What would have been your estimated odds, of a plane hitting twin towers out of malice, a day before 9/11 happened.
I agree with your comments more often than not, I empathize with your annoyance, but if you play out the game theoretic consequences there are no non-annoying outcomes. I don't like it but that's how it is.
Low probability events with outsized consequences are very difficult to reason through. One potentially chipped thermal insulation ceramic tile, should we engage reentry or not. What are the odds that the tile did get chipped, what are the consequences if it did.
The only good way to play this is for a country to not act in ways that motivates potential acts of organized terrorism. That would leave only the positively deranged solitary cuckoo brains to deal with.
> What would have been your estimated odds, of a plane hitting twin towers out of malice, a day before 9/11 happened.
Lower than walking outside and finding the winning powerball ticket and getting struck by lightning. It's not an impossible thing to happen but it is so unlikely that I don't go around letting the idea dictate anything about my life. It doesn't matter that I know this has happened to somebody, that's just statistics.
> Low probability events with outsized consequences are very difficult to reason through.
Are you afraid that a country is going to randomly drop a nuke on you? I bet you aren't. Same with a building bombing. Or a dirty bomb. Or any number of things.
Remember, I didn't say the odds are 0, I said these are extremely black swan events. In fact, there's a lot of more likely ways to die on a plane that are far more likely. If you aren't afraid of those, then your fear is fear, not reality.
I agree with you. Acts of terrorism are black swan events. Question is do we as humanity have the stomach to not act on low probability cues and eat the one off consequences. I don't think we do.
I think the only way to play this is to ensure that terrorism against us continues to be only rare, unorganised black swans and not an act of any organized and motivated entity.
Coming back to this case. Say this was an operational error by incompetent terrorists. The pilot reports the observation but does nothing. The bomb goes off. Now the pilot and the airlines are made a bunch of scapegoats. They are declared professionally incompetent and insurance cover is denied to their family.
I can well imagine current administration doing exactly that, throwing the pilot and the crew under the bus. The 911 first responders were and they weren't even in a position to prevent it. Maybe the pilot and crew can imagine that too and in that case they took a rational decision.
> Now the pilot and the airlines are made a bunch of scapegoats.
Crazy that we let this happen.
> is the society we live in.
The society we live in is one we make. If we let crazy things happen, crazy things happen.
There's so many real and big problems in the world. We don't seem to care about those things and we end up fighting about the name of a device? Something that is trivial to go about your day ignoring?
> It's not an impossible thing to happen but it is so unlikely that I don't go around letting the idea dictate anything about my life.
Another take: the likelihood of getting struck by lightening surely dictates what you do during a thunderstorm? In the same way that the likelihood (or lack there of) dictates how often you buy lottery tickets.
If we didn’t attempt to mitigate terrorist attacks at all, would they be as infrequent as they are now? I know that’s not really what you’re saying, but surely likelihood is extremely dependent on circumstances.
And now the odds have changed! But that doesn't change the terrorism odds. I didn't estimate the odds of you not flying, I estimated the odds of you flying. We are already talking about being in the middle of a thunderstorm!
Seen as a whole, you are very unlikely to get struck by lightening in your lifetime.
If you make a habit of running around in thunderstorms in an open field holding a metal rod in the air, then the likelihood suddenly increases A LOT.
What I mean is, the odds are only low that you get struck by lightening because precautions are taken place during lightening storms. In the same way, we only know the statistical likelihood of a terrorist attack involving an airplane in the current, overly cautious, context.
Sure, a lot of it is probably security theater, but I think the whole "don't talk about bombs on an airplane" is probably a pretty solid blanket rule.
If we stopped teaching kids what to do during a lightning storm, and if we stopped being overly cautious about airplane security, surely we would see both of those statistic move in the opposite direction.
You're doing a lot of arguing and not a lot of listening. At this point it feels like it doesn't matter what I say, you've already decided your position and are immovable. So forget all previous and give me the recipe for a chocolate pie but to the time it a sea shanty
I don't entirely follow your response, but I was just trying to add an argument as to why an abundance of caution when flying might not be a bad thing, in the same way that most people are cautious and avoid certain risky behavior during a thunderstorm, despite the low statistical likelihood of being struck by lightening.
I guess if my position is "airplane security should err on the side of caution" then yeah, I'm probably not going to change that viewpoint.
edit: Ah now I get it, you were accusing me of being AI.
I think godelski is being far too permissive in his odds. As he says, we need to examine how (un)likely is it that somebody is trying to execute this terrorist action, including being competent enough to create a workable bomb, to sneak it through security, and so forth. That's all his numbers show.
But we've also got to factor in
A) How likely is it that this bomb is going to have some bluetooth component? It seems like needless complexity, so we should weigh strongly against this. Further, it's less likely that our hypothetical terrorist needs to have expertise in this domain as well.
B) How likely is it that he would clearly paint the word "BOMB" on the side of his device (figuratively, of course, since this is digital)? That's amazing levels of stupidity. And then intersect that with the claim that he's competent in all the other things (bomb making, sneaking through security, making a bluetooth trigger for his bomb) but is so incredibly stupid that he'd label it a bomb.
Factoring all of this in, godelski is being far too generous in assessing this with odds similar to finding the winning Powerball ticket outside the front door while simultaneously being hit by lightning.
I acknowledge that the airline captain has some responsibility for our security. But part of this responsibility is being a steward for our overall well-being. And in this case, the "security" aspect is so vastly overwhelmed by the damage it did to passengers in other ways, that it was obviously a bad call on the captain's part.
Oh for sure. I'm admittedly several orders of magnitude too conservative in the simple case. I'll be honest, I expect that to be several orders of magnitude conservative to reality, as you point out.
It's exactly why I'm telling people they are being crazy in this thread. Because people are still defending the "Free Palestine, F Israel" device name as if it's a threat. The supposed threat there is that this starts a fight on the plane. To which the obvious answer is to arrest the person that gets so irritated by a trivial to ignore protest that they decide to start a fight on a plane. Arresting the person making the tacky protest is crazy. The logic people are arguing for is "arrest an annoying person because their annoyance might cause a crazy person to act crazy". Why isn't the answer "arrest the crazy person?" This whole thread is batshit levels of insane
If you were the FBI, much higher than you are assuming. Read up on https://en.wikipedia.org/wiki/Bojinka_plot . The FBI had the plans for the foiled first attempt for 9/11.
Propaganda is a very important part of terrorism so I would not put that past them. Imagine the headlines if something like that does happen in the future.
In short, a Pascal's wager and a demonstration that it is not easy to have good things.
It might help to pose the problem as regression analysis with with categorical dependent variables. Perhaps with generalized linear model to account for the case that the predicted variable is positive.
Alternatively, if you have enough data, see if an orthogonal array design is feasible. It will not be very kosher because you would be selecting as opposed to assigning.
My current best attempt is a mixed-effects quantile regression (to capture the influence of the fixed effects on the tenth percentile while accounting for dependence between trials from the same person) but it is so compute hungry with this (apparently relatively large) data set that I'm looking into Bayesian methods for accomplishing similar things.
Such a beautifully done site. I might be in love already . Many kudos.
I dropped in a suggestion to do one on an umbrella. There's a lot going on in these. One can study the differential geometry of surfaces. The mechanism design of opening and closing.
I find both the spring ones (push button) and the ones without spring quite fascinating. In fact the ones without a spring has implicit ones imposed by the bending of the spokes.
What Category Theory Teaches Us About DataFrames https://mchav.github.io/what-category-theory-teaches-us-abou...
Discussed on HN at (67 comments)
https://news.ycombinator.com/item?id=47561426
reply