Leader of Ealing Council says “Segregated cycle lanes are the answer”

BehindTheBikesheds wrote:

If segregation is the answer to all our problems how come the Dutch with their wonderful infra still manage to have circa 180 deaths annually? I get that they cycle a heck of a lot more miles but if the vast majority of that is on protected/segregated infra with specific light phases for people on bikes and the pro cycling thinking from governemnt etc how on earth do they get so many deaths?

They cycle a _lot_ more miles.  And, furthermore, the demographic of cyclists is very different.  Some of those deaths are of older people who could just as easily have died from a fall while walking (I think some are even heart-attacks that occur while on a bike).

Besides, I'm quite ready to agree that even the Dutch don't go far enough - they still have a fair bit of bad infrastructure, and personally I'd tax motoring far more and introduce more road tolls for motor vehicles, as well as creating still more dedicated cycle infrastructure but starting from where we are one has to at least be somewhat realistic.  If we were starting from where the Dutch are it might be plausible to be more ambitious.

BehindTheBikesheds wrote:

the answer is (in part) that those in cars are still not as good as people think they are, there are still many motorvehicle/cyclist interactions that are crossing points and people on bikes being on roads that simply do not have infra, so the point about driver education and restriction of motorists from areas where you are going to get (hopefully going to get) lots of cycling is massively important.

Personally I think far too much faith is put in 'education' as a cure for all ills.  I don't see much evidence that it works (for example, some I know think it will solve all problems of social mobility or lack of jobs - as if!).  I reckon this is because those proposing it as a solution are people who themselves have a lot of education and tend to overvalue it.

And rigorous legal enforcement is a fine idea in principle, but it's very, very expensive and realistically is never going to happen - socieity won't pay for that many police officers permanently monitoring the roads.

Restriction of motorists from certain areas is precisely what segregated infrastructure is all about, surely?

BehindTheBikesheds wrote:

In this country driver education is working better than cycling infra IMHO

Not sure where you get that from.  IMHO it isn't working at all.

BehindTheBikesheds wrote:

To think that infra as per the Dutch will happen in the next 20 years or even 40 years is utterly bonkers, it's a complete pipedream, it simply will not happen. it didn't happen in the 20s/30s when roads were being laid except for some parts of london and it was bloody awful so the cyclists of the day didn't want to use it and it was sparse and only in one big city (ring any bells).

To think that education is going to solve the problem is what is truly bonkers and a pipe-dream though.

BehindTheBikesheds wrote:

If we focus on restricting motorvehicle use in towns and cities, completely take back whole lengths of the highway as well as enforcing the law plus the most important bit of all driver education we can change things.

But that _is_ segregation, so I'm not sure what you are arguing about.  Except for 'driver education' which is mostly a waste of time.  And the only affordable way to enforce the law is to change the physical environment to make it physically difficult to break the law (i.e. infrastructure).  Doing it with cops on the street is a pipe-dream.

The only form of 'education' that will really help is that of drivers getting used to cyclists around them and getting used to cycling themselves.  And the best way to achieve that is by building more infrastructure so more people decide to cycle.

Rich_cb wrote:

Driverless cars will be with us long before any meaningful change in driver behaviour can be achieved.

They really won't.

Will we have cars made by several companies capable of navigating themselves perfectly autonomously? Of course - very soon.

But them becoming a reality on the roads is way off.

For that to happen, cars have to talk to each other, so we need agreed standards and platforms. That means the likes of VAG, Ford, Mercedes, Toyota, Tata, Nissan, BMW, Porsche, GM, Google, Apple, Dyson, Facebook, Tesla etc etc are yet to fight the Mother of all Standards Wars. It'll make Betamax vs VHS look quick and easy.

And then there's letting them loose on the streets. Do you blend them in with humans? You can't - either they're programmed to defer to humans or unpredictability, and then human drivers, cyclists and peds will just take the piss and cut them up, or they can only interact with each other and will need entirely separate infrastructure (so they take over the roads).

And what the carmakers will want as standard behaviour/infrastructure, they'll want replicated across all markets. That means that what goes in the US will have to go in Germany will have to go in China...

Technologically, it's a piece of piss.

Politically, decades away.

davel wrote:

They really won't.

