ISF Logo   IS Forum
Forum Index Register Members List Events Mark Forums Read Help

Go Back   International Skeptics Forum » General Topics » Science, Mathematics, Medicine, and Technology
 


Welcome to the International Skeptics Forum, where we discuss skepticism, critical thinking, the paranormal and science in a friendly but lively way. You are currently viewing the forum as a guest, which means you are missing out on discussing matters that are of interest to you. Please consider registering so you can gain full use of the forum features and interact with other Members. Registration is simple, fast and free! Click here to register today.
Reply
Old 23rd March 2018, 12:20 PM   #201
blutoski
Penultimate Amazing
 
blutoski's Avatar
 
Join Date: Jan 2006
Posts: 11,451
Originally Posted by JoeMorgue View Post
"I'm not against tech, I just want to make sure it's safe enough" is always the battlecry of anti-tech people.
With all due respect, it's also always the battlecry of people who want safe streets, drugs, and nuclear reactors.



Originally Posted by JoeMorgue View Post
"Safe enough" is an open ended standard you can keep going forever.
Yep. That's life, not everything is binary. Humans have to make cost/benefit judgements, risk/benefit judgements.

To me, this is no different than when I deal with quacks. There's an element out there that considers people like me who want to see clinical trials of efficacy before releasing drugs to the public to be Luddites. Well, screw 'em.

I agree that safety's a continuum with no boundaries, but that's no reason to scrap the discussion about acceptable thresholds. In fact: that line of reasoning is a named fallacy, called [Continuum Fallacy].
__________________
"Sometimes it's better to light a flamethrower than curse the darkness." - Terry Pratchett
blutoski is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 12:24 PM   #202
smartcooky
Philosopher
 
smartcooky's Avatar
 
Join Date: Oct 2012
Location: Nelson, New Zealand
Posts: 7,954
Originally Posted by phunk View Post
No. This started with you making a claim:
I asked you to support your claim and you tried to shift the burden of proof to me. It's your claim.
No, it started with Meadmaker making a claim (which you appear to support)

The burden of proof is on those making the claim (Meadmaker) and those supporting it (you).

Originally Posted by phunk View Post
Oops, you just blew your own claim out of the water.
No, I didn't

Originally Posted by phunk View Post
At best your analysis puts a lower bound on the time to react.
And I also put a best case on the PRT of 1.5 seconds instead of the more likely average of 2.5 seconds. I put an exceptional driver in the driver's seat and showed that he still could not have avoided a fatal collision.

Originally Posted by phunk View Post
The real view that the driver and the camera had was better than this crappy video by an unknown amount.
That is Meadmaker's (and your) claim. I disagree that it would be anything significant, certainly not significant enough to affect the outcome... even a half second (which would be an enormous amount under the circumstances) would not have been enough. It would add about 34 feet to the available stopping distance... not enough to avoid a fatal collision; the car would still travel past the point of impact before an average driver gets his foot on the brake.

Also, of significance is that an extra half a second puts the pedestrian further to the left and therefore further out of the driver's central vision, making it less likely that the driver would see her. Additionally, I would expect any good driver (in a right drive country) to be looking to the right of their driving line for pedestrians not the left, so would to take her even further out of the driver's central vision.

Originally Posted by phunk View Post
We don't know how much time someone would actually have to react
But we can make an estimate, and the best estimate is that there still would not be enough time for the average driver to avoid a fatal collision.

Meadmaker's claim is that "a human driver" would have avoided this accident. I have at best refuted that claim and an absolute worst, showed that it is extremely unlikely that even a very good driver could have done so.
__________________
► 9/11 was a terrorist attack by Islamic extremists; 12 Apollo astronauts really did walk on the Moon; JFK was assassinated by Lee Harvey Oswald,who acted alone.
► Never underestimate the power of the Internet to lend unwarranted credibility to the colossally misinformed. - Jay Utah
► Heisenberg's Law - The weirdness of the Universe is inversely proportional to the scale at which you observe it, or not.

Last edited by smartcooky; 23rd March 2018 at 12:26 PM.
smartcooky is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 12:29 PM   #203
theprestige
Penultimate Amazing
 
Join Date: Aug 2007
Posts: 26,637
Originally Posted by JoeMorgue View Post
"I'm not against tech, I just want to make sure it's safe enough" is always the battlecry of anti-tech people.

"Safe enough" is an open ended standard you can keep going forever.
This analysis does not seem to be informed by real-world experience.

For example, I work for a technology organization. Everybody I work with is 100% pro-technology. It's a basic job requirement for us.

But as stewards of our employer's success, we do not embrace new tech willy-nilly. We study it, find its flaws, and adapt it to meet our needs in a way that protects our interests. "I'm not against tech, I just want to make sure it's safe enough" could very appropriately be a battlecry of my team.

And it turns out that this principle is not very open-ended in practice. We have limited resources, business priorities, and opportunity costs whenever we choose to do one thing instead of another. So "safe enough" is actually a defined standard. The goal is to meet the standard without wasting resources by exceeding the standard. We decide what we need to do to make it work, and make it safe enough. Then we do that. Then we move on to the next project.

I think it is a little uncivil - and very unrealistic - to go straight from someone expressing safety concerns about new technology, to implying that they are intractably anti-technology. Slow your roll, man.
theprestige is online now   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 12:29 PM   #204
Elagabalus
Master Poster
 
Join Date: Dec 2013
Posts: 2,355
Originally Posted by theprestige View Post
...Driver's education in American public schools doesn't elevate my driving self that far above my natural state ...
I think hitting someone while taking the DMV driving test would probably result in an automatic fail.


