Hacker News Reader

Comment by airza | Hacker News Reader

This extensively documented site (http://www.killedbypolice.net/kbp2013.html) suggests that 768 people were killed by police in 2013, compared to the 784 people murdered by trains. If my math is correct, 784/768 is not five. Additionally, that site's statistics are only for may-december, so the actual number is probably around 50% higher.

View on HN

The authorities can already track the location and speed of all people with smartphones.

The location and speed of all trains is also known.

The solution to prevent crashes is thus obvious. Hit the car before the crash with a drone-fired missile.

"The location and speed of all trains is also known."

Just need an app that alerts you to oncoming trains.

Or pay attention to the gates and lights?

You got one inbuilt. It's logic goes something like

    if myself.is_near(train_tracks) and (myself.can_see(train) or myself.can_hear(train)):
        myself.risk_of_oncoming_train = True

risk_of_oncoming_train is a boolean? Sounds more like something that would be representing a chance, so I would be expecting something like a floating point number.

I was thinking that being hit by a train is a situation that I want a 0 probability of. So personally if the probability is non-zero I would immediately act to reduce it to zero. Hence a boolean does the job for me.

Apparently this doesn't work as well as you'd think. I just read a depressing article about the number of people that pose for photos on train tracks, when celebrating big life changes like marriage or graduation, and then get hit by trains.

https://www.washingtonpost.com/local/public-safety/taking-ph...

An app would be a further distraction and would perhaps be detriment to what you want to achieve.

yeah, an app for the car.

It seemed obvious to me that your comment was a joke. Please continue this style of humor despite the other comments.

"in 2013": It might be an outlier. We had an abnormally large group of terrorism-related deaths in 2001, but on average it's less, and the same type of thing may be in effect here. That year may have been bad for deaths by police, and better for train safety, than average.

over 10,000 died in alcohol-related crashes in 2013

http://www-nrd.nhtsa.dot.gov/Pubs/812102.pdf

I was, of course, using the "155" number that the OP used in his analysis, so we are both correct.

Reply on news.ycombinator.com