Latest Tesla self-driving software version doesn't stop for pedestrian in crossing
#76
Senior Member
Join Date: Jun 2008
Location: Louisville, KY
Posts: 13,446
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 4014 Post(s)
Liked 2,770 Times
in
1,673 Posts
This is the same argument as helmet people have been having for years. Supposedly, cars give cyclists who aren't wearing helmets more space, and helmet wearers engage in more risky riding behaviour because they trust in the helmet as a force field of protection. The idea that having extra safety aids in your car makes you more confident to do things you wouldn't otherwise do (and spend less attention on your driving because "the computer is there to do it") is exactly the same argument that's been going on in A&S (and has kept me out of there) for the last 15 years or more.
__________________
Bikes: 1996 Eddy Merckx Titanium EX, 1989/90 Colnago Super(issimo?) Piu(?),1990 Concorde Aquila(hit by car while riding), others in build queue "when I get the time"
Bikes: 1996 Eddy Merckx Titanium EX, 1989/90 Colnago Super(issimo?) Piu(?),
#77
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
You are mixing two completely separate things in the bolded type. 1) People are lazy. Once automatic transmissions became good enough to feel like a quality drive, wasn't hard at all for manufacturers to push it and (almost) everybody to buy into it. That is all about bottom line and marketing. Not needing to make two versions of the same car. 2) Safety - every study I've ever seen shows manual transmission drivers being more alert and better drivers.
Glad to take the correction.
Seriously, any cites to those studies? I'm not doubting you, I'm just curious, so if it's something easy at hand, I'd be grateful to have it.
Just a guess--is it because you can just accelerate without manually shifting
#79
Senior Member
Join Date: May 2021
Posts: 6,321
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 3214 Post(s)
Liked 3,519 Times
in
2,220 Posts
Likes For PeteHski:
#80
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
Certainly, the evidence for it is incredibly weak. It's all based on one dataset gathered by a single rider in one city riding with and without a helmet. That data has been reanalyzed twice with different interpretations : https://psyarxiv.com/nxw2k
I think the obvious methodological problem of having the researcher being his own subject is so obvious that I don't really see how any findings based on this could be useful. Obviously, if he knows what hypothesis he's testing, it could affect how he's riding.
#81
Senior Member
Join Date: May 2021
Posts: 6,321
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 3214 Post(s)
Liked 3,519 Times
in
2,220 Posts
Certainly, the evidence for it is incredibly weak. It's all based on one dataset gathered by a single rider in one city riding with and without a helmet. That data has been reanalyzed twice with different interpretations : https://psyarxiv.com/nxw2k
I think the obvious methodological problem of having the researcher being his own subject is so obvious that I don't really see how any findings based on this could be useful. Obviously, if he knows what hypothesis he's testing, it could affect how he's riding.
I think the obvious methodological problem of having the researcher being his own subject is so obvious that I don't really see how any findings based on this could be useful. Obviously, if he knows what hypothesis he's testing, it could affect how he's riding.
#82
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
According to the paper I linked to, he was actually primarily testing the effects of different lane positions, and just threw in this variable out of curiosity (don't think I've ever seen that admission in a scholarly study!). The reanalysis showed that any supposed helmet effect disappeared in the 3 foot from curb position. The statistical argument is over whether a small difference in average passng distance matters when there's evidence that the average passing distance is over 3 feet or right around that much.
What no one seems to be addressing is that although this study involves thousands of passes, it's really a case of n=1 because they're all the same rider riding in the same general area. I don't believe that riders are uniform in any number of possibly relevant factors, nor do I believe that drivers behave the same in different contexts. I have no idea how the habits of drivers in Bath relate to those of the rest of the UK let alone to the US.
#83
Senior Member
Join Date: May 2021
Posts: 6,321
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 3214 Post(s)
Liked 3,519 Times
in
2,220 Posts
According to the paper I linked to, he was actually primarily testing the effects of different lane positions, and just threw in this variable out of curiosity (don't think I've ever seen that admission in a scholarly study!). The reanalysis showed that any supposed helmet effect disappeared in the 3 foot from curb position. The statistical argument is over whether a small difference in average passng distance matters when there's evidence that the average passing distance is over 3 feet or right around that much.
What no one seems to be addressing is that although this study involves thousands of passes, it's really a case of n=1 because they're all the same rider riding in the same general area. I don't believe that riders are uniform in any number of possibly relevant factors, nor do I believe that drivers behave the same in different contexts. I have no idea how the habits of drivers in Bath relate to those of the rest of the UK let alone to the US.
