Join Bridge Winners
All comments by Richard Willey
You are ignoring the author of this comment. Click to temporarily show the comment.
I think that our standards regarding what it means to be “secure” are very different.

I would NEVER trust a system in which a single rogue individual could compromise the security of the system.


Dec. 9, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
They are. However, the generate, transmission, and storage aren't done in a particularly secure manner.

There have been cases where the ACBL has sent the same hands to different tournaments on different days. The existing system requires that we put trust in individuals rather than processes.
Dec. 9, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
> However, with correct software design, it isn't likely.

Well, the birthday paradox is in play… (I am actually sitting in a meeting where we are analyzing the implications of hash table collisions)
Dec. 9, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
> Discussion is related to dealing software and is only
> slightly relevant to the thread. Don't worry about it.

Huh. I would have felt that secure methods for hand generation would be fundamental to any proposal to move to an electronic playing environment.

Just last week, we saw a ridiculous screed that, in part, involved the security of the hands for the 2013 USBF team trials. I think that said thread does a good job showing why

1. The hand generation system must be secure
2. Equally important, it must be possible to demonstrate that the hand generation system is secure
Dec. 9, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
> To the best of my knowledge the tournaments where you go
> and play bridge are all played in the same ‘old school’ way:
> Four people sitting at one table each holding 13 cards.

I have never seen such an blindingly obvious example of “No true Scotsman”
Dec. 8, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
FWIW, I think that the mechanism for generating the seed and whether to provide a salt is much more interesting than the minimum seed size.

As I have noted before, I think that it is strongly desirable to have a mechanism in which

1. The seed/salt is generated in an open and secure fashion at the start of each round of play.

2. The seed/salt is keep secret until the close of the round of play

3. The source code for the executable is available so people can that the hands played are consistent with the seed / salt

(For aesthetic reasons, I think that there is some value in generating the seeds and the salt using physical desks of cards)
Dec. 8, 2015
Richard Willey edited this comment Dec. 8, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
> The number one question you have to answer is what happens
> when the system crashes, how do we carry on from where we
> are at. If the answer is “we can't” then this system is
> dead in the water.

All of the brains for this system are running on a server. Hot swapping a tablet should be pretty trivial.

If your local server crashes, life gets a bit more complicated. However, even here you should be able to recover pretty easily. Recall the system that I propose where your hands for a session are being generated from a known seed. This will allow you to recover all of your hands even if the server needs to get rebooted.

I would also expect that this system to feature a local server which is echoing information to the central BBO servers. The primary reason for this is to avoid overloading the local server with spectators, however, it also provides remote storage and recovery.

I (personally) would be far more worried about a network outage than a server crash.
Dec. 7, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
> Since beginning of (bridge)time there have been several
> changes, but one important thing has always stayed
> the same: Four people sitting at one table each
> holding 13 cards.

It is hard to imagine a statement that is so wrong.

I just logged into BBO. As I speak, there are 1900 tables playing bridge. If I log in again in a couple hours, I'll also see a couple thousand tables playing bridge. I'd bet dollars to donuts that the total number of hands of bridge played in a given week on BBO is larger than what you see in any national bridge organization. (Oh yeah, its worth noting that many of those players are earning master points)

Dec. 7, 2015
Richard Willey edited this comment Dec. 8, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
I agree that the existing recorder system appears to be badly broken. With this said and done, I think that any system based on reporting/recording “suspicious” hands is highly problematic.

Part of the reason that I have long supported the development of a system like Fred's BBOH is that this provides us with a mechanism to automatically record ALL hands.
Dec. 7, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
> What surprises me is that the statisticians seem to rule
> out that top players/top analysts can sense/prove cheating
> based on bridge logic and by looking carefully into the
> hands. Just because you don't get it yourself doesn't mean
> it isn't a fact.

Hi Boye

I never claimed that top players / analysts are able to sense cheating by looking carefully at hands. Where we differ is whether these same techniques are sufficient to prove cheating in any meaningful sense of the world.

I have heard that various top players were convinced for years that F+N, F+S, and the Doctors were cheating. Nothing was able to move forward on this front until the fortuitous introduction of video feeds provided a tool that careful observers could use to match various out of band signalling techniques with hand holding and defensive decisions.

Ultimately, the fate of these pairs will be determined by careful analysis of hands and presentation of facts. In turn, this means that do care needs to be exercised to ensure that the data was analyzed fairly and that the prosecution is not cherry picking hands.
Dec. 7, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Your “top players/analysts” were cheated blind for 15 years. They may have had a sense that something was wrong, but they certainly couldn't prove anything. In the process, almost every major event has been tainted or discredited.

