3 ms·

I feel sorry for the people who are going to have to die for this to happen. I try to explain this to friends who are far too optimistic on self-crashing car t

by kirse 8y ago

I feel sorry for the people who are going to have to die for this to happen. I try to explain this to friends who are far too optimistic on self-crashing car technology. Self-driving cars (SDC) ultimately trade one class of problems that result in death (human-attention deficit) for another class of growing issues (sensor malfunctions/incapabilities, software defects). Ultimately SDC deaths end up as bugs/features on some random devteam's backlog, and I have no desire to have a JIRA ticket named in my honor. In my opinion, by the time all the money and effort is spent making SDC's capable of successfully driving from point-A to point-B in the near infinite possible conditions they could encounter, it would have been 50x cheaper to simply build a fully modernized high-speed rail network over existing highways and roads.

1123581321 8y ago
You are saying that the cost to develop successful self-driving cars is 1-50 quadrillion dollars using an optimistic estimate of rail costs. That does not seem reasonable. Perhaps 200 billion across all companies have been invested in self-driving cars so far (I.e., Waymo is just a fraction of that.)
- kirse 8y ago
  Agree, SWAG was heavily W. Updated. I still think we're 50 years off from near-flawless SDC's though, assuming current LOE.
albntomat0 8y ago
> Ultimately SDC deaths end up as bugs/features on some random devteam's backlog, and I have no desire to have a JIRA ticket named in my honor. Instead you can be the victim in a vehicular manslaughter case due to DUI/texting/etc. As stated elsewhere on this thread, self-driving needs to be measurably better than a human. My state displays how many people have been killed thus far this year on the automatic traffic signs (used for amber alert, traffic info, etc as well). A 20% reduction in the 3000+ folks killed in 2018 would mean a whole lot for those saved.
tzs 8y ago
> Self-driving cars (SDC) ultimately trade one class of problems that result in death (human-attention deficit) for another class of growing issues (sensor malfunctions/incapabilities, software defects). One big difference between those classes is what you can do after an accident. With both, you can investigate why it happened and then make recommended changes to prevent it or reduce its chances in the future. But with driver attention problems, such as drunk driving or driving while texting, it is easy for those recommendations to be ignored. We've been telling people not to drive drunk, not to text while driving, and so on for probably a century in the case of drunk driving, and for as long as texting has existed for texting...and people still do them frequently. With a software defect, it is a lot easier to make sure that the fix actually gets deployed. Make it part of the annual registration renewal for cars that all safety updates have been applied to their self-driving systems.
mr_toad 8y ago
> it would have been 50x cheaper to simply build a fully modernized high-speed rail network Where do you get that number from?

by kirse 8y ago

1123581321 8y ago

You are saying that the cost to develop successful self-driving cars is 1-50 quadrillion dollars using an optimistic estimate of rail costs. That does not seem reasonable. Perhaps 200 billion across all companies have been invested in self-driving cars so far (I.e., Waymo is just a fraction of that.)

kirse 8y ago
Agree, SWAG was heavily W. Updated. I still think we're 50 years off from near-flawless SDC's though, assuming current LOE.

albntomat0 8y ago

> Ultimately SDC deaths end up as bugs/features on some random devteam's backlog, and I have no desire to have a JIRA ticket named in my honor. Instead you can be the victim in a vehicular manslaughter case due to DUI/texting/etc. As stated elsewhere on this thread, self-driving needs to be measurably better than a human. My state displays how many people have been killed thus far this year on the automatic traffic signs (used for amber alert, traffic info, etc as well). A 20% reduction in the 3000+ folks killed in 2018 would mean a whole lot for those saved.

tzs 8y ago

> Self-driving cars (SDC) ultimately trade one class of problems that result in death (human-attention deficit) for another class of growing issues (sensor malfunctions/incapabilities, software defects). One big difference between those classes is what you can do after an accident. With both, you can investigate why it happened and then make recommended changes to prevent it or reduce its chances in the future. But with driver attention problems, such as drunk driving or driving while texting, it is easy for those recommendations to be ignored. We've been telling people not to drive drunk, not to text while driving, and so on for probably a century in the case of drunk driving, and for as long as texting has existed for texting...and people still do them frequently. With a software defect, it is a lot easier to make sure that the fix actually gets deployed. Make it part of the annual registration renewal for cars that all safety updates have been applied to their self-driving systems.

mr_toad 8y ago

> it would have been 50x cheaper to simply build a fully modernized high-speed rail network Where do you get that number from?