Will we have cars made by several companies capable of navigating themselves perfectly autonomously? Of course - very soon.

But them becoming a reality on the roads is way off.

For that to happen, cars have to talk to each other, so we need agreed standards and platforms. That means the likes of VAG, Ford, Mercedes, Toyota, Tata, Nissan, BMW, Porsche, GM, Google, Apple, Dyson, Facebook, Tesla etc etc are yet to fight the Mother of all Standards Wars. It'll make Betamax vs VHS look quick and easy.

And then there's letting them loose on the streets. Do you blend them in with humans? You can't - either they're programmed to defer to humans or unpredictability, and then human drivers, cyclists and peds will just take the piss and cut them up, or they can only interact with each other and will need entirely separate infrastructure (so they take over the roads).

And what the carmakers will want as standard behaviour/infrastructure, they'll want replicated across all markets. That means that what goes in the US will have to go in Germany will have to go in China...

Technologically, it's a piece of piss.

Politically, decades away.

You're completely wrong.

They're already driving on public roads.

https://youtu.be/aaOB-ErYq6Y

The economic benefits of self driving technology are so huge that they will be commonplace far far sooner than you think.

Rich_cb wrote:

davel wrote:

They really won't.

Will we have cars made by several companies capable of navigating themselves perfectly autonomously? Of course - very soon.

But them becoming a reality on the roads is way off.

For that to happen, cars have to talk to each other, so we need agreed standards and platforms. That means the likes of VAG, Ford, Mercedes, Toyota, Tata, Nissan, BMW, Porsche, GM, Google, Apple, Dyson, Facebook, Tesla etc etc are yet to fight the Mother of all Standards Wars. It'll make Betamax vs VHS look quick and easy.

And then there's letting them loose on the streets. Do you blend them in with humans? You can't - either they're programmed to defer to humans or unpredictability, and then human drivers, cyclists and peds will just take the piss and cut them up, or they can only interact with each other and will need entirely separate infrastructure (so they take over the roads).

And what the carmakers will want as standard behaviour/infrastructure, they'll want replicated across all markets. That means that what goes in the US will have to go in Germany will have to go in China...

Technologically, it's a piece of piss.

Politically, decades away.

You're completely wrong.

They're already driving on public roads.

https://youtu.be/aaOB-ErYq6Y

The economic benefits of self driving technology are so huge that they will be commonplace far far sooner than you think.

The odd one, probably with a driver ready to step in, absolutely. But that video didn't look like London during a rainy winter evening, did it? And it definitely wasn't Mumbai during a religious holiday. What works in Silicon Valley in a quiet afternoon is not typical of what's needed across swathes of the world that will define the market. It didn't have to negotiate a deliveroo rider or tuktuk or ped listening to music while reading Facebook.

But that wasn't your original point anyway - your implication was critical mass, that driverless cars will step in and make the debates about human driver behaviour moot. That's decades away.

Honestly, this isn't pessimism or alarmism - just realism based on an understanding on the different levels of automation. We're currently at around level 1: we need to be 4 or 5 (what seems to be the waymo car. I'm skeptical about that, but as I said it isn't the technology that will be the main problem) or driverless cars to be a reality, and the implementation is even more difficult.

It takes a decade to get a few miles of decent bike infrastructure in London or a new bit of railway track laid. Let's revisit this thread in 30 years.

davel wrote:

The odd one, probably with a driver ready to step in, absolutely. But that video didn't look like London during a rainy winter evening, did it? And it definitely wasn't Mumbai during a religious holiday. What works in Silicon Valley in a quiet afternoon is not typical of what's needed across swathes of the world that will define the market. It didn't have to negotiate a deliveroo rider or tuktuk or ped listening to music while reading Facebook.

But that wasn't your original point anyway - your implication was critical mass, that driverless cars will step in and make the debates about human driver behaviour moot. That's decades away.

Honestly, this isn't pessimism or alarmism - just realism based on an understanding on the different levels of automation. We're currently at around level 1: we need to be 4 or 5 (what seems to be the waymo car. I'm skeptical about that, but as I said it isn't the technology that will be the main problem) or driverless cars to be a reality, and the implementation is even more difficult.

It takes a decade to get a few miles of decent bike infrastructure in London or a new bit of railway track laid. Let's revisit this thread in 30 years.