Originally Posted by theprestige View Post
...So if I'm willing to forgive a pedestrian for being stupid, careless, and inattentive and getting into a car accident as a result, then I'm also willing to forgive her doppelganger behind the wheel of the car.
I don't really fault the Uber Driver. I think being a watchdog over an autonomous vehicle would be a gig from hell. I have to be as attentive as if I were driving - only I'm not. I don't think humans can do it.
Elagabalus is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 12:31 PM   #205
theprestige
Penultimate Amazing
 
Join Date: Aug 2007
Posts: 26,637
Originally Posted by Elagabalus View Post
I don't really fault the Uber Driver. I think being a watchdog over an autonomous vehicle would be a gig from hell. I have to be as attentive as if I were driving - only I'm not. I don't think humans can do it.
Yeah, I don't think they can, either.
theprestige is online now   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 12:34 PM   #206
BobTheCoward
Penultimate Amazing
 
Join Date: Nov 2010
Posts: 12,730
In this concept of thresholds,I keep going back to the real chance that cops will declare that it is legally acceptable for a human of driver to kill that lady. In that case, how should thresholds account for that reality?
BobTheCoward is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 01:07 PM   #207
theprestige
Penultimate Amazing
 
Join Date: Aug 2007
Posts: 26,637
Originally Posted by BobTheCoward View Post
In this concept of thresholds,I keep going back to the real chance that cops will declare that it is legally acceptable for a human of driver to kill that lady. In that case, how should thresholds account for that reality?
Don't thresholds already count for that reality? We do recognize that not all car accidents are crimes.
theprestige is online now   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 01:09 PM   #208
BobTheCoward
Penultimate Amazing
 
Join Date: Nov 2010
Posts: 12,730
Originally Posted by theprestige View Post
Don't thresholds already count for that reality? We do recognize that not all car accidents are crimes.
Let's change crimes to fault.
BobTheCoward is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 01:20 PM   #209
theprestige
Penultimate Amazing
 
Join Date: Aug 2007
Posts: 26,637
Originally Posted by BobTheCoward View Post
Let's change crimes to fault.
I have no opinion about that.
theprestige is online now   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 01:21 PM   #210
BobTheCoward
Penultimate Amazing
 
Join Date: Nov 2010
Posts: 12,730
Originally Posted by theprestige View Post
I have no opinion about that.
It happens. I appreciate your honesty.
BobTheCoward is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 01:53 PM   #211
William Parcher
Show me the monkey!
 
William Parcher's Avatar
 
Join Date: Jul 2005
Posts: 19,986
Sensor firm Velodyne 'baffled' by Uber self-driving death

Originally Posted by BBC News
The firm that designed the sensors on the Uber self-driving car that killed a woman this week has said its technology was not to blame.

San Jose-based Velodyne told the BBC it was "baffled" by the incident, adding its equipment was capable of seeing in the dark...

Velodyne Lidar president Marta Hall told the BBC it would not be advising its customers to halt tests in the wake of the Arizona death because "we do not believe the accident was due to Lidar".

Instead, the company is pointing to Uber's on-board computer as potentially being to blame, Ms Hall said.

"Our Lidar can see perfectly well in the dark, as well as it sees in daylight, producing millions of points of information.

However, it is up to the rest of the system to interpret and use the data to make decisions. We do not know how the Uber system of decision-making works."

While it makes use of third-party hardware, Uber's self-driving cars use software developed in house...
http://www.bbc.com/news/technology-43523286
__________________
Bigfoot believers and Bigfoot skeptics are both plumb crazy. Each spends more than one minute per year thinking about Bigfoot.
William Parcher is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 02:11 PM   #212
rwguinn
Penultimate Amazing
 
rwguinn's Avatar
 
Join Date: Apr 2003
Location: 16 miles from 7 lakes
Posts: 11,046
Originally Posted by JoeMorgue View Post
"I'm not against tech, I just want to make sure it's safe enough" is always the battlecry of anti-tech people.

"Safe enough" is an open ended standard you can keep going forever.
To be fair, "not quite ready" is often the engineer's mantra.
We really do like futzing with thing...
__________________
"Political correctness is a doctrine,...,which holds forth the proposition that it is entirely possible to pick up a turd by the clean end."
"
I pointed out that his argument was wrong in every particular, but he rightfully took me to task for attacking only the weak points." Myriad http://forums.randi.org/showthread.php?postid=6853275#post6853275
rwguinn is online now   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 02:14 PM   #213
JoeMorgue
Self Employed
Remittance Man
 
JoeMorgue's Avatar
 
Join Date: Nov 2009
Location: Florida
Posts: 10,837
My point, for all the difference it makes at this point, is that people are using "It needs to be safer!" when they mean "I'm not comfortable with it."

Automated cars are already safer than human drivers. That's... one of those "fact" things.

You can't argue that automated cars "aren't there yet" while being completely comfortable with human drivers without their being an anti-tech factor in your argument somewhere.

Again it's the anti-vaxxer argument. "But vaccines aren't 100% safe in every scenario!"
__________________
"Ernest Hemingway once wrote that the world is a fine place and worth fighting for. I agree with the second part." - Detective Sommerset, Se7en
JoeMorgue is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 02:21 PM   #214
Dr. Keith
Not a doctor.
 
Dr. Keith's Avatar
 
Join Date: Jun 2009
Location: Texas
Posts: 15,616
Originally Posted by William Parcher View Post
Sensor firm Velodyne 'baffled' by Uber self-driving death
Originally Posted by BBC News
While it makes use of third-party hardware, Uber's self-driving cars use software developed in house...
http://www.bbc.com/news/technology-43523286
Is this a subtle dig at Waymo?
__________________
I once proposed a fun ban.

