted signals are ambiguous just because it is a tricky concept.

If you and your partner have an agreement in signalling, then that agreement is encrypted if it changes based on a key available to the defence but unavailable to declarer. That agreement is still encrypted even if part of the agreement is that one player's cards do not mean anything.

[pran]

I do come a long way with bluejak here, but I would like his specific comment on an agreement like:

High-low signals show the count of cards in the suit ignoring Aces (or alternatively Aces and Kings).

So {AKxxxx, Kxxxx,} Axxxx and xxxx will all be signalled as containing an even number of (small) cards.

I don't know if a signalling agreement like this would have any merit, but my question here is if such a signal can be considered encrypted?