The cars in the video don't need a backup driver.

Right now there are fully driverless cars carrying passengers on public roads.

They are therefore, by definition, level 5.
They do not require any special infrastructure.

The company featured in the video was created in 2009.

If you think it's going to take decades to get from that video to cars in all cities I think you fail to understand the speed of progress in the field.

My point was that changing driver behaviour will take multiple decades. Driverless cars will be commonplace, if not ubiquitous, within 1 decade.

Rich_cb wrote:

My point was that changing driver behaviour will take multiple decades. Driverless cars will be commonplace, if not ubiquitous, within 1 decade.

So the technology's there to scoot round California streets in the afternoon. Slightly different AI level need from that to drive across London at 5.15 on a Thursday, no?

I doubt it will reacht that AI level for a longtime - but as you say, I could be massively underestimating it. Fundamentally, that isn't what I think the delay will be down to. The implementation... think it through. Do you just think the volumes of driverless cars will increase as people replace their current vehicle?

davel wrote:

So the technology's there to scoot round California streets in the afternoon. Slightly different AI level need from that to drive across London at 5.15 on a Thursday, no?

I doubt it will reacht that AI level for a longtime - but as you say, I could be massively underestimating it. Fundamentally, that isn't what I think the delay will be down to. The implementation... think it through. Do you just think the volumes of driverless cars will increase as people replace their current vehicle?

Not a huge difference no.

We've gone from zero to fully autonomous cars on public roads in 8 years.

Do you really think it will take another 30 years to cope with busy city traffic?

I think private car ownership will collapse as driverless cars become more common, owning a private car in a city will become incredibly rare.

Each driverless car will replace at least 5 private cars.

Rich_cb wrote:

We've gone from zero to fully autonomous cars on public roads in 8 years. Do you really think it will take another 30 years to cope with busy city traffic?

I think if 8 years is zero to driverless in Caluifornia on a Sunday afternoon, then yes - to continue that trajectory to cope with London, Mumbai, Beijing, Sao Paulo... probably another 30 years.

But I keep saying, that's the easy bit. I don't think that's the route it will go, because I think it will be next to impossible. As soon as a driverless car hits a pedestrian (and they will, if they're implemented in that manner) there will be backlash by media and poltiicians and people just not ready for it.

The alternative is to go 'big bang' and move all the roads over to an agreed platform whereby all cars communicate with each other so they all know where the others are. That requires less AI, but standards that, again, won't be agreed in 30 years.

Then there's the updating of regulation to cover responsibility and what 'driverless' actually means. The manufacturers will want something fairly standard (as it is, largely, with current driving laws) for them to be able to launch everywhere and be economically viable. So, Europe, the US and China has to see these cars in the same manner and pass similar regulation.

Decades.

davel wrote:

I think if 8 years is zero to driverless in Caluifornia on a Sunday afternoon, then yes - to continue that trajectory to cope with London, Mumbai, Beijing, Sao Paulo... probably another 30 years.

But I keep saying, that's the easy bit. I don't think that's the route it will go, because I think it will be next to impossible. As soon as a driverless car hits a pedestrian (and they will, if they're implemented in that manner) there will be backlash by media and poltiicians and people just not ready for it.

The alternative is to go 'big bang' and move all the roads over to an agreed platform whereby all cars communicate with each other so they all know where the others are. That requires less AI, but standards that, again, won't be agreed in 30 years.

Then there's the updating of regulation to cover responsibility and what 'driverless' actually means. The manufacturers will want something fairly standard (as it is, largely, with current driving laws) for them to be able to launch everywhere and be economically viable. So, Europe, the US and China has to see these cars in the same manner and pass similar regulation.

Decades.

Well we'll have to wait and see.

I think 80-90% ofthe work is already done for major city driving in developed countries.

Driverless cars will be massively safer than normal cars so whilst any fatality will garner significant media attention it won't halt the change.

The ridiculous sums of money to be made by both companies and countries will ensure legislation is implemented swiftly.

Rich_cb wrote:

davel wrote:

I think if 8 years is zero to driverless in Caluifornia on a Sunday afternoon, then yes - to continue that trajectory to cope with London, Mumbai, Beijing, Sao Paulo... probably another 30 years.