Suffering is not a punishment not a fruit of sin, it is a gift of God.
He allows us to share in His suffering and to make up for the sins of the world. -Mother Teresa
Dr. Keith is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 02:29 PM   #215
William Parcher
Show me the monkey!
 
William Parcher's Avatar
 
Join Date: Jul 2005
Posts: 19,986
Was Uber's driverless car crash avoidable? Experts say yes

Originally Posted by CNET
After watching a dashboard video of the fatal collision, some autonomous-vehicle analysts say the car's sensors should've detected the pedestrian...

The Arizona police, along with federal investigators, have been examining the vehicle and accident site, while also gathering information about the technology in the car, the vehicle operator and the pedestrian. They've also been studying a video that captured the crash Sunday night from the car's dashboard camera...

Cortica, a technology company that develops autonomous artificial intelligence, analyzed the video and provided its evaluation exclusively to CNET. Its system detected Herzberg at 0.9 second before impact when the car was about 50 feet away. Cortica's CEO, Igal Raichelgauz, said that would have been enough time for an autonomous vehicle to react and save Herzberg's life.

"The advantage of machine response time and control, the right actions could be made to certainly mitigate the damage," Raichelgauz said.

Tempe police say the car didn't slow down or swerve as Herzberg appeared on the road. It hit her traveling at 38 mph...
https://www.cnet.com/news/was-ubers-...ld-have-braked
__________________
Bigfoot believers and Bigfoot skeptics are both plumb crazy. Each spends more than one minute per year thinking about Bigfoot.
William Parcher is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 02:39 PM   #216
blutoski
Penultimate Amazing
 
blutoski's Avatar
 
Join Date: Jan 2006
Posts: 11,451
Originally Posted by JoeMorgue View Post
My point, for all the difference it makes at this point, is that people are using "It needs to be safer!" when they mean "I'm not comfortable with it."
Reading minds, are we? Or is this just strawpersoning.


Originally Posted by JoeMorgue View Post
Automated cars are already safer than human drivers. That's... one of those "fact" things.
I doubt it. I'd love you to prove that claim.


Originally Posted by JoeMorgue View Post
You can't argue that automated cars "aren't there yet" while being completely comfortable with human drivers without their being an anti-tech factor in your argument somewhere.
Why not? What if automated cars are 10,000 times less safe? 10x less safe? Do we even know?


Originally Posted by JoeMorgue View Post
Again it's the anti-vaxxer argument. "But vaccines aren't 100% safe in every scenario!"
Strawperson. Nobody here has demanded 100% safe.

I'd settle for knowing any relative ratio. Are they 50% as safe? 10% as safe? Does anybody know?
__________________
"Sometimes it's better to light a flamethrower than curse the darkness." - Terry Pratchett
blutoski is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 02:44 PM   #217
blutoski
Penultimate Amazing
 
blutoski's Avatar
 
Join Date: Jan 2006
Posts: 11,451
Originally Posted by blutoski View Post
I'd settle for knowing any relative ratio. Are they 50% as safe? 10% as safe? Does anybody know?
Just as a level-setter... I work in medical research. If I came across a collegue with your eyeroll attitude about safety, ie: health canada wanting "quote" evidence "unquote" that my drugs didn't kill people before I ran clinical trials, I'd consider him a sociopath.

I'm always looking for that benefit/risk ratio. What's the proportional safety for these autonomous vehicles?

My concern is that the proprietary approach means they're pinkie swearing that their prior safety testing is sufficient to merit exposure to the public. I need convincing.
__________________
"Sometimes it's better to light a flamethrower than curse the darkness." - Terry Pratchett
blutoski is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 02:45 PM   #218
JoeMorgue
Self Employed
Remittance Man
 
JoeMorgue's Avatar
 
Join Date: Nov 2009
Location: Florida
Posts: 10,837
Originally Posted by blutoski View Post
I'd settle for knowing any relative ratio. Are they 50% as safe? 10% as safe? Does anybody know?
I've already linked to articles multiple times in this thread. People just either aren't listening or don't want to hear it.

Here so they can be ignored again.

http://bigthink.com/ideafeed/googles...iculously-safe

https://www.huffingtonpost.com/entry...b03b105b44bc6b

https://www.theatlantic.com/technolo...merica/407956/
__________________
"Ernest Hemingway once wrote that the world is a fine place and worth fighting for. I agree with the second part." - Detective Sommerset, Se7en

Last edited by JoeMorgue; 23rd March 2018 at 02:46 PM.
JoeMorgue is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 02:55 PM   #219
GlennB
Loggerheaded, earth-vexing fustilarian
 
GlennB's Avatar
 
Join Date: Sep 2006
Location: Pie City, Arcadia
Posts: 22,738
Originally Posted by JoeMorgue View Post
Automated cars are already safer than human drivers. That's... one of those "fact" things.
How could you possibly know that?
__________________
"Even a broken clock is right twice a day. 9/11 truth is a clock with no hands." - Beachnut

Last edited by GlennB; 23rd March 2018 at 02:57 PM.
GlennB is online now   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 03:05 PM   #220
blutoski
Penultimate Amazing
 
blutoski's Avatar
 
Join Date: Jan 2006
Posts: 11,451
Originally Posted by JoeMorgue View Post
I've already linked to articles multiple times in this thread. People just either aren't listening or don't want to hear it.