What no one seems to be addressing is that although this study involves thousands of passes, it's really a case of n=1 because they're all the same rider riding in the same general area. I don't believe that riders are uniform in any number of possibly relevant factors, nor do I believe that drivers behave the same in different contexts. I have no idea how the habits of drivers in Bath relate to those of the rest of the UK let alone to the US.
Likes For PeteHski:
#84
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
I had to change my post before sending when I translated 3 feet from the curb as being the "FRAP" position, then realized that in the original study, that would be the FLAP position.
#85
Been Around Awhile
Join Date: Oct 2004
Location: Burlington Iowa
Posts: 29,755
Bikes: Vaterland and Ragazzi
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked 1,391 Times
in
942 Posts
From ArsTechnica today: Huge Tesla leak reveals thousands of safety concerns, privacy problems
The shocking toll of Tesla’s Autopilot - Washington Post
Extract:
Former NHTSA senior safety adviser Missy Cummings, a professor at George Mason University’s College of Engineering and Computing, said the surge in Tesla crashes is troubling.
“Tesla is having more severe — and fatal — crashes than people in a normal data set,” she said in response to the figures analyzed by The Post. One likely cause, she said, is the expanded rollout over the past year and a half of Full Self-Driving, which brings driver-assistance to city and residential streets. “The fact that … anybody and everybody can have it. … Is it reasonable to expect that might be leading to increased accident rates? Sure, absolutely.”
“Tesla is having more severe — and fatal — crashes than people in a normal data set,” she said in response to the figures analyzed by The Post. One likely cause, she said, is the expanded rollout over the past year and a half of Full Self-Driving, which brings driver-assistance to city and residential streets. “The fact that … anybody and everybody can have it. … Is it reasonable to expect that might be leading to increased accident rates? Sure, absolutely.”
An interesting discussion that included Dr. Missy Cummings (cited in above quote) about human interaction to include complacency and distraction when dealing with so-called autonomous controls of machinery is below.
Discussion especially relevant to this thread can be found at timeline 10:15 through 27:00, though the entire presentation is worthwhile. It may prove illuminating to those who believe that collisions resulting from distracted driving will be eliminated or even reduced by so-called autonomous operation that depends on constant human monitoring for safe operation.
#86
Senior Member
Join Date: Mar 2012
Location: Tallahassee, FL
Posts: 4,654
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1455 Post(s)
Likes: 0
Liked 858 Times
in
471 Posts
Tesla is having more severe — and fatal — crashes than people in a normal data set,”
I doubt there is statistically significant data to back up that claim.
I doubt there is statistically significant data to back up that claim.
#87
Senior Member
Join Date: Oct 2015
Location: San Diego, California
Posts: 4,077
Bikes: Velo Orange Piolet
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 2227 Post(s)
Liked 2,006 Times
in
971 Posts
17 fatalities, 736 crashes: The shocking toll of Tesla’s Autopilot
WaPo gifted link:https://wapo.st/3oZDBsk
Likes For tyrion:
#88
Senior Member
Join Date: Mar 2012
Location: Tallahassee, FL
Posts: 4,654
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1455 Post(s)
Likes: 0
Liked 858 Times
in
471 Posts
The article presents both the quoted claim and Musk's claims to the contrary. Neither are actually supported.
I certainly believe that FSD is not ready for prime time, but this article is more of an opinion piece that a solid analysis.
I certainly believe that FSD is not ready for prime time, but this article is more of an opinion piece that a solid analysis.
#89
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
“Tesla is having more severe — and fatal — crashes than people in a normal data set,” she said in response to the figures analyzed by The Post"
Silly me, but I suspect she has a lot better sense of what a normal data set looks like than you.
#90
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
Well, that's an assertion. WaPo actually shows that Musk has been lying about the data, and that the real data is awful. It's not ready, but it was ok to put it on 400,000 Teslas is a pretty weird assertion.
#93
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
You've just committed a classic example of the fallacy fallacy. You offered no evidence for your assertion that her statement was unsupported other than your "doubt." By definition, that automatically pits your qualifications against hers.
