> Waymo probably won’t have any luck sticking a vandalism charge on someone who puts a cone on the hood of its vehicles.
Good luck automating society if the public sentiment starts to really turn against these kinds of initiatives. Us nerds can argue about the merits of markets all day long, but we can’t expect the society we’re leaving behind to willingly participate in this new techno world order.
Automation is so brittle. It’s hard enough to design a fault-tolerant web, I can’t even imagine how we’re gonna do it in the physical world at scale with things like self-driving cars when they’re so easy to disable.
It’s like infosec, the attack surface of the system is far too big to properly guard against unless you’re willing to spend considerably and make the necessary sacrifices.
Markets are not even the thing. Nerds need to get to grips with ethics or at least social signaling.
E.g. how much of an burden is a service allowed to be on society at large? Most of the "disruption"-based startup models are based around exploiting some rule that societies had before that. If you have to trick society into using your service, consider you just might be on the wrong side of history.
Without knowing the details, when a nerd friend talks to me about a fictional self driving taxi start up, the first thing they would need to convince me of is that such a thing is even benefitial given the fact that it is space that is limited on city streets. Then I would have worries a out how that corp is socializing part of it's costs (by using public roads, by usings ways of attracting particular workers etc).
Don't get me wrong, those might all be ways to make good money, I don't dispute that. But as someone living in a city my stakes are different. I want to be able to get from A to B without being stuck in traffic with 100 empty cars, where the workers receive low pay and the road wear costs are paid by my taxes.
Sure, it is harder to find a way to earn money from which all people profit, but it is not impossible. If your corp is acting like a leech on society, don't be surprised if it is treated as such.
Playing video games I use to joke about people getting angry about cheats, I argued the game must be really important to them if they get so emotional.
A thing I keep thinking: A culture isn't weird until compared or confronted with a different one.
You got the impression that people don't like nerds building ridiculous things but nothing could be further from the truth.
Normals absolutely love these things, they define their being more than anything. They love to see nerds fool around AND FAIL. It's the same with investors (who are grouped along with get rich quick schemes) they love to see investors lose everything. Bitcoin is twice as exciting! Cant wait to see the pyramid implode.
I see a picture one time of a street in China and mistakenly thought it was my city in the Netherlands. It was just the context that gave it away. All the houses in those streets are different they have tiles with round holes in the side walk, the same trees and even people cycling wearing the usual... the usual cloths everyone else around the world is wearing.
This is why people take vacations far away, if you don't go far enough everything looks exactly the same, an endless regurgitation of the same mindless replication.
The answer is to do public transport with rollercoasters or turn all streets into half pipes, replace the main road with 2 looooong wave pools, use the AI to make self driving pony carts, busses with 10 floors.
I know, I know, people didn't want running water or a sewage system. Sadly for them it didn't fail but there are plenty of failures for them to laugh at, get them though their day at their useless job.
So much text and you didn't give me one substantial argument why I should change my mind.
Space is a concern in cities. It is one of the concerns. If you have the same roads-to-building-ratio public transport will always beat individual transport, because that is just basic geometry.
You like to frame this as progress, but being able to transport less people while using more resources and space isn't progress.
People are going to adjust. I remember the local forum in town with people freaking out about the Google streetview being a massive invasion of privacy and saying how they're going to chase those cars and destroy the cameras. Some people are going to have fun here and there, but I think we'll be collectively over it very soon.
> in the physical world at scale with things like self-driving cars when they’re so easy to disable.
Cars are easy to disable in general. You can permanently disable lots of cars on a random street before you get caught. This is temporary and harmless, so people aren't really serious about it.
> People are going to adjust. I remember the local forum in town with people freaking out about the Google streetview being a massive invasion of privacy and saying how they're going to chase those cars and destroy the cameras
Open Google Maps and try to streetview over Germany. People (or provacy laws) did not adjust, and I can’t say I’m sad about that - even though not having SV can be annoying.
I thought the underlying reason behind the laws and lack of StreetView in Germany was the general higher sensitivity to privacy, not just “laws”. Laws, after all, are created by elected people who should interpret the spirit of the population.
Of course people change over time to live with a technology, but I would not take it for granted that it _has_ to happen always, and not without reshaping the tech to make it more palatable to people’s concerns and values.
I'm reminded of jaywalking. Pedestrians had the right of way until the marketing departments of automotive companies decided to turn walking on the street into a crime.
