The best place to spot a Waymo in the wild. Plus: New London Overground lines edge closer, the end of the night czar, and an underground swimming pool in East Dulwich.
This Stilgoe professor sounds like a bit of an ****. Looking at his quotes:
“Moving around London isn’t the same as moving around Phoenix, Arizona - what happens if police put tape across the road and ask vehicles to turn around?”
Well, while there definitely are many differences between the two places, that surely isn’t one of them. I’m confident that happens in Phoenix as well.
“Zebra crossings often involve eye contact between pedestrian and driver.”
They often do, but they don’t need to. The Highway Code says that the car must stop if the pedestrian is on the crossing, and that will be straightforward for Waymo to cope with. It might take a little while for people to get used to not being able to make eye contact any more, but it won’t be a showstopper, especially as most pedestrians in London seem to step out into the road without looking or even bothering to find a crossing first.
“Most people will think what problems do these cars solve?”
Well there’s the opportunity to massively reduce the accident rate, and to increase capacity in an under-served market, for starters. But in any case, this objection reminds me of every Luddite anywhere ever. “What problem does email solve anyway? We have a perfectly good postal system!”
"Your self-driving cars may look impressive, but what about [obvious technical challenge that hundreds of engineers have spent the last fifteen years investigating and ultimately overcoming]? Bet you never thought about that, did you?"
I understood the zebra crossings comment as the part where the car stops, the pedestrian stops, the pedestrian is telling the driver to go, the driver respects the highway code and does not advance and waits, and the pedestrian crosses with a bit of shame and the head down and saying thank you like 100 times.
Mostly I imagine these self driving cars routines as the very funny behaviours you see in GTA4 (at max traffic option) or Cyberpunk 2077 when someone does not behave in a predictable manner. Hope I am wrong.
A thing I noticed earlier, when the Wayve CEO was boasting of his trip through Central London in a massive, single-occupant, automated vehicle, is that they leave a generous amount of space to the car in front. Multiply this out to a large fleet and they surely increase congestion significantly compared to the bumper-to-bumper practices favoured by cars trying to navigate vehicle-choked streets. Just another way these companies plan to take up public space, potentially to the detriment of everyone else.
The spacing is such an interesting point I hadn’t considered. It’s fascinating that fundamentally there’s a slider equivalent within all the systems of “how much of an arse do you want the AI driver to be”.
But autonomous cars can respond much more smoothly to the flow of traffic around them – they don't do sudden starts, sudden stops, distracted by phone etc. (and of course they don't get in crashes). This ends up being much better for congested roads. From an article in MIT Technology Review:
"The team’s results show that by having an autonomous vehicle control its speed intelligently when a phantom jam starts to propagate, it’s possible to reduce the amount of braking performed further back down the line. The numbers are impressive: the presence of just one autonomous car reduces the standard deviation in speed of all the cars in the jam by around 50 percent, and the number of sharp hits to the brakes is cut from around nine per vehicle for every kilometer traveled to at most 2.5—and sometimes practically zero."
An academic - and govt adviser - being relentlessly negative about innovation is a turn-up for the books. If we're going to point all the ways in which Waymos could go wrong we should probably also mention the US evidence to date suggests they are much safer than human drivers.
The point about zebra crossings is a good one. But it's worth remembering in the US they have four way stops which Waymo cars manage to negotiate, which similarly rely on rules and behavioural cues.
The ridiculous attacks by Islington on Lime, for things they can barely control like rider behaviour or the fact the council have provided a fraction of the parking spaces ebikes need, is depressing. Totally self-defeating for anyone that wants to tackle car-caused congestion. Looking like a Zipcar situation on thr way.
Meanwhile SUVs keep getting larger and driver behaviour of these - far more lethal to anyone outside them than bicycle - or pavement parking is barely addressed.
I need to revisit the Great Lime Debate properly and look at the regulatory battle. Space on the road taken from car parking spaces to bike parking spaces is the thing that's lacking.
Yes, completely agree. The massive variations on how councils treat ebike rental firms is also so reminiscent of the same Zipcar talked as one of their main reasons to exist. Notice RBKC appear as one to charge Lime a lot to retrieve bikes. Again reminiscent.
Good content idea too! Watching your comments to James OB about content that does well. Lime Bikes are guaranteed to exercise people from both sides of the debate.
Waymos might come with a higher premium, but a more pleasant experience (smoother ride; no awkward or forced human interaction), higher safety standards, and more reliability will be worth it.
I regularly take Uber and Bolt, and it's a complete gamble. Drivers accept my request, waste 10 mins of my life, then cancel for a better job. And that usually happens multiple times. And don't get me started on the ridiculous pre-booking a ride situation: Uber never shows up, and I have to fear missing my flight.