The only reason that things moved at all was a combination of

1. The novelty of video feeds
2. The folks who were cheating were really stupid about how they were communicating information

Now that the cheaters are aware of the strengths and weakness of video technology, you aren't going to be able to catch folks any more (and I doubt that the reason is that the game is suddenly completely clean)

Personally, I'll put my faith in the statisticians. However, I also know that these techniques need large and clean data sets to operate on.



Dec. 6, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
> Create a database where every bridge player in the world
> can send in peculiar hands (must be easy to contribute with
> hands). The contributor has to identify himself/herself so
> it is possible to verify the facts (and avoid rumours and
> false accusations). I believe this database should be open
> to the public, as should all video tapes. Transparancy is a
> key.

Speaking as one of the stats types on this list, I need to emphasize that any system that only records “suspicious” hands is much less useful than a system that records all hands.

> If the thought is to use this mechanism for top level
> bridge, why don't we ask the top players? If they prefer to
> continue playing a card game (even knowing that it gives
> their opponents a better chance to cheat), don't you think > that speaks volumes about how great a card game bridge is?

The conclusion that I would draw is that people - particularly old people - don't like change. At the same time, I also believe that folks acclimate pretty quickly once they are dragged (kicking a screaming) out of their comfort zone.

One point that is worth noting: I think that we've reached the tipping point at which more hands of bridge are being played online than in F2F tournaments. You might think of bridge as a card game, however, I don't think that demographics agree with you.
Dec. 6, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Tom Peters might be on to something when he mentioned cardboard.

Triple wall card board is extremely strong, heavy enough that you don't need to worry about it getting knocked over, and much much cheaper than wood. (A decent number of folks are actually using cardboard for low cost furniture these days).

In a perfect world, you might be able to drop the costs low enough that disposable cardboard screens might be a viable substitute for shipping.

Don't discount the possibility of constructing an integrated screen / table out of cardboard rather than renting those. If designed properly, you'd also be able to block toe tappers.
Dec. 6, 2015
Richard Willey edited this comment Dec. 6, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Sadly, I don't think that players can be trusted to bring their own hardware. Its too easy to add some kind of app to facilitate cheating.

I do believe that the app itself (maybe even a tablet) should be provided to participants well in advance of the event so players have the option to practice with the new technology.
Dec. 6, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
David / Sabine,

You might want to check out a new game called Crowfall. (Between this and Warhammer: Total War, 2016 is looking really exciting)
Dec. 5, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
> why make me share in this cost?
> for 90% of the ACBL they don't care why make them pay these > costs?

The ACBL claims that the nonprofit's mission is to promote the game of bridge.

This system represents an enormous improvement in broadcasting and recording games. WRT Vugraph you automatically get the ability to live stream every single table being played. You also get a perfect record of every bid that is made and every card that is played. I can not help but believe that this will be good for the game. (Imagine what Eric Kokish or some other author could do to create a world championship book).

While this system doesn't go as far as I would like to prevent electronic cheating, it make it much more difficult to use more traditional methods. In addition, if we can force players to use electronic mechanisms to cheat, it does make the trial phase relatively simple. If a player is caught with a transmitter in his shoe, its going to be very hard to explaining this one away.

(BTW, I still strongly believe that you're going to want to delay hands by five minutes or so)

Dec. 5, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
I think Karl hit it on the head.

Let's assume that you're dealer. You get to see when LHO bids, but that's no different than with a normal screen.

After that, you get to see when RHO bid (same information you get with a screen). However, unless there is an audible signal, you can't tell how much time was spent by partner and how much by RHO.

Dec. 4, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
> Finally, and not very seriously since I suspect technical
> measures could be devised, it seems to me that any sort of
> network communication, essential to the method, might be
> susceptible to being hacked.

As shocking as it might seem, there are well established protocols to encrypt tcp connections.
Dec. 4, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
I have no first hand knowledge of the architecture that Fred is using. However, I am presuming that

1. The tablets are all running client processes.
2. These clients are all communicating with a local server that is running the tournament.
3. That same server is relaying information back to BBO's main system which is responsible for handling Vugraph.

This type of system would mean that the tournament can continue to function even if the Internet feed goes pear shaped. At the same time, you don't need to worry about thousands of spectators clogging the Internet feed to whatever hotel you are playing at.
Dec. 4, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Why would anyone ever do that?

1. It is much less secure
2. The security of the system relies you trusting individuals rather than processes
3. It is un-necessary



Dec. 4, 2015
Richard Willey edited this comment Dec. 4, 2015
.

Bottom Home Top