If you actually posted something specific about the data and why it didn't support the conclusion, my reference to her qualifications to refute that would have been fallacious, but you didn't, and it wasn't. Nice try though
Likes For livedarklions:
#94
Been Around Awhile
Join Date: Oct 2004
Location: Burlington Iowa
Posts: 29,755
Bikes: Vaterland and Ragazzi
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked 1,391 Times
in
942 Posts
Ignoring the evidence presented by an Authority on the subject of human interaction with autonomous controls is another.
Professor Mary (Missy) Cummings received her B.S. in Mathematics from the US Naval Academy in 1988, her M.S. in Space Systems Engineering from the Naval Postgraduate School in 1994, and her Ph.D. in Systems Engineering from the University of Virginia in 2004. A naval officer and military pilot from 1988-1999, she was one of the U.S. Navy’s first female fighter pilots. She is a Professor in the George Mason University Mechanical Engineering, Electrical and Computer Engineering and Computer Science departments. She is an American Institute of Aeronautics and Astronautics (AIAA) Fellow, and recently served as the senior safety advisor to the National Highway Traffic Safety Administration. Her research interests include embedded artificial intelligence in safety-critical systems, assured autonomy, human-systems engineering, and the ethical and social impact of technology.
Dr. Missy Cummings, Professor at George Mason University and former Senior Safety Advisor to the National Highway Traffic Safety Administration (NHTSA), joined the CCAT Distinguished Lecture Series to discuss her lessons learned in that role and the use of systems engineering principles.The recent presentation at the Center for Connected and Automated Transportation, University of Michigan Transportation Research Institute may be informative for those interested in why a Tesla with the latest self-driving "Beta" software version didn't stop for pedestrian in crossing. More importantly it addresses issues created when drivers depend on such software to safely drive the car without their input or constant attention It also addresses the issue of manufacturers that demonstrate low regard for safety standards or regulations. It probably will mean nothing to people who prefer to shut out information that doesn't fit their own perception of reality.
A few slides from the presentation are extracted:


The presentation is summarized in an article at Autoblog
Autonomy expert Missy Cummings explains why even driver assists can be trouble
Extract:
The NYT article that discusses the subject and Dr. Cummings experience cited in the Autoblog article above is at:
Carmakers Are Pushing Autonomous Tech. This Engineer Wants Limits.
Relevant to any discussion of so-called autonomous vehicles, is the reaction and rhetoric of some people to any discussion that throws shade on Tesla's cavalier approach to testing "beta software" on public streets.
Extract:
Professor Mary (Missy) Cummings received her B.S. in Mathematics from the US Naval Academy in 1988, her M.S. in Space Systems Engineering from the Naval Postgraduate School in 1994, and her Ph.D. in Systems Engineering from the University of Virginia in 2004. A naval officer and military pilot from 1988-1999, she was one of the U.S. Navy’s first female fighter pilots. She is a Professor in the George Mason University Mechanical Engineering, Electrical and Computer Engineering and Computer Science departments. She is an American Institute of Aeronautics and Astronautics (AIAA) Fellow, and recently served as the senior safety advisor to the National Highway Traffic Safety Administration. Her research interests include embedded artificial intelligence in safety-critical systems, assured autonomy, human-systems engineering, and the ethical and social impact of technology.
Dr. Missy Cummings, Professor at George Mason University and former Senior Safety Advisor to the National Highway Traffic Safety Administration (NHTSA), joined the CCAT Distinguished Lecture Series to discuss her lessons learned in that role and the use of systems engineering principles.The recent presentation at the Center for Connected and Automated Transportation, University of Michigan Transportation Research Institute may be informative for those interested in why a Tesla with the latest self-driving "Beta" software version didn't stop for pedestrian in crossing. More importantly it addresses issues created when drivers depend on such software to safely drive the car without their input or constant attention It also addresses the issue of manufacturers that demonstrate low regard for safety standards or regulations. It probably will mean nothing to people who prefer to shut out information that doesn't fit their own perception of reality.
A few slides from the presentation are extracted:


The presentation is summarized in an article at Autoblog
Autonomy expert Missy Cummings explains why even driver assists can be trouble
Extract:
Yet the very nature of these systems makes constant vigilance hard, as Dr. Missy Cummings will tell you. Fresh off a New York Times profile, the George Mason University engineering professor, former NHTSA senior adviser for safety, and longtime autonomy and robotics researcher delivered a lecture last week at the University of Michigan’s Center for Connected and Automated Transportation. We’ve attached a video, above, and as college lectures go, it’ll hold your attention.
