[hrothgar]

TimG, on Jun 25 2009, 03:46 PM, said:

awm, on Jun 25 2009, 02:38 AM, said:

Only the last method of cheating (electronic device) does not require another person. Potentially vugraph allows someone to cheat using an iphone, without having another complicit person. If vugraph is delayed, the potential for such cheating is quite limited.

If players must remain at the table during a segment, the potential for this sort of cheating is quite limited. How many people can pull up results on their iPhone while sitting at the table without the other players noticing? Especially if electronic devices are banned from the playing room.

I'll note in passing that one of the competitors during this years Scripps spelling bee was using the spell check functions on his iPhone during the contest.

http://www.huffingtonpost.com/andy-borowit...i_b_209013.html

He made it quite a ways through before getting noticed.

This was just some random kid cheating in a fairly transparent manner. Imagine just what someone could do if they were willing to invest $20-30K on some custom hardware.

Here's how I'd do it.

I'd invest in something extremely simple. A tiny little coil or crystal or what have you that would heat up or vibrate when it heard the right frequency.

I'd place this somewhere VERY sensitive where a small temperature change would get noticed very easily. Under my hair, right on my temple might be reasonable. (I can suggest other locations that are probably even more sensitive plus they're MUCH less likely to get searched)

All that I'd need to do then is sit back and wait for things to heat up.

I could only receive a single bit of information at a time.

Does partner have a minimum or a maximum?
is the queen on my left or on my right?
Is a sacrifice good or too expensive?

Admittedly, my game is bad enough that I wouldn't be competitive even with something like this in "hand". However, I suspect that there are a number of pros who might find something like this worth their while...

Anyone want to remind us how much some of the top pros make for a single weekend of play?

[fred]

Codo, on Jun 25 2009, 01:01 PM, said:

Tim,

I doubt that you can ban people from using the restrooms when they need to go there. This is close to waterboarding.

The USBF has been doing this right for at least as long as I have been playing in the USBC each year (since 2002 I think).

Each table is in its own hotel room. Each hotel room has its own washroom.

I am quite sure that the USBF's original decision to isolate all of the tables in the USBC had nothing to do with vugraph (because this one-table-per-room policy has been in place since before online vugraph was much of a reality).

The USBF's motivation for this policy was likely general security, but in the process they managed to eliminate some of the possibilities for cheating using live vugraph (such as #7 in my original list) while making it easier on themselves to control other possibilities (such as #9 in my original list).

Of course it would not be practical to do this for tournaments in which there are 100s of tables in play at once, but if it is vugraph-related cheating you are concerned about then you only have to do this for tables that are featured on vugraph.

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[TimG]

hrothgar, on Jun 25 2009, 08:02 AM, said:

This was just some random kid cheating in a fairly transparent manner. Imagine just what someone could do if they were willing to invest $20-30K on some custom hardware.

Here's how I'd do it.

I'd invest in something extremely simple. A tiny little coil or crystal or what have you that would heat up or vibrate when it heard the right frequency.

I'd place this somewhere VERY sensitive where a small temperature change would get noticed very easily. Under my hair, right on my temple might be reasonable. (I can suggest other locations that are probably even more sensitive plus they're MUCH less likely to get searched)

All that I'd need to do then is sit back and wait for things to heat up.

I could only receive a single bit of information at a time.

Does partner have a minimum or a maximum?
is the queen on my left or on my right?
Is a sacrifice good or too expensive?

Admittedly, my game is bad enough that I wouldn't be competitive even with something like this in "hand". However, I suspect that there are a number of pros who might find something like this worth their while...

Anyone want to remind us how much some of the top pros make for a single weekend of play?

I think you've gone off the deep end!

You're also talking about a system that needs an accomplice rather than a solo cheat.

[TimG]

fred, on Jun 25 2009, 08:36 AM, said:

Of course it would not be practical to do this for tournaments in which there are 100s of tables in play at once, but if it is vugraph-related cheating you are concerned about then you only have to do this for tables that are featured on vugraph.

But, it would add an expense for the tournament organizers to set up a special vugraph room (or rooms).

Perhaps I am significantly out of touch with reality, but asking a competitor to do without the washroom for about 120 minutes does not seem at all unreasonable to me. Especially if there is a protocol for emergencies.

[hrothgar]

TimG, on Jun 25 2009, 04:46 PM, said:

I think you've gone off the deep end!

You're also talking about a system that needs an accomplice rather than a solo cheat.

Comment 1: This type of system certainly requires an accomplice

Comment 2: I'm not sure where the "Off the Deep End" comment is coming from...

If I wanted a "wire" I'd use precisely this type of method.

This type of system would be almost impossible to detect.
You couldn't sweep for the bug.
The transmitter only needs to burst for a short length of time on a few critical hands.
I suspect that this type of system would be pretty cheap, easy to build, and still provide highly actionable information.