A protest against self-driving vehicles is a protest for more dead cats, dogs and humans due to road accident. I look forward to the day I will never have to get in a taxi being driven by a maniacal driver with little regard for road safety again.
I have a lot of sympathy of course with the Uber (and other taxi) drivers who are concerned about their future.
But the other "expert" opinions on driverless cars seem to lack much weight - they're all problems that can be solved or worked through, and I don't doubt that driverless cars will deal with pedestrians, cyclists, zebra crossings, in a safer way than human drivers.
The streets of London as a whole will be safer as more cars become driverless, which is something that we all should welcome, including TfL and the Mayor.
Others have mentioned the question of space. In traffic, they might take up more space than 'regular' cars due to leaving bigger gaps, but they will also 'flow' more smoothly. But crucially, due to not needing drivers, who need to eat/sleep/see their families, they will spend more time on the road and less time parked, and when they are parked that can be at a single specific location with chargers rather than spread out across residential streets.
In the longer term, this probably opens up all sorts of opportunities that are hard to conceive of now. Things like that ill-fated Citymapper bus service, or the shared-ride thing Uber does - if there are cars that are near-constantly on-the-go, they can be much more efficient at picking up passengers and routing them. This in turn will improve things for buses.
I saw a Wayve on test driving into London past Apex Corner a few months ago, and it seemed a little jumpy dealing with other hurrying drivers changing lanes abruptly. Having driven in central London for thirty-odd years, I'm not convinced these cars are going to cope well with some of the counterintuitive stuff you have to do not to get stuck in traffic.
The airport runs make a lot more sense though. The idea of a smooth run to Gatwick and no one asking you about your holiday itinerary at 5am does sound appealing.
To be honest, that was a wonderful idea, half-arsedly executed. On the way to Gatwick or Heathrow, amazing, we used it every time. On the way back, you'd always find a whole bunch of cars with 15% battery life, because it didn't occur to them to have someone on site to charge the cars. Also, if you arrived and had to park in a non-Zipcar space because the official ones were full, no one could book that car, and it was basically defunct. Twice this happened to us, and both times the car was still sitting there when we returned, days later.
That's not even to mention getting marooned in Croydon once due to a Golf with a knackered battery and no charging cables.
I do massively miss Zipcar, but they were buggers for coming up with great new features, implementing them badly, then not bothering to advertise them.
I used to see a lot of Waymos being driven near the corner of brewery road and York Way - I guess that’s where the corporate office might have been? On them being slightly more expensive, I wonder if that includes tips? You absolutely wouldn’t tip a robotaxi when quite a lot of folks do tip drivers
Robotaxis in London are a great idea until you need to make a right turn onto a main road without traffic lights. London drivers know you have to edge out when theres an opening to your right. No way a risk averse robot is doing that. Likewise a Robot isn't going to flash you out of a sidestreet. People who don't drive don't realise London roads run on courtesy.
Handily the Waymos have those massive sensors, so all the Ubers, Cabbies and truck drivers whose jobs are under threat from the robots know to identify them and not to let them out of a side street. Justifiably so, as there's no driver to make eye contact with.
So you'll be sat in peaceful tranquility not having a driver to speak to while the human drivers backed up behind you honk more and more aggressively and your arrival time gets later and later.
I’m one of the early Waymo supporters that’s signed up for the surveys. I can’t wait for the Waymos to arrive! Safer drivers and smoother driving experience. I cant wait for it, bring on the progress
I wonder if the Uber drivers who are outraged at the prospect of technology reducing their worth were equally outraged on behalf of black cabbies ten years ago, who faced the prospect of technology reducing their worth by...Uber
Those Accident stats really need more digging into.
They say that they are 10x safer than a human driver. Is this measured in the US? A quick look on Wikipedia suggests that the road death rate in the US is 7x that of the UK by population, and about 2x by distance driven. (Similar ratios for other developed countries outside North America).
Hard to know the best comparison for city driving in somewhere like London - given distances will be relatively low and accident rate higher given density, I'd suggest it is closer to the population number.
That means the robots are about as safe as a human over here. Given the safety argument evaporates, and all of the other externalities, it is hard to see the benefit.
This Stilgoe professor sounds like a bit of an ****. Looking at his quotes:
“Moving around London isn’t the same as moving around Phoenix, Arizona - what happens if police put tape across the road and ask vehicles to turn around?”
Well, while there definitely are many differences between the two places, that surely isn’t one of them. I’m confident that happens in Phoenix as well.
“Zebra crossings often involve eye contact between pedestrian and driver.”