Public sentiment can be easily manipulated with sufficient resources.
Social tolerance is a market. We are seeing the results of running a debt in that market.
Small scale automation is brittle. Large scale automation is far easier, especially if you're willing to purpose build infrastructure for it. This would be light rail, trains, and busses.
Also like infosec... if these cars "behaved better" then these faults would remain obscure and possibly unpatched for longer periods of time. It should make everyone wonder what other fault modes these vehicles have.
These cars that have millions of dollars in R&D sunk into them can be rendered useless by simply covering the sensors. It reminds me of the Keurig situation where they tried limiting the functionality of the machine to Keurig-signed (barcoded) K-cups. People immediately figured out how to bypass this by taping a keurig lid to the sensor so it would run every time. People are going to look for ways to get around this.
> especially if you're willing to purpose build infrastructure for it.
These companies are gonna have a hard time pulling that off politically. Public transit serves everyone (which this technology also kills), why should governments support infra built for private companies like this? The technology is not public by design. It would be political suicide.
Yes, any change that's expected to go on a social scale needs buyin from the average people. If these AVs were more expected to be used to help the more vulnerable people and in general improve everyone's life it would be cast in a different light.
Right now it has more of a technobro and rich people vibe, hoping it doesn't fall into the same kind of rabbit hole as the Google Glass and other useful innovations that just had horrendous PR and the worst people to evangelise it.
I agree with you regarding the brittleness, the small things that count.
This is the typical automation cycle. Automation is a pipeline and you keep adding/removing tasks in it.
The cone placement discovery reminds me of the feared birds, that can literally bring down an airplane when sucked into the turbine. Large steel construct vs small bird - bird wins. Large steel construct vs plastic thing - plastic wins. ;)
I’ve wondered by watching the SF rollout of more autonomous vehicles if there is central monitoring of many cars by humans.
While not fully automated, one or two people could monitor a fleet of hundreds of cars for problems with the vehicles and send one or two human agents in the field to investigate and fix.
Still stands to be wildly more efficient than a normal car ownership.
This is a social phenomenon with other autonomous urban robots already; cars aren't the first. I don't have links handy right now but I know you see security robots being pushed over, or food delivery bots boxed into places. Perhaps for various similar reasons: They're recording, they're annoying, they're slow, they're in the way, etc...
I don't know how much the "we don't consent to being on camera" argument holds water, since there's already lots of private cameras monitoring public urban places and roads. Maybe the fact that the cameras are up close and in your face is valid, though.
I don't condone mischief, but I also understand being annoyed. IMO, all vehicles on public roads should have a driver for now. But I welcome our autonomous-driving future.
I think there’s probably some game theory at play. I can imagine the mindset of the vandals mirroring a train of thought that I share somewhat:
1. This autonomous robot is not doing anything for me, and is helping someone else realize their capitalistic aspirations.
2. By vandalizing the robot, a victimless crime, I can do my part to just slightly close the gap between I, one who has gotten the short end of the economic stick, and some corporation/ceo who is on the other end of the economic spectrum, profiteering.
Thus slightly narrowing the gap between the haves and have nots, which for most individuals isn’t realizable in any other way
IANAL, but if you are recorded without consent, the video evidence may not be presentable in court. In CA, and about 15 other states, you need two-party consent for the evidence to be admissible. I remember talking about it to a former DA who told me this, so I believe them.
Of course, you can’t be expected to have privacy in places like public parks according to CA law, so I’m not sure where the law stands on that.
This wasn’t organic. It was organized by a tiny group of activists looking for an anti-tech flashpoint.
EDIT: I’ll be more direct. It’s a hoax. A fiction. An attempt to portray widespread safety concerns among the populace that do not exist.
There’s a hearing coming up at the board the state of California appointed regarding expansion of the robotaxis.
Alongside that vote, the city’s quality of operation has noticeably deteriorated. Encampments are more common, drug intoxication to the point of collapse and physical decay is highly visible. Robberies have become commonplace.
These effects are now beyond the point where they be explained by pointing at flat trend data, or as a temporary effect of more severe pandemic measures.
A number of local politicians have been facing significant heat over this dysfunction
National and international media publications have begun to distrust local publications for papering over issues and have amplified that pressure.
At that point, those SF politicians began a sudden unified push attacking self-driving cars. Self-driving cars have recently become more reliable than manually driven taxis in SF in terms of accidents per miles driven.