Folks have done MUCH sillier things in pursuit of filthy lucre...

[fred]

TimG, on Jun 25 2009, 01:55 PM, said:

But, it would add an expense for the tournament organizers to set up a special vugraph room (or rooms).

Sure but...

1) As I understand it, when tournaments (at least in the USA) make deals with the host hotel, it is easy for the tournament organizers to get some free rooms thrown in as part of the deal.

2) Hopefully the time will come in which tournament organizers start leveraging the possibilities of promotion through vugraph to attract more corporate sponsors and to get such sponsors to donate more money than they otherwise would.

IMO my point #2 is not just a fantasy and it is also one of the less obvious reasons why vugraph is important and why it is smart for tournament organizers to do what is necessary to commit resources to make sure there is vugraph and that the vugraph attracts a large audience.

Quote

Perhaps I am significantly out of touch with reality, but asking a competitor to do without the washroom for about 120 minutes does not seem at all unreasonable to me.  Especially if there is a protocol for emergencies.

You are out of touch with reality. I can easily go 120 minutes without having to go to the washroom, but in my experience this is far from being the case for many men who are 20+ years older than I am.

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[Codo]

TimG, on Jun 25 2009, 10:55 PM, said:

fred, on Jun 25 2009, 08:36 AM, said:

Of course it would not be practical to do this for tournaments in which there are 100s of tables in play at once, but if it is vugraph-related cheating you are concerned about then you only have to do this for tables that are featured on vugraph.

But, it would add an expense for the tournament organizers to set up a special vugraph room (or rooms).

Perhaps I am significantly out of touch with reality, but asking a competitor to do without the washroom for about 120 minutes does not seem at all unreasonable to me. Especially if there is a protocol for emergencies.

My experience is that your are way out of touch.

I do not like to go into details, but:
A lot of people drink a lot of water/juice.
A lot of people are seniors.
A lot people are nervous because of the pressure or whatelse.

So to have a restroom for every table is a step in the right direction. But as long as you do not have one, everything is possible.

[Echognome]

Why not have the players themselves "donate" the use of their rooms for the top level events? Most of the top level players stay at the hotel onsite and often in pretty decent rooms. You could make it so that players don't play in their own or their partner's or teammates room. Someone could do a quick scan of the bathroom ahead of time. As the tables fold up, they shouldn't be too much of a burden.

All you do is make sure play doesn't start too early and work out with the hotel to have the rooms serviced each day before play starts.

You obviously have more work to do on the organization side, with making sure people know which room to go to, etc. But it should only be for the "vugraph rounds", so it shouldn't involve too many tables.

Also, if trained properly, why not have the vugraph operator be the monitor in general for the room? There could be a set of roving TD's (I imagine two would be enough). If the players have an issue, they could tell the monitor/operator to call the TD's, who are then dialed on their cell phone (or called on a walkie talkie) who then go to the room.

Yet another alternative is to have the rooms be set up as the BBO vugraph rooms. They are comprised of TD and operator rooms and act as compensation for their services. Imagine getting free hotel in exchange for being a vugraph operator/monitor and getting to watch the top quality bridge? I'm sure many people would sign up, even if you slept two to a room and they both had to work some. (I can imagine a couple might do this or two juniors, as examples)

[Hanoi5]

I'm sure some designer could come up with (in case it doesn't exist already) some sort of cabin for the vugraph operator to be in in which he can see/hear everything but he is not seen or heard. It can be the size of a telephone cabin and foldable and would be used in tournaments where it is demanded. The operator would be inside just doing the Vugraph and some troubles might be solved.

[fred]

hrothgar, on Jun 25 2009, 01:02 PM, said:

Anyone want to remind us how much some of the top pros make for a single weekend of play?

I agree that this is a relevant question, but I think you shoot down one of your own arguments by asking it.

Sorry if I misunderstood what you were saying before, but I believe you argued that cheating with the help of a non-partner/teammate is more likely than cheating with a partner/teammate because the potential cost, in terms of getting caught, to your partner in crime is less if that person is not a player. Therefore that person is more likely to be willing to help you cheat.

I agree with that, but the money issue balances this factor at least to some extent:

If your partner in crime is also your partner/teammate, that person has more to gain from helping you cheat than a non-partner/teammate.

Anyways, I personally think it is a distraction to spend a lot of time thinking about exactly how likely various scenarios are. As long as scenarios exist that are sufficiently realistic that they might happen and as long as it is realistic to stop such scenarios from happening, efforts should be made to do just that.

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[fred]

hrothgar, on Jun 25 2009, 01:58 PM, said:

If I wanted a "wire" I'd use precisely this type of method.