First, let’s just interject here that when Cummings was appointed to her role at NHTSA, Elon Musk called her “extremely biased against Tesla,” which had the effect of siccing his fans on her — she received death threats from Tesla-stans, and her family had to move out of their home for a time. Now, for a credible understanding of how human beings interact with high-performance technology, who are you going to believe, an auto company CEO, or an Annapolis grad who was one of the Navy’s first female fighter pilots? Only one of these human beings can land an F/A-18 on an aircraft carrier.
When she went to NHTSA, Cummings said, she had “been complaining about NHTSA for years” and relished the chance to attempt fixing it. Musk is no fan of the agency either, so you'd think the enemy of Elon’s enemy is his friend. And the two of them presumably share a common goal: safer cars. Musk might even appreciate her sense of humor. When at Duke University, she named its Humans and Autonomy Lab — HAL.
Yet she had to have a security assessment done on the venue where she gave the U of M lecture. “Trust me, everyone hates me today,” she told the audience. “All the manufacturers, NHTSA, my 15-year-old daughter, everyone hates me.” (Her daughter has her learner’s permit; imagine being taught to drive by one of the foremost authorities on vehicle safety, your mom.)
But, Cummings says, “I’m good with that. I don’t mind being the bad cop because I do think we’re at the most important time in history since we figured out brake lights and headlights. I think that we are in the scariest time in transportation with autonomy and technology.
“I think we’re making some big mistakes.”
Her fat-chance goal is for the perpetually put-upon and glacially paced NHTSA to impose some order on the Wild West of these technologies. Musk put an embryonic FSD on the streets because he could. There was nothing to stop Tesla from beta-testing the system among an unwitting public.
In an analysis Cummings sent NHTSA last fall of driver-assist systems from GM, Ford, Tesla, et al, she determined that cars using these systems that were involved in fatal crashes were traveling over the speed limit in 50% of the cases, while those with serious injuries were speeding in 42% of cases. In crashes that did not involve ADAS, those figures were 29% and 13%. So one simple solution would be for NHTSA to mandate speed limiters on these systems. “The technology is being abused by humans,” she told the Times. “We need to put in regulations that deal with this.”
The technology by definition lulls you. Cummings described an experiment at Duke in which she and other researchers placed 40 test subjects behind the wheel of a driving simulator for a four-hour “trip” using adaptive cruise. At the 2½-hour mark, a moose ambled slowly across the road. Only one test subject had presence enough to avoid the moose — the other 39 clobbered it.
Autopilot, Super Cruise, Blue Cruise ... she's grown to dislike active driver assists that are, or can be tricked into being, hands-free. "They put it on, whatever version of 'autopilot' they have, and then they relax. They relax, because indeed, that's what they've been told." They might be paying attention "for the most part," she said. For the most part is not enough.
As for full autonomy, Cummings laid out a description of the learning curve that humans and now technology have to climb, from first acquiring a basic skill to ultimately full expertise, when we've mastered the skill-based reasoning that helps us know when to break a rule to get out of an unusual situation safely. Handling uncertainty is the hump that technology, which is inherently rules-based, can't seem to get over.
She showed how an autonomous vehicle was stopped in its tracks during testing because it interpreted a movers’ truck as not just a truck, but as a collection of a truck, four poles, traffic signs, a fence, a building, a bus, and “a gigantic person who was about to attack.”
That’s an eight-year-old example, she admits, but “still very much a problem,” as illustrated by the now-infamous phantom-braking crash in the Bay Bridge tunnel in San Francisco last Thanksgiving, in which the driver of a Tesla blamed "Full Self-Driving" for mysteriously changing lanes and then slamming on the brakes, resulting in an eight-car pileup that injured a 2-year-old child.
“And for all you Tesla fanboys who are jumping on Twitter right now so you can attack me, I’m here to tell you that it’s not just a Tesla problem,” she said. “All manufacturers who are dealing in autonomy are dealing with this problem” — her next examples involving GM Cruise cars in San Francisco, including one that apparently attempted to drive through an active firefighting scene. "San Francisco, oh boy, they're at their wits' end with Cruise."
But despite the problems, she also offers praise. "Even though I just complained a lot about Cruise, I am in awe of Cruise and of Waymo and of all the other car companies out there that have not had any major crashes. They have not killed anybody since the Uber issue, So I am very amazed ... I think the self-driving community has done a very good job of policing themselves."