They often do, but they don’t need to. The Highway Code says that the car must stop if the pedestrian is on the crossing, and that will be straightforward for Waymo to cope with. It might take a little while for people to get used to not being able to make eye contact any more, but it won’t be a showstopper, especially as most pedestrians in London seem to step out into the road without looking or even bothering to find a crossing first.
“Most people will think what problems do these cars solve?”
Well there’s the opportunity to massively reduce the accident rate, and to increase capacity in an under-served market, for starters. But in any case, this objection reminds me of every Luddite anywhere ever. “What problem does email solve anyway? We have a perfectly good postal system!”
Go to the bottom of the class, professor.
"Your self-driving cars may look impressive, but what about [obvious technical challenge that hundreds of engineers have spent the last fifteen years investigating and ultimately overcoming]? Bet you never thought about that, did you?"
You know you’re talking about a fellow Londoner there? Some might say you sound like a bit of an ****
I understood the zebra crossings comment as the part where the car stops, the pedestrian stops, the pedestrian is telling the driver to go, the driver respects the highway code and does not advance and waits, and the pedestrian crosses with a bit of shame and the head down and saying thank you like 100 times.
Mostly I imagine these self driving cars routines as the very funny behaviours you see in GTA4 (at max traffic option) or Cyberpunk 2077 when someone does not behave in a predictable manner. Hope I am wrong.
A thing I noticed earlier, when the Wayve CEO was boasting of his trip through Central London in a massive, single-occupant, automated vehicle, is that they leave a generous amount of space to the car in front. Multiply this out to a large fleet and they surely increase congestion significantly compared to the bumper-to-bumper practices favoured by cars trying to navigate vehicle-choked streets. Just another way these companies plan to take up public space, potentially to the detriment of everyone else.
The spacing is such an interesting point I hadn’t considered. It’s fascinating that fundamentally there’s a slider equivalent within all the systems of “how much of an arse do you want the AI driver to be”.
But autonomous cars can respond much more smoothly to the flow of traffic around them – they don't do sudden starts, sudden stops, distracted by phone etc. (and of course they don't get in crashes). This ends up being much better for congested roads. From an article in MIT Technology Review:
"The team’s results show that by having an autonomous vehicle control its speed intelligently when a phantom jam starts to propagate, it’s possible to reduce the amount of braking performed further back down the line. The numbers are impressive: the presence of just one autonomous car reduces the standard deviation in speed of all the cars in the jam by around 50 percent, and the number of sharp hits to the brakes is cut from around nine per vehicle for every kilometer traveled to at most 2.5—and sometimes practically zero."
So less accidents. Obviously. Bumper to bumper is the way imbeciles drive. Another argument in favour of driverless vehicles. Thank you.
An academic - and govt adviser - being relentlessly negative about innovation is a turn-up for the books. If we're going to point all the ways in which Waymos could go wrong we should probably also mention the US evidence to date suggests they are much safer than human drivers.
The point about zebra crossings is a good one. But it's worth remembering in the US they have four way stops which Waymo cars manage to negotiate, which similarly rely on rules and behavioural cues.
The ridiculous attacks by Islington on Lime, for things they can barely control like rider behaviour or the fact the council have provided a fraction of the parking spaces ebikes need, is depressing. Totally self-defeating for anyone that wants to tackle car-caused congestion. Looking like a Zipcar situation on thr way.
Meanwhile SUVs keep getting larger and driver behaviour of these - far more lethal to anyone outside them than bicycle - or pavement parking is barely addressed.
I need to revisit the Great Lime Debate properly and look at the regulatory battle. Space on the road taken from car parking spaces to bike parking spaces is the thing that's lacking.
Yes, completely agree. The massive variations on how councils treat ebike rental firms is also so reminiscent of the same Zipcar talked as one of their main reasons to exist. Notice RBKC appear as one to charge Lime a lot to retrieve bikes. Again reminiscent.
Good content idea too! Watching your comments to James OB about content that does well. Lime Bikes are guaranteed to exercise people from both sides of the debate.
Camden New Journal front page a couple of weeks ago had a rant from a local councillor about Lime, continued page 2.
Waymos might come with a higher premium, but a more pleasant experience (smoother ride; no awkward or forced human interaction), higher safety standards, and more reliability will be worth it.
I regularly take Uber and Bolt, and it's a complete gamble. Drivers accept my request, waste 10 mins of my life, then cancel for a better job. And that usually happens multiple times. And don't get me started on the ridiculous pre-booking a ride situation: Uber never shows up, and I have to fear missing my flight.
A protest against self-driving vehicles is a protest for more dead cats, dogs and humans due to road accident. I look forward to the day I will never have to get in a taxi being driven by a maniacal driver with little regard for road safety again.
Ok the fact that zipcar has folded and instead we have waymo is making me just a bit angry.