Here so they can be ignored again.

http://bigthink.com/ideafeed/googles...iculously-safe

https://www.huffingtonpost.com/entry...b03b105b44bc6b

https://www.theatlantic.com/technolo...merica/407956/
I confess I only read the first piece, and it was a fact devoid op/ed. ie: industry fluff. Zero statisics in it. Are the others any better? (I only have so many years to live)
__________________
"Sometimes it's better to light a flamethrower than curse the darkness." - Terry Pratchett
blutoski is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 03:06 PM   #221
bruto
Penultimate Amazing
 
bruto's Avatar
 
Join Date: Jun 2005
Location: Way way north of Diddy Wah Diddy
Posts: 22,570
It gets a little confusing with two robocar threads going at the same time, but I think I forgot to post here earlier on the "safe enough" issue. If I somehow missed it I apologize for redundancy.

I think statistics can be read differently depending on who reads them and for what purpose, and I've gone on about this some already. That might include the idea of what is "safe enough." The factors involved in car accidents are many and varied and complex, and many drivers are undoubtedly more convinced of their ability than is warranted. Nonetheless, a simple "safer by x percent" statistic does not really answer directly.

I was trying to figure out a simplistic analogy for this, and came up with one:

Here in the US, about 600 thousand people die each year from heart disease. Of course some of those are just old folks and everyone has to die, but many are younger. It's a terrible scourge, and if we could reduce it significantly many things would improve - health care costs would go down, people would live longer, hearts would be broken less often, fewer jobs left unfinished, operas and works of art uncompleted, etc. etc.

So, imagine that a magical answer is found. A benevolent god or something says "I'll hold a plebiscite, and if you elect for it, I guarantee that in the wink of an eye the rate of death by heart disease will be reduced by 75 percent. Only 150 thousand people will die of it each year. The only down side is that those 150 thousand will be chosen completely at random. Nothing they do or do not do will change their fate.

Now some people would undoubtedly go ahead and vote for it as it means they can smoke and eat bacon and get fat with impunity and others would undoubtedly vote against it because they believe, rightly or not, that their style of living is heart healthy, and perhaps because they believe, rightly or wrongly, that their family history is favorable. Some would take the social point of view and vote for it for ideological reasons, and some would vote against it for other ideological reasons. Some would do it from ignorance, some from a preference for playing the odds.

Would you vote for it or against it?

The question, and I think it's a real one, or bound eventually to become a real one, is not at what point the overall statistics favor self driving cars, but at what point your social altruism accepts putting your fate as a driver in the hands of something over which you have no vestige of control.
__________________
I love this world, but not for its answers. (Mary Oliver)

Quand il dit "cuic" le moineau croit tout dire. (When he's tweeted the sparrow thinks he's said it all. (Jules Renard)
bruto is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 03:08 PM   #222
blutoski
Penultimate Amazing
 
blutoski's Avatar
 
Join Date: Jan 2006
Posts: 11,451
Originally Posted by blutoski View Post
I confess I only read the first piece, and it was a fact devoid op/ed. ie: industry fluff. Zero statisics in it. Are the others any better? (I only have so many years to live)
OK, out of charity, I read the 2nd one, which opens with this paragraph:

Quote:
If driverless cars deliver on their promise to eliminate the vast majority of fatal traffic accidents, the technology will rank among the most transformative public-health initiatives in human history.
I've highlighted the relevant part. This article is about estimates for the promise that this technology has if it succeeds. Which I think it will. My question is are these safe enough to put on the road right now and this article does not shed any light on that question.

That's 2 out of 3 articles. I'm not very motivated to read the 3rd.
__________________
"Sometimes it's better to light a flamethrower than curse the darkness." - Terry Pratchett
blutoski is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 03:12 PM   #223
smartcooky
Philosopher
 
smartcooky's Avatar
 
Join Date: Oct 2012
Location: Nelson, New Zealand
Posts: 7,954
Originally Posted by Elagabalus View Post
I don't really fault the Uber Driver. I think being a watchdog over an autonomous vehicle would be a gig from hell. I have to be as attentive as if I were driving - only I'm not. I don't think humans can do it.
Agree.

IMO, the human "watchdog" is nothing more than a homoeopathic panacea for the masses.
__________________
► 9/11 was a terrorist attack by Islamic extremists; 12 Apollo astronauts really did walk on the Moon; JFK was assassinated by Lee Harvey Oswald,who acted alone.
► Never underestimate the power of the Internet to lend unwarranted credibility to the colossally misinformed. - Jay Utah
► Heisenberg's Law - The weirdness of the Universe is inversely proportional to the scale at which you observe it, or not.
smartcooky is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 03:14 PM   #224
jimbob
Uncritical "thinker"
 
jimbob's Avatar
 
Join Date: Jan 2007
Location: UK
Posts: 17,658
Originally Posted by Elagabalus View Post


I don't really fault the Uber Driver. I think being a watchdog over an autonomous vehicle would be a gig from hell. I have to be as attentive as if I were driving - only I'm not. I don't think humans can do it.
Indeed, one intervention every 5,000 miles is about the worst level... Days of driving, doing nothing but alsso sufficiently frequent to cause problems.
__________________
OECD healthcare spending
Expenditure on healthcare
http://www.oecd.org/els/health-systems/health-data.htm
link is 2015 data (2013 Data below):
UK 8.5% of GDP of which 83.3% is public expenditure - 7.1% of GDP is public spending
US 16.4% of GDP of which 48.2% is public expenditure - 7.9% of GDP is public spending
jimbob is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 03:17 PM   #225
blutoski
Penultimate Amazing
 
blutoski's Avatar
 
Join Date: Jan 2006
Posts: 11,451
Originally Posted by smartcooky View Post
Agree.

