Help
barmar, on Mar 31 2008, 02:12 PM, said:
Joshs, you seem to be saying that unless you can remove ALL vulnerabilities, there's little point in removing ANY. E.g. putting locks on your doors is a waste of time and money, because burglars can break windows.
But there's no such thing as 100% security, it's always a matter of degree. You want to make things harder on the bad guys, because they're likely to prefer taking the easy way. If there are two cars, and one is locked while the other one isn't, they'll prefer to steal the unlocked one. Even a sticker saying "Premises monitored by XXX Security" probably lessens the likelihood that a house will be broken into, even if it's false (why take the chance that it's true when there are other houses to break into?).
Cheating and accusations of cheating are similar. Every step taken to remove a means of cheating makes it harder for the cheaters, and less likely that they'll attempt it. And accusers usually have to state the means that they think the cheaters were using; if there are fewer ways to cheat, the accusers are less likely to come up with a plausible means.
I'm not saying that I agree with the cellphone ban the way the ACBL is doing it, but the general principle of trying to prevent means of cheating seems admirable.
But there's no such thing as 100% security, it's always a matter of degree. You want to make things harder on the bad guys, because they're likely to prefer taking the easy way. If there are two cars, and one is locked while the other one isn't, they'll prefer to steal the unlocked one. Even a sticker saying "Premises monitored by XXX Security" probably lessens the likelihood that a house will be broken into, even if it's false (why take the chance that it's true when there are other houses to break into?).
Cheating and accusations of cheating are similar. Every step taken to remove a means of cheating makes it harder for the cheaters, and less likely that they'll attempt it. And accusers usually have to state the means that they think the cheaters were using; if there are fewer ways to cheat, the accusers are less likely to come up with a plausible means.
I'm not saying that I agree with the cellphone ban the way the ACBL is doing it, but the general principle of trying to prevent means of cheating seems admirable.
I guess I need to repeat myself.
If I have two gates to a complex, and put an alarm at one but not the other it doesn't help at all. If I put alarms at both, but the fence is 2 feet high, and has no alarm, it still doesn't help. By, it doesn't help, I do not mean that it only improves security by "only" 1%. I mean it improves security by 0.
I also brought up the issue of psychological deterrence. I do beleive that there are some things which do not help security at all, but reduce the likelihood of an attack. For instance, I can put up a sign on my house that says "protected by XXX alarm company" and not have an alarm. That clearly will reduce the number of breakins (at least initially) by non-professional thiefs. Its unclear if a specialist will be effected. Furthermore, does it decrease breakins? No, you merely break into a house without the sign (just as you would cheat a different way). How about if every house had such a sign? Probably that will have some detterence effect, if somehow every house had a sign, but none of the crooks knew that the houses did not have any real security (which really is not beleivable)...In any case, it really is much more cost effective to have security measures that actually improve security, rather than spending resourses on only those that give the appearence (to an untrained eye) of improved security but don't help at all.
Effective security methods are things that either
a. eliminates the n most effective ways to cheat such that the n+1'th method is measurably less effective than any of the top n
b. improves your security against all forms of cheating (like paying soemone to walk around and look for cheating taking place)
To make this concreate suppose there were 5 ways to cheat:
A will be successful 90% of the time
B will be successful 90% of the time
C will be successful 89% of the time
D will be successful 70% of the time
E will be successful 20% of the time
Eliminating A gains you nothing (I can do B instead)
Eliminatating B gains you nothing (I can do A instead)
(Same for C, D and E alone)
Eliminating A and B gains you 1% which may or may not be worthwhile, depending on the costs of eliminating A and B, and the costs of successful cheating. Its a pretty exceptional case when this kind of improvement is really worth it.
Eliminating A, B and C gains you 20%, which is probably worth it, but again you been to look at costs and benefits. I certainly wouldn't do this if it cost every ACBL member $5000/year.
And so on.
Most "improvements to security" do not decrease the probability that someone will be able to cheat and get away with it (and thus are a complete waste of money). The goal is not a 100% reduction. Its a credible ability to catch more cheaters (and a cost benefit analysis is required to determine if decreasing cheating by say 1% is worth the assocaited cost). Merely forcing people to cheat in a different way does not in and of itself improve your probability of catching cheaters.
Having said that, I repeat that there are trivial ways to cheat right now that do not involve any technology. If you can't prevent those, eliminating things like cellphone use will clearly have no benefit...
In your analagy, we had a sign that said "we can detect all cheating with cellphones". So lets grant that it will reduce the amount of cheating with cell phones even if the detection claims are phony (which is a claim that is not necessarily true). Reducing one form of cheating does not reduce cheating unless that form is easier an/or more effective than others, which in this case its neither.