The attacks escalated to the point that false data was submitted against robotaxis to the California board, mischaracterizing accidents where another driver impacted the self-driven car while stopped. That falsehood was later discovered.
When those same people began promoting this “night of the cone” meme, the origin was obvious to even the most casual observer.
Robotaxis have been up and operating for some time. The city is in trouble and it needs the revenue innovation brings.
All movements need their initiators. The question is if they are still actively driving it, recruiting, funding, promoting. If not, then I feel it was an organic thing...
> “Not only is this understanding of how AVs operate incorrect, but this is vandalism and encourages unsafe and disrespectful behavior on our roadways,” the company said in a statement. “We will notify law enforcement of any unwanted or unsafe interference of our vehicles on public roadways.”
In London as in many other cities, on a Friday night, large parts of town basically turn into a giant open air pub. No one would drive there voluntarily but there’s the occasional Uber / Taxi / Delivery / Maintenance driver passing through at a crawl. People will get out of the way but it’ll sometimes require creativity to get through.
It would take a real asshole to deliberately cause trouble for some low paid driver and fortunately there are few of those in the general population. People can empathise, “guy’s just doing a job, no need to be a dick, hang on I’ll move my pint…”
However, given the amount of crap people have to put up with with the likes of ‘unexpected item in bagging area’ from the supposedly smart and better machines then I feel like the ‘asshole to a machine’ threshold is much much lower.
> Other opponents like the San Francisco Taxi Workers Alliance and the Alliance for Independent Workers have protested the spread of robotaxis, which they say will eliminate the need for taxi and ride-hail drivers.
Ummm… that’s clearly the point. What’s the controversial bit? Stable owners decried that house and buggy drivers were being competed by cars and driving them out of business. I’d much rather have safer streets because taxis are ever vigilant rather than be in an Uber with a driver boasting about how he figured out how to drive 18 hours a day and gets almost no sleep.
Sorry. You have regulations you can think of that prevent accidents by professional drivers? Trucking is already insanely regulated and still sees all sorts of corner cutting and safety violations. And even in ostensibly safe companies, accidents happen. It’s near impossible to simultaneously compete on cost and have safety - squeezing margins is at odds with following the rules (and that’s assuming a huge leap that the rules actually are helpful - very often no analysis is done on the ROI of any particular investment).
No I'm specifically addressing OPs point about drivers pulling 14 hour shifts - that's easily regulated everywhere. Anything beyond that is your interpretation.
>>It’s near impossible to simultaneously compete on cost and have safety
....airlines exist and compete on cost with extremely stringent safety requirements?
That regulations should be strict and they should be followed strictly. If you live somewhere where they are not enforced then that's not a problem of regulation. But I also don't know why we moved onto this topic from me saying that robotaxis shouldn't be allowed. I got the impression that the argument was that they should because they are competing against overworked Uber drivers pulling 14 hour shifts - to which I'll say again that Uber drivers doing 14 hour shifts isn't some unmovable law of the universe, it can and should be prevented.
I find it interesting how on HN is verboten to even suggest that maybe some innovation pushed on us and our cities by these companies isn't actually great and isn't the same as progress of humanity. It's just shit tech used to make some techbros money at the cost of making our cities worse. But apparently that's a "luddite" view now?
What's your basis for the conclusion that it's "shit tech"? Have you tried riding in one? I've found it to be an incredible product, like using the iPhone for the first time. There's one lady who's posted a lot of YouTube videos of Waymo. She's taken over 500 rides. So there are clearly people who get a lot of value out of it.
Also, no one is making money off self driving tech any time soon, people have been working on it for more than a decade and it costs billions. It might be another decade before it is profitable.
Then there is this lady who highlights fundamental issues with the way these vehicles operate. This isn't a teething problem that will be resolved with an OTA - it shows the tech isn't ready to be used on public streets and probably shouldn't ever be.
I’m an SF resident that is an anti-car as anyone but this is so misguided.
The real problems are:
- vehicles owned and operated by a single driver. These take up precious space for parking instead of living
- auto accidents due to human error. Tired, drunk, texting or just aggressive drivers cause auto accidents.
The slow streets movement is great. Put up cones where all cars can’t use certain streets.
Self driving cars and the companies behind them haven’t earned all our trust yet, but the sooner society shares robo taxis as a replacement for single owner vehicles the better.
Exactly, these are the things that count. The cones are a short term nuisance. Basically, the people putting them will probably get bored/distracted pretty soon. Or run out of cones. Or get caught and fined. Etc. They'll grumble and stamp their feet for a while and that will be it.