IMO, the human "watchdog" is nothing more than a homoeopathic panacea for the masses.
I'm still fine with it being a requirement. Not all malfunctions are this shocking. If the vehicle goes on the friz and starts driving into a front yard, I'm sure a human will have the presence of mind to resume control, and that's a good backup IMO.

Having a human for the broad range of unforseeables is a good idea I think.
__________________
"Sometimes it's better to light a flamethrower than curse the darkness." - Terry Pratchett
blutoski is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 03:50 PM   #226
bruto
Penultimate Amazing
 
bruto's Avatar
 
Join Date: Jun 2005
Location: Way way north of Diddy Wah Diddy
Posts: 22,570
Originally Posted by BobTheCoward View Post
In this concept of thresholds,I keep going back to the real chance that cops will declare that it is legally acceptable for a human of driver to kill that lady. In that case, how should thresholds account for that reality?
I am not sure (as often is the case) what you're after here. First of all, it is not up to the cops to declare anything with regard to liability. They can decide whether an accident warrants a criminal violation, but that is separate from civil responsibility and follows its own rules. Sure the cops could, and probably do, determine whether an act is acceptable, though I think that's a poor choice of terms, when they determine whether a criminal act or offense occurred. They might deplore the incident and believe one thing or another, but if they see no evidence of a crime that would hold up in court, they likely won't charge it. If they see a fairly simple offense, they probably will write a ticket. It would then be up to the victim or the victim's heirs to determine whether to sue for a civil tort. This would probably be a civil suit against Uber's insurer. In cases like that liability is often apportioned, and how that's done varies from place to place. In some cases, often notoriously and often not so notoriously, a defendant with very deep pockets and a clear ability to prevent a wrong may be left liable far beyond his (or its) contribution to the act. If, for example, a drunk driver plows into a tree, he's liable for having the accident, but if when he does so a defectively designed gas tank explodes and burns him to death, the fact of who caused the accident is not identical to the fact of what caused the accident to become fatal. If it turns out the car maker could have prevented the death with a gas cap modification or the like, then they may be on the hook even though millions of safe drivers who did not hit trees never had a problem with the defective equipment. Car makers are notorious for doing the math on such things, and deciding that the cost of paying out judgments for those few accidents is less than the cost of fixing a problem. Exploding Pinto gas tanks, capsizing Corvairs and Explorers and delaminating Firestone tires, take a bow.

In the case under consideration here, I am guessing that a civil case could be quite complicated, and it would be up to Uber to convince a jury that the fault not only for the accident but for the death that ensued was that of the pedestrian, and that the result was not reasonably avoidable. I suspect that even though the initial fault was obviously that of the pedestrian, a substantial share of the responsibility for her death would fall on Uber - for both good and bad reasons, partly distrust and ignorance of technology, and partly a judgment of the likelihood that better avoidance would have mitigated the result.
__________________
I love this world, but not for its answers. (Mary Oliver)

Quand il dit "cuic" le moineau croit tout dire. (When he's tweeted the sparrow thinks he's said it all. (Jules Renard)
bruto is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 04:13 PM   #227
BobTheCoward
Penultimate Amazing
 
Join Date: Nov 2010
Posts: 12,730
Originally Posted by bruto View Post
I am not sure (as often is the case) what you're after here. First of all, it is not up to the cops to declare anything with regard to liability. They can decide whether an accident warrants a criminal violation, but that is separate from civil responsibility and follows its own rules. Sure the cops could, and probably do, determine whether an act is acceptable, though I think that's a poor choice of terms, when they determine whether a criminal act or offense occurred. They might deplore the incident and believe one thing or another, but if they see no evidence of a crime that would hold up in court, they likely won't charge it. If they see a fairly simple offense, they probably will write a ticket. It would then be up to the victim or the victim's heirs to determine whether to sue for a civil tort. This would probably be a civil suit against Uber's insurer. In cases like that liability is often apportioned, and how that's done varies from place to place. In some cases, often notoriously and often not so notoriously, a defendant with very deep pockets and a clear ability to prevent a wrong may be left liable far beyond his (or its) contribution to the act. If, for example, a drunk driver plows into a tree, he's liable for having the accident, but if when he does so a defectively designed gas tank explodes and burns him to death, the fact of who caused the accident is not identical to the fact of what caused the accident to become fatal. If it turns out the car maker could have prevented the death with a gas cap modification or the like, then they may be on the hook even though millions of safe drivers who did not hit trees never had a problem with the defective equipment. Car makers are notorious for doing the math on such things, and deciding that the cost of paying out judgments for those few accidents is less than the cost of fixing a problem. Exploding Pinto gas tanks, capsizing Corvairs and Explorers and delaminating Firestone tires, take a bow.

In the case under consideration here, I am guessing that a civil case could be quite complicated, and it would be up to Uber to convince a jury that the fault not only for the accident but for the death that ensued was that of the pedestrian, and that the result was not reasonably avoidable. I suspect that even though the initial fault was obviously that of the pedestrian, a substantial share of the responsibility for her death would fall on Uber - for both good and bad reasons, partly distrust and ignorance of technology, and partly a judgment of the likelihood that better avoidance would have mitigated the result.
I guess I'm giving too much weight to a finding by the police that a driver is not criminally liable for hitting someone.
BobTheCoward is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 06:02 PM   #228
EHocking
Philosopher
 
EHocking's Avatar
 
Join Date: Apr 2004
Posts: 6,547
Originally Posted by Dr. Keith View Post
...I trust the car more than the news report saying it was speeding. Mainly because Uber drivers seem to be trained to only drive within the speed limit,...
Uber trains its drivers?
I can only presume you're being facetious.