But I keep saying, that's the easy bit. I don't think that's the route it will go, because I think it will be next to impossible. As soon as a driverless car hits a pedestrian (and they will, if they're implemented in that manner) there will be backlash by media and poltiicians and people just not ready for it.

The alternative is to go 'big bang' and move all the roads over to an agreed platform whereby all cars communicate with each other so they all know where the others are. That requires less AI, but standards that, again, won't be agreed in 30 years.

Then there's the updating of regulation to cover responsibility and what 'driverless' actually means. The manufacturers will want something fairly standard (as it is, largely, with current driving laws) for them to be able to launch everywhere and be economically viable. So, Europe, the US and China has to see these cars in the same manner and pass similar regulation.

Decades.

Well we'll have to wait and see. I think 80-90% ofthe work is already done for major city driving in developed countries. Driverless cars will be massively safer than normal cars so whilst any fatality will garner significant media attention it won't halt the change. The ridiculous sums of money to be made by both companies and countries will ensure legislation is implemented swiftly.

I spent a couple of days working with a company called UK Autodrive last week - they're currently testing self driving Range Rovers and Tatas on the streets of Coventry. They're not driving on quiet backstreets but a circuit of busy roads around the university, which have lots of pedestrians and junctions with no priority, which the cars navigate impressively well. I chatted to a number of the engineers involved and the consensus opinion was that this technology will be availble with five years and commonplace within ten.

Rich_cb wrote:

davel wrote:

So the technology's there to scoot round California streets in the afternoon. Slightly different AI level need from that to drive across London at 5.15 on a Thursday, no?

I doubt it will reacht that AI level for a longtime - but as you say, I could be massively underestimating it. Fundamentally, that isn't what I think the delay will be down to. The implementation... think it through. Do you just think the volumes of driverless cars will increase as people replace their current vehicle?

Not a huge difference no. We've gone from zero to fully autonomous cars on public roads in 8 years. Do you really think it will take another 30 years to cope with busy city traffic? I think private car ownership will collapse as driverless cars become more common, owning a private car in a city will become incredibly rare. Each driverless car will replace at least 5 private cars.

But all of what you say here seems very unlikely to me.  So far autonomous cars have gone from zero to negligable.  They don't seem remotely ready for mainstream use.  Personally I think they are overhyped techno-evangelism, and if they ever do break through it will probably be for the worse anyway.

FluffyKittenofTindalos wrote:

But all of what you say here seems very unlikely to me.  So far autonomous cars have gone from zero to negligable.  They don't seem remotely ready for mainstream use.  Personally I think they are overhyped techno-evangelism, and if they ever do break through it will probably be for the worse anyway.

Did you watch the video?

You think a car driving completely autonomously in a suburban environment is negligible progress?

If they can do that they can do motorways with ease and major cities with a bit more development.

I don't really see how they can make things worse.

Far less driver error, less congestion and less pollution all seem like huge benefits for cyclists and broader society.

Rich_cb wrote:

FluffyKittenofTindalos wrote:

But all of what you say here seems very unlikely to me.  So far autonomous cars have gone from zero to negligable.  They don't seem remotely ready for mainstream use.  Personally I think they are overhyped techno-evangelism, and if they ever do break through it will probably be for the worse anyway.

Did you watch the video?

You think a car driving completely autonomously in a suburban environment is negligible progress?

If they can do that they can do motorways with ease and major cities with a bit more development.

I don't really see how they can make things worse.

Far less driver error, less congestion and less pollution all seem like huge benefits for cyclists and broader society.

It's whether they will ever really be able to handle London deliveroos and couriers... If they can't, what's the logical step? Give roads over to cars completely: make them cyclist-free and ped-free.

Given the state of US roads anyway, the scarcity of cyclists, the jaywalking laws, the AI leap from being able to judge other cars on big roads, to then being assertive but safe with unpredictable peds and cyclists, it might be the next logical step regardless.

I think that was FKoT's point: cyclists and peds have already lost a few battles in road ownership; given the lobbying, propaganda and absurd logic that went into jaywalking laws being passed, it's not a huge leap of imagination to see the roads being dedicated to autonomous cars completely. That'd be a really shit outcome.

davel wrote:

It's whether they will ever really be able to handle London deliveroos and couriers... If they can't, what's the logical step? Give roads over to cars completely: make them cyclist-free and ped-free.