Those are just some highlights. It's a fascinating lecture from someone who seems to be sincerely working to keep you and me safe. And it’s well worth an hour of your time.
Meanwhile, take a cue from AAA and Missy Cummings: Don’t trust, don't let your guard down, don't relax behind the wheel — that’s never been more true than it is now.
First, let’s just interject here that when Cummings was appointed to her role at NHTSA, Elon Musk called her “extremely biased against Tesla,” which had the effect of siccing his fans on her — she received death threats from Tesla-stans, and her family had to move out of their home for a time. Now, for a credible understanding of how human beings interact with high-performance technology, who are you going to believe, an auto company CEO, or an Annapolis grad who was one of the Navy’s first female fighter pilots? Only one of these human beings can land an F/A-18 on an aircraft carrier.
When she went to NHTSA, Cummings said, she had “been complaining about NHTSA for years” and relished the chance to attempt fixing it. Musk is no fan of the agency either, so you'd think the enemy of Elon’s enemy is his friend. And the two of them presumably share a common goal: safer cars. Musk might even appreciate her sense of humor. When at Duke University, she named its Humans and Autonomy Lab — HAL.
Yet she had to have a security assessment done on the venue where she gave the U of M lecture. “Trust me, everyone hates me today,” she told the audience. “All the manufacturers, NHTSA, my 15-year-old daughter, everyone hates me.” (Her daughter has her learner’s permit; imagine being taught to drive by one of the foremost authorities on vehicle safety, your mom.)
But, Cummings says, “I’m good with that. I don’t mind being the bad cop because I do think we’re at the most important time in history since we figured out brake lights and headlights. I think that we are in the scariest time in transportation with autonomy and technology.
“I think we’re making some big mistakes.”
Her fat-chance goal is for the perpetually put-upon and glacially paced NHTSA to impose some order on the Wild West of these technologies. Musk put an embryonic FSD on the streets because he could. There was nothing to stop Tesla from beta-testing the system among an unwitting public.
In an analysis Cummings sent NHTSA last fall of driver-assist systems from GM, Ford, Tesla, et al, she determined that cars using these systems that were involved in fatal crashes were traveling over the speed limit in 50% of the cases, while those with serious injuries were speeding in 42% of cases. In crashes that did not involve ADAS, those figures were 29% and 13%. So one simple solution would be for NHTSA to mandate speed limiters on these systems. “The technology is being abused by humans,” she told the Times. “We need to put in regulations that deal with this.”
The technology by definition lulls you. Cummings described an experiment at Duke in which she and other researchers placed 40 test subjects behind the wheel of a driving simulator for a four-hour “trip” using adaptive cruise. At the 2½-hour mark, a moose ambled slowly across the road. Only one test subject had presence enough to avoid the moose — the other 39 clobbered it.
Autopilot, Super Cruise, Blue Cruise ... she's grown to dislike active driver assists that are, or can be tricked into being, hands-free. "They put it on, whatever version of 'autopilot' they have, and then they relax. They relax, because indeed, that's what they've been told." They might be paying attention "for the most part," she said. For the most part is not enough.
As for full autonomy, Cummings laid out a description of the learning curve that humans and now technology have to climb, from first acquiring a basic skill to ultimately full expertise, when we've mastered the skill-based reasoning that helps us know when to break a rule to get out of an unusual situation safely. Handling uncertainty is the hump that technology, which is inherently rules-based, can't seem to get over.
She showed how an autonomous vehicle was stopped in its tracks during testing because it interpreted a movers’ truck as not just a truck, but as a collection of a truck, four poles, traffic signs, a fence, a building, a bus, and “a gigantic person who was about to attack.”
That’s an eight-year-old example, she admits, but “still very much a problem,” as illustrated by the now-infamous phantom-braking crash in the Bay Bridge tunnel in San Francisco last Thanksgiving, in which the driver of a Tesla blamed "Full Self-Driving" for mysteriously changing lanes and then slamming on the brakes, resulting in an eight-car pileup that injured a 2-year-old child.
“And for all you Tesla fanboys who are jumping on Twitter right now so you can attack me, I’m here to tell you that it’s not just a Tesla problem,” she said. “All manufacturers who are dealing in autonomy are dealing with this problem” — her next examples involving GM Cruise cars in San Francisco, including one that apparently attempted to drive through an active firefighting scene. "San Francisco, oh boy, they're at their wits' end with Cruise."