Uber's background checks on their drivers is exceptional.
As demonstrated by this incident.
__________________
"A closed mouth gathers no feet"
"Ignorance is a renewable resource" P.J.O'Rourke
Prayer: "a sophisticated way of pleading with thunderstorms." T.Pratchett
"It's all god's handiwork, there's little quality control applied", Fox26 reporter on Texas granite
Forum Birdwatching Webpage
EHocking is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 06:25 PM   #229
Meadmaker
Penultimate Amazing
 
Meadmaker's Avatar
 
Join Date: Apr 2004
Posts: 17,718
Originally Posted by smartcooky View Post
No, it started with Meadmaker making a claim (which you appear to support)
Sort of. You made a claim that no human could have stopped in time, and you based that claim on analysis of the video that showed 1.72 seconds (IIRC) from the time the woman's tennis shoes became visible until the time she was escorted out of this plane of existence.

So, that's a claim. That you made.

I made a counterclaim that the video does not show what the human eye would have seen. That ought to be an uncontroversial statement. And a further claim that a human would have perceived the motion more than 1.72 seconds before the crash, based on what I saw on the video. That part is more controversial, and indeed there are some exceptions you pointed out.

It's true that a person with poor night vision might have seen the person no sooner than she is visible in the video. It's true that a person who had been blinded by looking at a device might have suffered a temporary loss of night vision ability. It's true that all sorts of circumstances could have contributed to a human running into that pedestrian under similar circumstances.

On the other hand, an awful lot of experts have weighed in on the subject by now and, having seen the video, said that this is exactly the sort of accident that self driving cars ought to be able to avoid, and have expressed a desire to see the results of the inevitable investigation into exactly what happened to explain why the car did not react to avoid the accident.

So, I did indeed make a claim, which is that an alert human with normal eyesight, or eyesight corrected to normal, would have perceived a hazard earlier than the moment that her shoes are visible on that tape. Alas, I have no scientific data that I can point to to prove my case. We'll have to await more expert opinion on that subject.

I believe that your claim, which is that no human could have avoided the accident under the same circumstances, is suspect, and that the analysis backing it up (the 1.72 second time which came from the public video) is not a legitimate source of evidence, because the video from which the calculation is made is not the same as what a healthy human eye would see in those circumstances.
Meadmaker is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 11:03 PM   #230
bruto
Penultimate Amazing
 
bruto's Avatar
 
Join Date: Jun 2005
Location: Way way north of Diddy Wah Diddy
Posts: 22,570
Originally Posted by BobTheCoward View Post
I guess I'm giving too much weight to a finding by the police that a driver is not criminally liable for hitting someone.
I think the police may find that a driver is not always criminally liable for hitting someone.
__________________
I love this world, but not for its answers. (Mary Oliver)

Quand il dit "cuic" le moineau croit tout dire. (When he's tweeted the sparrow thinks he's said it all. (Jules Renard)
bruto is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 23rd March 2018, 11:11 PM   #231
bruto
Penultimate Amazing
 
bruto's Avatar
 
Join Date: Jun 2005
Location: Way way north of Diddy Wah Diddy
Posts: 22,570
Originally Posted by Meadmaker View Post
Sort of. You made a claim that no human could have stopped in time, and you based that claim on analysis of the video that showed 1.72 seconds (IIRC) from the time the woman's tennis shoes became visible until the time she was escorted out of this plane of existence.

So, that's a claim. That you made.

I made a counterclaim that the video does not show what the human eye would have seen. That ought to be an uncontroversial statement. And a further claim that a human would have perceived the motion more than 1.72 seconds before the crash, based on what I saw on the video. That part is more controversial, and indeed there are some exceptions you pointed out.

It's true that a person with poor night vision might have seen the person no sooner than she is visible in the video. It's true that a person who had been blinded by looking at a device might have suffered a temporary loss of night vision ability. It's true that all sorts of circumstances could have contributed to a human running into that pedestrian under similar circumstances.

On the other hand, an awful lot of experts have weighed in on the subject by now and, having seen the video, said that this is exactly the sort of accident that self driving cars ought to be able to avoid, and have expressed a desire to see the results of the inevitable investigation into exactly what happened to explain why the car did not react to avoid the accident.

So, I did indeed make a claim, which is that an alert human with normal eyesight, or eyesight corrected to normal, would have perceived a hazard earlier than the moment that her shoes are visible on that tape. Alas, I have no scientific data that I can point to to prove my case. We'll have to await more expert opinion on that subject.

I believe that your claim, which is that no human could have avoided the accident under the same circumstances, is suspect, and that the analysis backing it up (the 1.72 second time which came from the public video) is not a legitimate source of evidence, because the video from which the calculation is made is not the same as what a healthy human eye would see in those circumstances.
I agree, but would just add again that there is also a difference between avoiding any accident at all and mitigating it. Of course we cannot be sure, and probably can never be sure, what might have been done, but the issue is not just a binary matter of avoiding or not avoiding an accident. There's a big difference between being injured and being killed.
__________________
I love this world, but not for its answers. (Mary Oliver)

Quand il dit "cuic" le moineau croit tout dire. (When he's tweeted the sparrow thinks he's said it all. (Jules Renard)
bruto is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 24th March 2018, 01:58 AM   #232
shuttlt
Philosopher
 
Join Date: Aug 2008
Posts: 5,628
Originally Posted by Meadmaker View Post
I made a counterclaim that the video does not show what the human eye would have seen. That ought to be an uncontroversial statement. And a further claim that a human would have perceived the motion more than 1.72 seconds before the crash, based on what I saw on the video. That part is more controversial, and indeed there are some exceptions you pointed out.