This type of system would be almost impossible to detect. 
You couldn't sweep for the bug.
The transmitter only needs to burst for a short length of time on a few critical hands.
I suspect that this type of system would be pretty cheap, easy to build, and still provide highly actionable information.

Folks have done MUCH sillier things in pursuit of filthy lucre...

Richard,

You seem to know at least as much about relevant technology considerations as anyone here (or else you are very good at BSing and the people who really know what they are talking about are remaining silent!).

For now I will take your word for it that scanning for electonic transmitters/receivers is more complex than I would have guessed.

Do you know anything about technologies that block the signals that such devices send and receive?

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[hrothgar]

fred, on Jun 25 2009, 05:54 PM, said:

hrothgar, on Jun 25 2009, 01:58 PM, said:

If I wanted a "wire" I'd use precisely this type of method.

This type of system would be almost impossible to detect. 
You couldn't sweep for the bug.
The transmitter only needs to burst for a short length of time on a few critical hands.
I suspect that this type of system would be pretty cheap, easy to build, and still provide highly actionable information.

Folks have done MUCH sillier things in pursuit of filthy lucre...

Richard,

You seem to know at least as much about relevant technology considerations as anyone here (or else you are very good at BSing and the people who really know what they are talking about are remaining silent!).

For now I will take your word for it that scanning for electonic transmitters/receivers is more complex than I would have guessed.

Do you know anything about technologies that block the signals that such devices send and receive?

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

Hi Fred

I'm not an RF engineer by any stretch of the imagination. However, I have the luxury of working with a rather eclectic group of individuals.

A few monthes back, I picked some of their brains about some of these topics. I am simply reporting what they said...

Few points to keep in mind

1. It's much easier to detect a transmitter than a dirt simple receiver.

By their very nature, receivers can be small and passive. It's certainly possible to detect a passive receiver (You blast our some energy and you listen for a harmonic - The aforementioned Non Linear Junction Detectors do exactly that). However, doing so in a way that you can safely scan people... Thats a nasty problem.

In contrast, receivers need to transmit. This means that they're larger, they're going to have power sources, and they're going to be easier to detect.

2. The small, simple receivers that I described can easily be thwarted

You can use a Faraday cage or some such to shield players
Alternatively, you can pump out enough electromagnetic noise to degrade the signal

I'm not sure how feasible either of these courses of actions might be.

A Faraday cage is large and bulky
Pumping out lots of noise gets you in trouble with the FCC

3. You might be able to design a system that was listening for transmissions. However, if folks are only transmitting tiny little chirps every once and a while I doubt that this would work.

As I mentioned, all of this is based on an hour or so of discussions over beers at the Cambridge Brewing Company. None of us have any skin in the game, but we were able to come up with some fairly simple ways to cheat without all that much difficulty. From my perspective, the important point isn't the specific claims, but rather the fact that people can come up with some very VERY clever ideas.

Personally, I'd be very worried that folks that DO have a strong financial incentive have figured out something MUCH MUCH more clever.

Playing defense against these folks is going to be a royal pain in the butt.

[TimG]

fred, on Jun 25 2009, 09:03 AM, said:

Quote

Perhaps I am significantly out of touch with reality, but asking a competitor to do without the washroom for about 120 minutes does not seem at all unreasonable to me.  Especially if there is a protocol for emergencies.

You are out of touch with reality. I can easily go 120 minutes without having to go to the washroom, but in my experience this is far from being the case for many men who are 20+ years older than I am.

My wife is an attorney. She tells me that in Federal Court here, the trial day is generally 8:30-2:30 with two 15 minute recesses at approximately 10:30 and 12:30. There is some flexibility, but it is generally expected that all parties make a good faith effort to stick to the schedule.

Perhaps senior events should be run in 7 or 8 board segments instead of 15 or 16 board segments. :-)

[mycroft]

If you search me on electronic cheating here, you'll see I'm the champion of "one-bit transmission" as a strategy. Yes, it not only requires vugraph, it requires an accomplice who is good enough to know what bit of information will be useful to a world-class expert (or just determine unambiguously beforehand what to transmit - two-way guesses on my left, maybe. But that's much easier to detect, and much less useful).

But it's simple, it's almost undetectable (because, as Richard says, the passive component is the one in the room) and one bit of information on every hand would make a second-level WCE a champion, if a third- or fourth-level WCE were able to double-dummy analyse in real-time and determine what the "need-to-know piece of information" on this hand was. Very likely even counting transmission errors, translation errors, and occasional deliberate misinformation (to screw up the tracers; the active player would need to know that this would happen sometimes - say 15% of the time, usually in situations where the cost isn't too high - or eir reaction to such a failure may be more detectable than the pattern).