But despite the problems, she also offers praise. "Even though I just complained a lot about Cruise, I am in awe of Cruise and of Waymo and of all the other car companies out there that have not had any major crashes. They have not killed anybody since the Uber issue, So I am very amazed ... I think the self-driving community has done a very good job of policing themselves."
Those are just some highlights. It's a fascinating lecture from someone who seems to be sincerely working to keep you and me safe. And it’s well worth an hour of your time.
Meanwhile, take a cue from AAA and Missy Cummings: Don’t trust, don't let your guard down, don't relax behind the wheel — that’s never been more true than it is now.
Carmakers Are Pushing Autonomous Tech. This Engineer Wants Limits.
Relevant to any discussion of so-called autonomous vehicles, is the reaction and rhetoric of some people to any discussion that throws shade on Tesla's cavalier approach to testing "beta software" on public streets.
Extract:
Dr. Cummings has long warned that this can be a problem — in academic papers, in interviews and on social media. She was named senior adviser for safety at NHTSA in October 2021, not long after the agency began collecting crash data involving cars using driver-assistance systems.
Mr. Musk responded to her appointment in a post on Twitter, accusing her of being “extremely biased against Tesla,” without citing any evidence. This set off an avalanche of similar statements from his supporters on social media and in emails to Dr. Cummings.
She said she eventually had to shut down her Twitter account and temporarily leave her home because of the harassment and death threats she was receiving at the time. One threat was serious enough to be investigated by the police in Durham, N.C., where she lived.
Many of the claims were nonsensical and false. Some of Mr. Musk’s supporters noticed that she was serving as a board member of Veoneer, a Swedish company that sells sensors to Tesla and other automakers, but confused the company with Velodyne, a U.S. company whose laser sensor technology — called lidar — is seen as a competitor to the sensors that Tesla uses for Autopilot.
“We know you own lidar companies and if you accept the NHTSA adviser position, we will kill you and your family,” one email sent to her said.
Jennifer Homendy, who leads the National Transportation Safety Board, the agency that investigates serious automobile crashes, and who has also been attacked by fans of Mr. Musk, told CNN Business in 2021 that the false claims about Dr. Cummings were a “calculated attempt to distract from the real safety issues.”
Mr. Musk responded to her appointment in a post on Twitter, accusing her of being “extremely biased against Tesla,” without citing any evidence. This set off an avalanche of similar statements from his supporters on social media and in emails to Dr. Cummings.
She said she eventually had to shut down her Twitter account and temporarily leave her home because of the harassment and death threats she was receiving at the time. One threat was serious enough to be investigated by the police in Durham, N.C., where she lived.
Many of the claims were nonsensical and false. Some of Mr. Musk’s supporters noticed that she was serving as a board member of Veoneer, a Swedish company that sells sensors to Tesla and other automakers, but confused the company with Velodyne, a U.S. company whose laser sensor technology — called lidar — is seen as a competitor to the sensors that Tesla uses for Autopilot.
“We know you own lidar companies and if you accept the NHTSA adviser position, we will kill you and your family,” one email sent to her said.
Jennifer Homendy, who leads the National Transportation Safety Board, the agency that investigates serious automobile crashes, and who has also been attacked by fans of Mr. Musk, told CNN Business in 2021 that the false claims about Dr. Cummings were a “calculated attempt to distract from the real safety issues.”
Likes For I-Like-To-Bike:
#95
Senior Member
Join Date: Mar 2012
Location: Tallahassee, FL
Posts: 4,654
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1455 Post(s)
Likes: 0
Liked 858 Times
in
471 Posts
Absent support, her assertion is by definition unsupported. Show me the valid and reliable data.
#96
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
You seem to think it's a fallacy to believe that expertise matters. Sorry, I don't think you've shown any reason anyone should care about your "doubts".
#97
Senior Member
Join Date: Mar 2012
Location: Tallahassee, FL
Posts: 4,654
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1455 Post(s)
Likes: 0
Liked 858 Times
in
471 Posts
I will freely admit that I have never seen such data sets, let alone analyzed them, but I'm not the one equating the value of my opinion with that of a trained expert who actually works with such data sets, you are.
You seem to think it's a fallacy to believe that expertise matters. Sorry, I don't think you've shown any reason anyone should care about your "doubts".
You seem to think it's a fallacy to believe that expertise matters. Sorry, I don't think you've shown any reason anyone should care about your "doubts".