It's true that a person with poor night vision might have seen the person no sooner than she is visible in the video. It's true that a person who had been blinded by looking at a device might have suffered a temporary loss of night vision ability. It's true that all sorts of circumstances could have contributed to a human running into that pedestrian under similar circumstances.

On the other hand, an awful lot of experts have weighed in on the subject by now and, having seen the video, said that this is exactly the sort of accident that self driving cars ought to be able to avoid, and have expressed a desire to see the results of the inevitable investigation into exactly what happened to explain why the car did not react to avoid the accident.

So, I did indeed make a claim, which is that an alert human with normal eyesight, or eyesight corrected to normal, would have perceived a hazard earlier than the moment that her shoes are visible on that tape. Alas, I have no scientific data that I can point to to prove my case. We'll have to await more expert opinion on that subject.

I believe that your claim, which is that no human could have avoided the accident under the same circumstances, is suspect, and that the analysis backing it up (the 1.72 second time which came from the public video) is not a legitimate source of evidence, because the video from which the calculation is made is not the same as what a healthy human eye would see in those circumstances.
Does this really matter? The speed in the conditions was chosen by the computer. I'm pretty sure in the UK, if a human hit somebody because visibility was so poor that they were unable to see somebody in the road ahead of them until it was too late, then the driver would be at fault for driving inappropriately fast in the conditions. Is that not the case in the US?
shuttlt is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 24th March 2018, 02:18 AM   #233
jimbob
Uncritical "thinker"
 
jimbob's Avatar
 
Join Date: Jan 2007
Location: UK
Posts: 17,658
Originally Posted by shuttlt View Post
Does this really matter? The speed in the conditions was chosen by the computer. I'm pretty sure in the UK, if a human hit somebody because visibility was so poor that they were unable to see somebody in the road ahead of them until it was too late, then the driver would be at fault for driving inappropriately fast in the conditions. Is that not the case in the US?
I have had this discussion before, talking about a previous accident and several posters thought it unreasonable to require drivers to drive at a speed where they could see beyond their stopping distance.

I also once had to avoid a tyre on the inside lane of the M40 in the dark - it wasn't a problem even downhill at 70mph, but was unexpected and was black on black. It did ram home to me that the unexpected can happen even on major roads.
__________________
OECD healthcare spending
Expenditure on healthcare
http://www.oecd.org/els/health-systems/health-data.htm
link is 2015 data (2013 Data below):
UK 8.5% of GDP of which 83.3% is public expenditure - 7.1% of GDP is public spending
US 16.4% of GDP of which 48.2% is public expenditure - 7.9% of GDP is public spending

Last edited by jimbob; 24th March 2018 at 02:26 AM.
jimbob is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 24th March 2018, 02:58 AM   #234
Klimax
NWO Cyborg 5960x (subversion VPUNPCKHQDQ)
 
Klimax's Avatar
 
Join Date: Feb 2008
Location: Starship Wanderer - DS9
Posts: 12,252
Originally Posted by Disbelief View Post
You can't make the assumption that since this particular application failed, that the technology isn't ready. One company or incident does not define the drive to autonomous, so you have to look at what is being done as a whole industry.
They are testing in (almost) best conditions. By definition they are far from ready. heavy rain/snow, mountain roads, roads in old inner cities of Europe,... There is far more fun ahead of them.

Originally Posted by JoeMorgue View Post
"I'm not against tech, I just want to make sure it's safe enough" is always the battlecry of anti-tech people.

"Safe enough" is an open ended standard you can keep going forever.
There is no evidence they are at least as safe as human drivers who don't have various advanced technologies available,. (LIDAR, night vision, fast processing...)

==

Frankly, give us access to data from LIDAR and night vision and it will improve driving on its own.
__________________
ModBorg

Engine: Ibalgin 400
Klimax is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 24th March 2018, 04:04 AM   #235
jimbob
Uncritical "thinker"
 
jimbob's Avatar
 
Join Date: Jan 2007
Location: UK
Posts: 17,658
Originally Posted by Klimax View Post
They are testing in (almost) best conditions. By definition they are far from ready. heavy rain/snow, mountain roads, roads in old inner cities of Europe,... There is far more fun ahead of them.



There is no evidence they are at least as safe as human drivers who don't have various advanced technologies available,. (LIDAR, night vision, fast processing...)

==

Frankly, give us access to data from LIDAR and night vision and it will improve driving on its own.
Like many others, my car has a radar that apparently would apply the brakes to prevent certain types of collision. It's not perfect - the sensor can get obscured by driving snow, for example but as an additional safety feature, supervising the human it's good. Far better than a human supervising a machine when you need quick reactions.
__________________
OECD healthcare spending
Expenditure on healthcare
http://www.oecd.org/els/health-systems/health-data.htm
link is 2015 data (2013 Data below):
UK 8.5% of GDP of which 83.3% is public expenditure - 7.1% of GDP is public spending
US 16.4% of GDP of which 48.2% is public expenditure - 7.9% of GDP is public spending
jimbob is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 24th March 2018, 04:09 AM   #236
3point14
Pi
 
3point14's Avatar
 
Join Date: Nov 2005
Posts: 14,850
Originally Posted by jimbob View Post
I have had this discussion before, talking about a previous accident and several posters thought it unreasonable to require drivers to drive at a speed where they could see beyond their stopping distance.
Okay, that's mental.



