4 ms·

Same BS as usual, failing to control for where, when, and how crashes and incidents occur will alway make self driving look safer: they disable themselves in pr

by olliej 3y ago

Same BS as usual, failing to control for where, when, and how crashes and incidents occur will alway make self driving look safer: they disable themselves in pretty much every situation where crash rates increase: night, rain, types of road, etc. A legitimate comparison would only compare self driving rates to human driven accidents in the same conditions. That these companies continue to repeat this kind of BS makes me think that the benefits, if any, are much smaller than anything they’re claiming.

Reubend 3y ago
Seems like you wrote all of this out without realizing that they drive 24/7, regardless of weather, and they drive in all roads except highways (granted that's a big difference!) in the cities they're in. Which is to say that most of what you just said is really misleading. The only part that seems relevant to me is the highway driving versus city driving.
- olliej 3y ago
  Waymo's description. Sentence one. [1] "Waymo’s system is currently designed so each vehicle operates only within pre-mapped zones under certain conditions." They don't drive all roads, they require roads to be fully mapped down to a full 3d map of the surface, and if the realtime surface deviates from the mapped surface they immediately reroute to a different road. [1] https://ltad.com/about/waymo-zones.html https://ltad.com/about/waymo-zones.html
huytersd 3y ago
Almost nothing in your comment is true. Moreover you didn’t bother reading the article let alone the research data.
- olliej 3y ago
  I did read the article, so I can see that nothing I said was wrong, given the content of the article. Given one of the first things I said was that they don't drive in all conditions, and the first thing on Waymo's "lets hype self driving" site is that they don't operate in all conditions, I'm fairly sure I'm right here. Unless you can point to the part of the article where it talked about how they controlled for driving conditions and locations of accidents, rather than the part of the article complaining about how their data lumped together unrelated regions.

by olliej 3y ago

Reubend 3y ago

Seems like you wrote all of this out without realizing that they drive 24/7, regardless of weather, and they drive in all roads except highways (granted that's a big difference!) in the cities they're in. Which is to say that most of what you just said is really misleading. The only part that seems relevant to me is the highway driving versus city driving.

olliej 3y ago
Waymo's description. Sentence one. [1] "Waymo’s system is currently designed so each vehicle operates only within pre-mapped zones under certain conditions." They don't drive all roads, they require roads to be fully mapped down to a full 3d map of the surface, and if the realtime surface deviates from the mapped surface they immediately reroute to a different road. [1] https://ltad.com/about/waymo-zones.html https://ltad.com/about/waymo-zones.html

huytersd 3y ago

Almost nothing in your comment is true. Moreover you didn’t bother reading the article let alone the research data.

olliej 3y ago
I did read the article, so I can see that nothing I said was wrong, given the content of the article. Given one of the first things I said was that they don't drive in all conditions, and the first thing on Waymo's "lets hype self driving" site is that they don't operate in all conditions, I'm fairly sure I'm right here. Unless you can point to the part of the article where it talked about how they controlled for driving conditions and locations of accidents, rather than the part of the article complaining about how their data lumped together unrelated regions.