3 ms·

Grass shoulder: http://www.wired.com/wp-content/uploads/2015/07/IMG_0724-1024x768.jpg http://www.wired.com/wp-content/uploads/2015/07/IMG_0724-102... I agree t

Grass shoulder: http://www.wired.com/wp-content/uploads/2015/07/IMG_0724-1024x768.jpg http://www.wired.com/wp-content/uploads/2015/07/IMG_0724-102... I agree that it probably presented some level of danger to the public, but I maintain that (i) the added danger was small relative to the normal everyday danger of driving with humans; and (ii) the media exposure they've achieved by doing this on a public road has the potential to pressure Chrysler to remove tens of thousands of hazards (read: vulnerable Jeep Cherokees) from the road, which could ultimately reduce danger and save lives. It's not clear-cut when you're dealing with a vendor who chooses to ignore and/or litigate upon an initial disclosure instead of fix their product.

pyre 11y ago
> the danger was small relative to the normal everyday danger of driving with humans The danger of a car having its transmission crap out on the freeway is non-zero, yes. On the other hand, they explicitly cut a vehicle's transmission on the freeway. The danger for the people in the immediate area of this vehicle was increased because the situation (a cut transmission) went from "maybe it will happen" to "it is definitely happening." At that point, whether or not an accident happened depended entirely on the skill and attention of the drivers around this vehicle something completely out of the control of the researchers.
tlrobinson 11y ago
That photo is next to the parking lot where he "found an empty lot where [he] could safely continue the experiment" and then "they cut the Jeep’s brakes, leaving [him] frantically pumping the pedal as the 2-ton SUV slid uncontrollably into a ditch" Not the highway.
kbenson 11y ago
I think the point is that you, and the researchers, have no right to make that decision for other people. It's illegal to purposefully stop in the middle of the highway. It's illegal to cause someone else to do so as well. People who routinely test things that have the capability for real harm learn to take precautions for as many of the the things you don't think off as you can. For example, the Mythbusters are routinely testing things with cars and other bits of machinery that can cause physical harm if something goes wrong. They also routinely retire to abandoned airforce bases, remote locations, and the salt flats to test things.
- facetube 11y ago
  Point taken. Maybe I'm just being horribly jaded and/or emotionally invested from too much driving, but after a while it gets hard to discern malice from incompetence, and you start (wisely or unwisely) worrying more about actual outcomes than intent. If I'm hit by a security researcher or a drunk driver or a million miler who had one black swan of a bad day, the result is the same to me: I'm hurt or dead. But perhaps a little more objectively, there's certainly a moral hazard here; if every security researcher did this on public roads it'd likely be chaos. An appropriate response, IMO, would be for the police to make a phone call and tell them not to do it again.
  - kbenson 11y ago
    I agree, I don't necessarily want them to go to jail, but I would be happy if they were suitably scared shitless for a while as the enormity of how bad they fucked up (if the facts are as they seem) hits them. Part of the benefit of the extreme reactions from the people here is that future security researchers working on interactions between software and hardware appliances that may pose a physical threat will have an example of exactly why you should show your proof in a controlled environment. It's actually not that different than pure software security research. You don't show a POC for your new DNS exploit by doing it against Comcast or AT&T public DNS servers without expecting some blowback. You set up a test environment.
kbenson 11y ago
FYI, here's a link to a video[1] of the situation (which bengali3 posted up-thread). The grass shoulder was later, cutting the power was actually done on a fairly busy stretch of highway with no shoulder. The reporter's words during that incident are particularly telling. 1: http://dp8hsntg6do36.cloudfront.net/55ad80d461646d4db7000005/9aca04ce-083b-4756-bda9-a37a148bf90flow.webm http://dp8hsntg6do36.cloudfront.net/55ad80d461646d4db7000005...

pyre 11y ago

> the danger was small relative to the normal everyday danger of driving with humans The danger of a car having its transmission crap out on the freeway is non-zero, yes. On the other hand, they explicitly cut a vehicle's transmission on the freeway. The danger for the people in the immediate area of this vehicle was increased because the situation (a cut transmission) went from "maybe it will happen" to "it is definitely happening." At that point, whether or not an accident happened depended entirely on the skill and attention of the drivers around this vehicle something completely out of the control of the researchers.

tlrobinson 11y ago

That photo is next to the parking lot where he "found an empty lot where [he] could safely continue the experiment" and then "they cut the Jeep’s brakes, leaving [him] frantically pumping the pedal as the 2-ton SUV slid uncontrollably into a ditch" Not the highway.

kbenson 11y ago

I think the point is that you, and the researchers, have no right to make that decision for other people. It's illegal to purposefully stop in the middle of the highway. It's illegal to cause someone else to do so as well. People who routinely test things that have the capability for real harm learn to take precautions for as many of the the things you don't think off as you can. For example, the Mythbusters are routinely testing things with cars and other bits of machinery that can cause physical harm if something goes wrong. They also routinely retire to abandoned airforce bases, remote locations, and the salt flats to test things.

facetube 11y ago
Point taken. Maybe I'm just being horribly jaded and/or emotionally invested from too much driving, but after a while it gets hard to discern malice from incompetence, and you start (wisely or unwisely) worrying more about actual outcomes than intent. If I'm hit by a security researcher or a drunk driver or a million miler who had one black swan of a bad day, the result is the same to me: I'm hurt or dead. But perhaps a little more objectively, there's certainly a moral hazard here; if every security researcher did this on public roads it'd likely be chaos. An appropriate response, IMO, would be for the police to make a phone call and tell them not to do it again.
- kbenson 11y ago
  I agree, I don't necessarily want them to go to jail, but I would be happy if they were suitably scared shitless for a while as the enormity of how bad they fucked up (if the facts are as they seem) hits them. Part of the benefit of the extreme reactions from the people here is that future security researchers working on interactions between software and hardware appliances that may pose a physical threat will have an example of exactly why you should show your proof in a controlled environment. It's actually not that different than pure software security research. You don't show a POC for your new DNS exploit by doing it against Comcast or AT&T public DNS servers without expecting some blowback. You set up a test environment.

kbenson 11y ago

FYI, here's a link to a video[1] of the situation (which bengali3 posted up-thread). The grass shoulder was later, cutting the power was actually done on a fairly busy stretch of highway with no shoulder. The reporter's words during that incident are particularly telling. 1: http://dp8hsntg6do36.cloudfront.net/55ad80d461646d4db7000005/9aca04ce-083b-4756-bda9-a37a148bf90flow.webm http://dp8hsntg6do36.cloudfront.net/55ad80d461646d4db7000005...