If Waymo introduces a “pick up a table you bought from Enfield on Facebook Marketplace” option then I’m on board.
Or an "I can't be arsed to catch five buses to Hackney Wick for a birthday party, but I have to, so I'd rather drive than drink" option.
Sir we have Lime for a reason.
😂 Sir, I have Muswell Hill and Colney Hatch Lane between me and Hackney, and I'm a long way from King of the Mountains.
If it does not rain, that is.
I have a lot of sympathy of course with the Uber (and other taxi) drivers who are concerned about their future.
But the other "expert" opinions on driverless cars seem to lack much weight - they're all problems that can be solved or worked through, and I don't doubt that driverless cars will deal with pedestrians, cyclists, zebra crossings, in a safer way than human drivers.
The streets of London as a whole will be safer as more cars become driverless, which is something that we all should welcome, including TfL and the Mayor.
Others have mentioned the question of space. In traffic, they might take up more space than 'regular' cars due to leaving bigger gaps, but they will also 'flow' more smoothly. But crucially, due to not needing drivers, who need to eat/sleep/see their families, they will spend more time on the road and less time parked, and when they are parked that can be at a single specific location with chargers rather than spread out across residential streets.
In the longer term, this probably opens up all sorts of opportunities that are hard to conceive of now. Things like that ill-fated Citymapper bus service, or the shared-ride thing Uber does - if there are cars that are near-constantly on-the-go, they can be much more efficient at picking up passengers and routing them. This in turn will improve things for buses.
I saw a Wayve on test driving into London past Apex Corner a few months ago, and it seemed a little jumpy dealing with other hurrying drivers changing lanes abruptly. Having driven in central London for thirty-odd years, I'm not convinced these cars are going to cope well with some of the counterintuitive stuff you have to do not to get stuck in traffic.
The airport runs make a lot more sense though. The idea of a smooth run to Gatwick and no one asking you about your holiday itinerary at 5am does sound appealing.
Especially given RIP the Zipcar Flex self-drive to the airport option.
To be honest, that was a wonderful idea, half-arsedly executed. On the way to Gatwick or Heathrow, amazing, we used it every time. On the way back, you'd always find a whole bunch of cars with 15% battery life, because it didn't occur to them to have someone on site to charge the cars. Also, if you arrived and had to park in a non-Zipcar space because the official ones were full, no one could book that car, and it was basically defunct. Twice this happened to us, and both times the car was still sitting there when we returned, days later.
That's not even to mention getting marooned in Croydon once due to a Golf with a knackered battery and no charging cables.
I do massively miss Zipcar, but they were buggers for coming up with great new features, implementing them badly, then not bothering to advertise them.
I used to see a lot of Waymos being driven near the corner of brewery road and York Way - I guess that’s where the corporate office might have been? On them being slightly more expensive, I wonder if that includes tips? You absolutely wouldn’t tip a robotaxi when quite a lot of folks do tip drivers
Those are probably Wayves, I think! A name confusion which is going to cause ABSOLUTE CARNAGE for the next few years.
Ahh yes, you’re right! Carnage for sure…
Robotaxis in London are a great idea until you need to make a right turn onto a main road without traffic lights. London drivers know you have to edge out when theres an opening to your right. No way a risk averse robot is doing that. Likewise a Robot isn't going to flash you out of a sidestreet. People who don't drive don't realise London roads run on courtesy.
Handily the Waymos have those massive sensors, so all the Ubers, Cabbies and truck drivers whose jobs are under threat from the robots know to identify them and not to let them out of a side street. Justifiably so, as there's no driver to make eye contact with.
So you'll be sat in peaceful tranquility not having a driver to speak to while the human drivers backed up behind you honk more and more aggressively and your arrival time gets later and later.
Progress!
I’m one of the early Waymo supporters that’s signed up for the surveys. I can’t wait for the Waymos to arrive! Safer drivers and smoother driving experience. I cant wait for it, bring on the progress
I can imagine street piracy. Hold ups on self-driving taxis. Stand and deliver!
I wonder if the Uber drivers who are outraged at the prospect of technology reducing their worth were equally outraged on behalf of black cabbies ten years ago, who faced the prospect of technology reducing their worth by...Uber
Those Accident stats really need more digging into.
They say that they are 10x safer than a human driver. Is this measured in the US? A quick look on Wikipedia suggests that the road death rate in the US is 7x that of the UK by population, and about 2x by distance driven. (Similar ratios for other developed countries outside North America).
Hard to know the best comparison for city driving in somewhere like London - given distances will be relatively low and accident rate higher given density, I'd suggest it is closer to the population number.
That means the robots are about as safe as a human over here. Given the safety argument evaporates, and all of the other externalities, it is hard to see the benefit.