Originally Posted by Klimax View Post
Frankly, give us access to data from LIDAR and night vision and it will improve driving on its own.

Would it? Or would it make people seem safer so they drive more dangerously leaving the overall risk about the same? It seems fairly well documented that people adopt riskier behavior when they know there's something in place to moderate that risk.
__________________
Up the River!
3point14 is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 24th March 2018, 04:20 AM   #237
jimbob
Uncritical "thinker"
 
jimbob's Avatar
 
Join Date: Jan 2007
Location: UK
Posts: 17,658
Originally Posted by 3point14 View Post
Okay, that's mental.



I don't think that I am misrepresenting the discussion here

http://www.internationalskeptics.com...ce#post9977017
__________________
OECD healthcare spending
Expenditure on healthcare
http://www.oecd.org/els/health-systems/health-data.htm
link is 2015 data (2013 Data below):
UK 8.5% of GDP of which 83.3% is public expenditure - 7.1% of GDP is public spending
US 16.4% of GDP of which 48.2% is public expenditure - 7.9% of GDP is public spending
jimbob is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 24th March 2018, 05:21 AM   #238
shuttlt
Philosopher
 
Join Date: Aug 2008
Posts: 5,628
Originally Posted by jimbob View Post
Like many others, my car has a radar that apparently would apply the brakes to prevent certain types of collision. It's not perfect - the sensor can get obscured by driving snow, for example but as an additional safety feature, supervising the human it's good. Far better than a human supervising a machine when you need quick reactions.
My car also. This is one of the things that puzzles me about the case for self drive cars. You have a relatively simple problem - dealing quickly with dangerous situations. And a much more complicated problem - driving a car without supervision in all the circumstances that humans do. Surely if we are arguing that self drive is safer, that is the comparison, not with an unassisted human? What does self drive really add in terms of safety to humans driving, with automated systems that step in in an emergency? Wouldn't the best system to deal with this accident be a human actively driving and automation monitoring and stepping in as needed, then you have the best of both?

Is safety really the motivation for self drive?
shuttlt is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 24th March 2018, 05:39 AM   #239
W.D.Clinger
Illuminator
 
W.D.Clinger's Avatar
 
Join Date: Oct 2009
Posts: 3,355
Originally Posted by William Parcher View Post
Was Uber's driverless car crash avoidable? Experts say yes

Originally Posted by CNET
After watching a dashboard video of the fatal collision, some autonomous-vehicle analysts say the car's sensors should've detected the pedestrian...

The Arizona police, along with federal investigators, have been examining the vehicle and accident site, while also gathering information about the technology in the car, the vehicle operator and the pedestrian. They've also been studying a video that captured the crash Sunday night from the car's dashboard camera...

Cortica, a technology company that develops autonomous artificial intelligence, analyzed the video and provided its evaluation exclusively to CNET. Its system detected Herzberg at 0.9 second before impact when the car was about 50 feet away. Cortica's CEO, Igal Raichelgauz, said that would have been enough time for an autonomous vehicle to react and save Herzberg's life.

"The advantage of machine response time and control, the right actions could be made to certainly mitigate the damage," Raichelgauz said.

Tempe police say the car didn't slow down or swerve as Herzberg appeared on the road. It hit her traveling at 38 mph...
https://www.cnet.com/news/was-ubers-...ld-have-braked
The highlighted numbers are consistent.

Although an autonomous vehicle might have been able to react in time to save Herzberg's life, that is not the same as saying it wouldn't have hit her.

Furthermore, if those numbers are correct, then a human driver would probably not have been able to stop in time to avoid hitting her. If the car was traveling at 38 mph, that's 61 kph, or 17 meters per second, or 56 feet per second. The 50-foot distance at time of recognition is just over 15 meters. According to a braking distance calculator, a human driver probably wouldn't have been able to stop before the vehicle would travel 45 meters: 25 meters before the driver would react (in about 1.5 seconds) plus 20 meters with the brakes applied.

That's for a "medium-sized car with good tyres on a dry road".

The disturbing thing here is that the autonomous system apparently did not even try to stop. That implies some kind of malfunction.

Last edited by W.D.Clinger; 24th March 2018 at 05:41 AM.
W.D.Clinger is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Old 24th March 2018, 06:39 AM   #240
GlennB
Loggerheaded, earth-vexing fustilarian
 
GlennB's Avatar
 
Join Date: Sep 2006
Location: Pie City, Arcadia
Posts: 22,738
Originally Posted by W.D.Clinger View Post
The disturbing thing here is that the autonomous system apparently did not even try to stop. That implies some kind of malfunction.
I also find it disturbing that the systems didn't detect her (a substantial moving object) much earlier. I hasten to add I'm not comparing the car's performance here to when an attentive human might have detected her, just a competent AV in absolute terms.
__________________
"Even a broken clock is right twice a day. 9/11 truth is a clock with no hands." - Beachnut
GlennB is online now   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Reply With Quote Back to Top
Reply

International Skeptics Forum » General Topics » Science, Mathematics, Medicine, and Technology

Bookmarks

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -7. The time now is 12:30 PM.
Powered by vBulletin. Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.

This forum began as part of the James Randi Education Foundation (JREF). However, the forum now exists as
an independent entity with no affiliation with or endorsement by the JREF, including the section in reference to "JREF" topics.

Disclaimer: Messages posted in the Forum are solely the opinion of their authors.