The more autonomous cars fail to do any harm whatsoever, the less of a case against them people will have. Most of the current fears are fueled by minor incidents, weird little bugs that these systems still have, and lots of irrational outrage and fears. And of course there are lots of armchair "experts" providing all sorts of reasons why the autonomous cars should not and cannot possibly work, even while they can easily be observed to be working just fine. That disconnect only stretches so far before it breaks.
Raw economics will do the rest. Insurers will penalize manual driving and scrutinize things like liability when accidents happen. The raw math insurers use is already against manual driving and they already like cars with lots of safety features. Cities will want to reclaim the space occupied by useless parking spaces (i.e. expect to pay for those). And drivers licenses will become much more expensive as those are no longer required to get from A to B and will come with a lot more scrutiny.
Many of the supposed minor incidents aren't even real. Two recent examples that got a lot of attention were the mass shooting incident where a Cruise allegedly blocked first responders to a mass shooting (it didn't, the police later clarified), and a Waymo that supposedly blocked a bus and forced passengers to disembark (it was the other way around, the bus broke down and blocked the AV). Also, SF officials got caught misrepresenting safety numbers, and were called out by the state.
At this point I'm very skeptical of issues that get attention. There definitely have been legitimate problems, but it seems like they're rare enough that people would tolerate them, thus fake incidents need to be created. This cone thing also might backfire because now even when there's a real incapacitated AV people will speculate it was due to external mischief.
It’s ironic that “safe street” activists are protesting against technology that produces safer streets.
“That one time the self driving car blocked a fire truck”, sure, but statistically on a per-rider basis are we seeing more or less harm/mischief than with human drivers or taxis? I haven’t seen any credible analysis of this question.
Similarly “they killed one person”, ok, how many human drivers killed pedestrians last year? You’ll find it’s more than self-driving cars, but again we need the per-rider (really, per passenger-mile driven) statistics.
A general pattern - even if self-driving cars were 10x safer, there would still be pushback if they are sub-human in any single niche case. We will cause many (thousands) people to die by being irrational like this.
Not to dismiss the concerns - they just need to be quantified and taken in context.
Because it has more sensors, 360° view and a 1000x faster reaction time, over time - if not already - cars with AI are safer. The perception of an AI going rogue is much more impactful than a drunken driver on the highway though, as people can't relate to an AI (yet?).
Or murder/kill. The novel verb unalive is used for “kill” to evade filters regardless of whether the agent and patient of that action are the same or different individuals.
But they clearly didn't intend the word "suicide" because it wouldn't even fit in that sentence grammatically. If they are referring to the death of Elaine Herzberg, they may have intended it to mean "murdered" or "killed", but certainly not "suicide[d]". How could an SDC suicide a human being???
Well I'm not a TikTok user, but it wouldn't surprise me if such a filter was equally uncomfortable with the words "murdered" or "killed" in videos. They probably really don't want death threats going viral.
Audiovisual media seems increasingly sensitive to "naughty words". I've seen a lot of YouTubers who feel that it is necessary to censor words like "m*rder" or "r*pe" or "f*ck" -- exactly this way in their text descriptions, and they literally bleep the words if mentioned on video.
I'm confused, because I thought that the scanners which detect TOS violations were becoming smarter, and it doesn't seem like an isolated swear word--or mention of sexual assault using plain English--would trigger an actual TOS violation, but it seems that these creators are wary of attracting attention by using the words uncensored.
> And most importantly they require streets that are designed for cars, not people or transit
I think this is one of the most interesting point the protestors make.
This kind of solution doesn't seem to integrate very well with new practice of urbanisms where pedestrian, bike and traffic intermingled in the same space and travel distance are shorter.
But I am European so I am probably bias because our city are already partially car free.
How did we go from SDC (which is not in Wikipedia) https://en.wikipedia.org/wiki/SDC to "AV" which is a wildly overused acronym that will only sow confusion? Can we nip that one in the bud please?
Just install a 'hood popper' that automatically opens and closes the hood remotely. The cone will just fall off. You can use it for repairs at the same time.
The only thing I like about robotaxis is that we are calling them "robotaxis". I'm sure that won't last, but I am really sick of the referring to a taxi as an "uber" or the somehow even more ridiculous "ride hailing service".
https://news.ycombinator.com/item?id=36624172