#98
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
As an attorney, you should at least be familiar with the concept of evidence. Doesn't matter how esteemed an attorney is, nor how many cases they have prevailed in. They still have to provide evidence. It is not incumbent on opposing counsel to prove the evidence does not exist.
You don't get it--this is a newspaper story, not a trial or a scientific journal article. You made an allegation that one of their sources was likely talking through her hat without "statistically significant" proof. If you're going to make such an allegation, the burden of proof is on you. You can look at the evidence yourself to make the case. Compare the Tesla data set to a "normal" data set. No one's stopping you.
Hate to break it to you, but you're not anyone's counsel here, you're just a guy talking through his hat because somebody taught you the phrase "statistically significant". You know as well as I do that neither of us is particularly qualified to evaluate her claims. All I know is that she is an extremely qualified person who has actually seen the evidence and characterizes it one way, and you're just some guy who hasn't looked at the evidence telling us what you expect it will probably show or not show. So from my perspective as a consumer of information, which of you two should I find more credible? There's nothing fallacious about that--it's not an appeal to authority, it's that I know you're likely basing your "opinion" on nothing other than your presuppositions as a Musk fan-boy.
Likes For livedarklions:
#99
Senior Member
Join Date: Jun 2008
Location: Louisville, KY
Posts: 13,446
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 4014 Post(s)
Liked 2,770 Times
in
1,673 Posts
Who exactly in this little analogy of yours do you think you are? The judge or someone's counsel? News flash--you're neither.
You don't get it--this is a newspaper story, not a trial or a scientific journal article. You made an allegation that one of their sources was likely talking through her hat without "statistically significant" proof. If you're going to make such an allegation, the burden of proof is on you. You can look at the evidence yourself to make the case. Compare the Tesla data set to a "normal" data set. No one's stopping you.
Hate to break it to you, but you're not anyone's counsel here, you're just a guy talking through his hat because somebody taught you the phrase "statistically significant". You know as well as I do that neither of us is particularly qualified to evaluate her claims. All I know is that she is an extremely qualified person who has actually seen the evidence and characterizes it one way, and you're just some guy who hasn't looked at the evidence telling us what you expect it will probably show or not show. So from my perspective as a consumer of information, which of you two should I find more credible? There's nothing fallacious about that--it's not an appeal to authority, it's that I know you're likely basing your "opinion" on nothing other than your presuppositions as a Musk fan-boy.
You don't get it--this is a newspaper story, not a trial or a scientific journal article. You made an allegation that one of their sources was likely talking through her hat without "statistically significant" proof. If you're going to make such an allegation, the burden of proof is on you. You can look at the evidence yourself to make the case. Compare the Tesla data set to a "normal" data set. No one's stopping you.
Hate to break it to you, but you're not anyone's counsel here, you're just a guy talking through his hat because somebody taught you the phrase "statistically significant". You know as well as I do that neither of us is particularly qualified to evaluate her claims. All I know is that she is an extremely qualified person who has actually seen the evidence and characterizes it one way, and you're just some guy who hasn't looked at the evidence telling us what you expect it will probably show or not show. So from my perspective as a consumer of information, which of you two should I find more credible? There's nothing fallacious about that--it's not an appeal to authority, it's that I know you're likely basing your "opinion" on nothing other than your presuppositions as a Musk fan-boy.

__________________
Bikes: 1996 Eddy Merckx Titanium EX, 1989/90 Colnago Super(issimo?) Piu(?),1990 Concorde Aquila(hit by car while riding), others in build queue "when I get the time"
Bikes: 1996 Eddy Merckx Titanium EX, 1989/90 Colnago Super(issimo?) Piu(?),
Likes For himespau:
#100
Tragically Ignorant
Join Date: Jun 2018
Location: New England
Posts: 15,613
Bikes: Serotta Atlanta; 1994 Specialized Allez Pro; Giant OCR A1; SOMA Double Cross Disc; 2022 Allez Elite mit der SRAM
Mentioned: 62 Post(s)
Tagged: 0 Thread(s)
Quoted: 8183 Post(s)
Liked 9,077 Times
in
5,047 Posts
Well, the explaining how evidence works to an attorney by telling him what attorneys do is a classic Dunning Kruger move. If he really wants to borrow a legal concept, I don't think he has standing to make his objection.
Not getting into the vaxx and mask stuff cuz that'll just get the thread locked and I don't remember what he posted on the subject.