Slashdot Mirror

no, it's really not

http://www.meretrix.com/~harry...

from 2004, but essentially should be still good,

looking at per mile driven vs flown, or looking at per hour driven vs flown. general aviation is still a couple times more fatal than driving.

GA _is_ far more dangerous than road travel. According to this analysis, it is nearly 9 times so on a per-mile basis:
http://www.meretrix.com/~harry...

You can't compare nominal numbers like that because there are many, many more vehicle miles traveled by auto than by airplane. Also, the air travel numbers include commercial planes, which comprise the vast majority of air miles traveled, and have different standards than General Aviation, which is what TFA is talking about.

General Aviation has 9x more fatalities annually than cars per vehicle mile traveled (link). I'll take my chances on the ground, thanks.

General Aviation results in 9x more fatalities per mile traveled than driving (link). I'll take my chances on the ground, thanks.

The stats for non-commercial flying do not support your assertion that flying is somehow safer than driving. GA Flying vs driving uses 2004 numbers, but you can see the updated data with similar numbers over at AOPA

Aircraft injuries/fatalities are only better than cars because the vast majority of traffic is on commercial airlines that are rigorously maintained, with pilots who must pass (relatively) stringent qualifications.

Little private planes are much more dangerous, mile for mile, than cars

Even if the bugs were worked out so that a car "experienced a bug" only once every 100,000 miles, given the number of vehicles presently on the road and how much they are driven every day, that would still be too many "crashes" for society to find acceptable.

Well, the Google cars have logged over 140,000 miles as of last fall without an accident*, though that is with "occasional human intervention." Looking up human driver statistics, I found that there is about 2 accidents per million miles. This statistic probably doesn't include minor fender benders that aren't reported to police/insurance. I think the Google robot car probably has a little more proving of itself to do, but I think if they can get to 140k, getting to 500k won't be too difficult.

The most impressive thing about this, and the thing that makes this idea viable in the real world, is that the car doesn't need any infrastructure to support it. Most robot car ideas I've seen that are intended for public use (e.g. to ease freeway congestion, etc.) involve placing sensors or emitters in the roadway or in other cars for the robot car to use as guides. The Google car can, potentially, be deployed regardless of other users adoption or government investment.

*OK, there was one accident where they were rear-ended while stopped at at traffic light. Oh, the irony.

I don't know that your second statement is true (re: on way to airport).

I could be wrong, but I have read that over-all driving vs flying is of equal safety per person hour (making flying significantly safer than driving per person-mile due to fewer injuries, and faster speeds). I suppose airports could be in a more dangerous driving area of the road system skewing things, but in general i think you're worse off over the course of a flight than the drive there.

Also, this is the first page that came up looking for it, and I find equal fatalities for person-distance, thought 1/3 as likely to happen, but unless you're driving a total of 1/3 of your flights distance you would be safer on the drive.

http://www.meretrix.com/~harry/flying/notes/safetyvsdriving.html

False. These stats cherry pick the type of airplane/pilot (large, commercial) but do not allow you to cherry pick the type of auto/driver (safe, sober). This is a good link. Another good link.

There are a lot of ways to skew the statistics. For example, pedestrian deaths are included in cars as opposed to the "shoe vehicle" category. However, ground victims of 9/11 are not considered. So, hit by a plain and your death doesn't count against airplane safety, but hit by a car and your death does count against auto safety. Weird.