I'm more concerned, frankly, with the work done to properly seed the RNG for the hand generator, and the procedures in place to ensure it is done (and, of course, to ensure that the amount of random information is high enough to encompass all possible board*sets*). Computing is getting cheap enough that probably $250K would buy supercomputer-class hardware that could crunch through the information provided by the first set (or previous sets) of boards and reverse-engineer the seed in "real-time" - say the first 10 boards of the set. I *think* we have that problem down, after several major public mistakes, but the correct (cryptographic science) answer doesn't seem to be happening; we don't see the seed used to create the hand record set on the hand record set. We need a copy of the hand record generator (don't tell me that it's totally secret and totally secure - *someone's* got a copy for sale to the right bidder, so we have to assume that our hypothetical nefarious WCSponsor has a copy) and the seed on the set, each set, for verification that the system is pristine; we need each seed to determine if there's a pattern to the seeds that could reduce the search space from "all possible sets" to "uniquely determinable given a few hands" (likely pointing to an error in implementation rather than an error in system); we need each seed to determine if there is truly enough random information in it to allow each possible boardset to be produced; we need the Bazaar of Eyeballs on the system and the seeds for two reasons: 1) Schneier's Rule (anyone can create a security system *they* can't break into); 2a) 500, 5000, whatever many interested people with some knowledge emulate, for free, a major security expert or two (Linus' "given enough eyeballs, all bugs are shallow"), 2b) the computing power held by that same whatever many people can emulate, for free, a $250k supercomputer (especially if real-time-criticality isn't an issue; see SETI@Home et al.)

Man, that was dense. Sorry. But there is a difference between the BB and the finals of the Canadian Open Trials (and certainly between the BB and the Sheardown Memorial KOs at the Toronto Easter Regional); it may be worth doing the whole gamut (both physical and cryptanalytic) for the BB and the Cavendish, and relax it somewhat for other VGs. The problem is that I don't believe what's being done at the moment, even for the BB and the Cavendish, is at the level where open source cryptanalysis wouldn't find a flaw exploitable by people with the resources that our world class sponsors have (again, I'm not suggesting that any of them would, or do; but if I can buy a $2k computer effectively "because I feel like it", then someone making 500 times what I do could easily afford a $1m cryptanalysis effort "because she feels like it"). And I certainly haven't been shown any proof other than "It works. Trust us." Is it necessary? For me, no. I'm willing to live with that trust, and the honour of the players, and the instincts of the honest players to determine who to pay more attention to. But I'm a professional paranoid, working in computer security, with (most of) a graduate degree in cryptographic engineering, so I see the holes that I'm trained to, and they're bigger than they look to the untrained. They're also smaller than they look to the paranoid, of course.

As I said, I think a consistent 10, 15 minutes, long enough to let the board go through, would be okay. More than that, or doing the compress rebroadcast, would be an issue. I agree that any delay would hinder the self-correction ability of the VG process, though, so a less-error-prone method of VG input would become very desireable very quickly.

[TimG]

I wonder what different precautions are taken at the Cavendish as opposed to the USBF Open Trials. It seems to me that cheating at the Cavendish could be quite lucrative and must be on the minds of the organizers, perhaps even more so that at the USBF Open Trials.

[Phil]

Why do I have the Mission Impossible theme playing in my head as I read Mycroft's post?

[hotShot]

I don't know why so many think that signaling would be done every board.
Do decide a team match between top teams, it would be enough to signal that the next board is a makeable but unbidable game or slam. A player could use a non playing friend without the knowledge of his partner to get that information and just bid a little more aggressive on the next board.

If the aid can see what board his partner is playing and what the result off the next board was in the other room, all what is needed is a little time to signal that information. During a live vugraph there is enough time for that.

[Rain]

Delayed vugraph doesn't prevent VO from sending info to players, only vugraph watchers. Unless you mean VO isn't seated at table, but slightly away, and keys in the cards brought to him by a caddy after every 2 boards.

[barmar]

Hanoi5, on Jun 25 2009, 10:42 AM, said:

I'm sure some designer could come up with (in case it doesn't exist already) some sort of cabin for the vugraph operator to be in in which he can see/hear everything but he is not seen or heard. It can be the size of a telephone cabin and foldable and would be used in tournaments where it is demanded. The operator would be inside just doing the Vugraph and some troubles might be solved.

The operator often needs to communicate with the players, usually to request information he missed: how many tricks were claimed, what was the explanation of the alert, etc.

Maybe we could get players to provide this information more clearly, but I doubt it.

[TimG]

barmar, on Jun 25 2009, 07:15 PM, said:

The operator often needs to communicate with the players, usually to request information he missed: how many tricks were claimed, what was the explanation of the alert, etc.

Maybe we could get players to provide this information more clearly, but I doubt it.

It has always been my impression that the vugraph operators do not bother the players with this sort of thing, that the vugraph operator enters as best he can what he sees, but does not disrupt the game by asking any questions of the players. Same as any other kibitzer.