Given the state of US roads anyway, the scarcity of cyclists, the jaywalking laws, the AI leap from being able to judge other cars on big roads, to then being assertive but safe with unpredictable peds and cyclists, it might be the next logical step regardless.

I think that was FKoT's point: cyclists and peds have already lost a few battles in road ownership; given the lobbying, propaganda and absurd logic that went into jaywalking laws being passed, it's not a huge leap of imagination to see the roads being dedicated to autonomous cars completely. That'd be a really shit outcome.

I honestly think they'll be able to manage any city in the UK within a few years.

Once they are established there won't be any need for deliveroo cyclists or couriers.

Driverless cars will eliminate most on street parking so there will be swathes of currently wasted road that will become available for cyclist and pedestrians.

The difference will be transformative for urban living.

HarrogateSpa wrote:

Segregated cycle lanes in towns and cities are a large part of the answer, to encourage utility cycling - as successful European cycling nations demonstrate.

Not the answer. Segregated cycle lanes are only going to give numpty car drivers another excuse to drive badly/agressively if you are not using them.

Is Germany a successful European cycling nation? I hope not because it really doesn't feel like it when you are often required by law to use really crappy bits of pavement or potholed gravel track because it has been designated "cycling infrastructure". Or when a reasonable stretch of cycle path abruptly ends or continues on the opposite side of a busy road.

Cycling infrastructure is just a ploy by the pro-car lobby to get us cyclists off the road (I admit I could be a teeny-weeny bit paranoid!):

Jem PT wrote:

Agree. Segregated cycle lanes are not The Answer. They can help in some circumstances/situations but can never replace education of drivers to drive better and be more considerate towards more vulnerable users. Introducing presumed liability (on drivers) in the event of an 'accident' would help a lot, be cheaper and quicker than education, but would still not on its own be The Answer. 

Presumed liability will just mean more dashcams in cars. This will result in more cyclists ending up like the cyclist sued by the taxi company. 

Bluebug wrote:

Jem PT wrote:

Agree. Segregated cycle lanes are not The Answer. They can help in some circumstances/situations but can never replace education of drivers to drive better and be more considerate towards more vulnerable users. Introducing presumed liability (on drivers) in the event of an 'accident' would help a lot, be cheaper and quicker than education, but would still not on its own be The Answer. 

Presumed liability will just mean more dashcams in cars. This will result in more cyclists ending up like the cyclist sued by the taxi company. 

Like it does in every other country in Europe?

Except of course Cyprus, Malta, Ireland and Romania - exalted company indeed.

davel wrote:

Bluebug wrote:

Jem PT wrote:

Agree. Segregated cycle lanes are not The Answer. They can help in some circumstances/situations but can never replace education of drivers to drive better and be more considerate towards more vulnerable users. Introducing presumed liability (on drivers) in the event of an 'accident' would help a lot, be cheaper and quicker than education, but would still not on its own be The Answer. 

Presumed liability will just mean more dashcams in cars. This will result in more cyclists ending up like the cyclist sued by the taxi company. 

Like it does in every other country in Europe? Except of course Cyprus, Malta, Ireland and Romania - exalted company indeed.

I was actually thinking of Russia where drivers have dashcams to protect themselves primarily from other drivers but also from pedestrians who want a pay off.  

Insurers are businesses so if they have a chance of getting dashcam footage to sue a cyclist to recover their costs they will.

Bluebug wrote:

Jem PT wrote:

Agree. Segregated cycle lanes are not The Answer. They can help in some circumstances/situations but can never replace education of drivers to drive better and be more considerate towards more vulnerable users. Introducing presumed liability (on drivers) in the event of an 'accident' would help a lot, be cheaper and quicker than education, but would still not on its own be The Answer. 

Presumed liability will just mean more dashcams in cars. This will result in more cyclists ending up like the cyclist sued by the taxi company. 

As it should be if the cyclist is at fault - the whole point of presumed liability. The cyclist sued by the taxi company was at fault (his words) therefore he needs to pay damages.

LastBoyScout wrote:

Segregated cycles lanes are not the answer - being given the space and consideration that cyclists are legally entitled to is.

..and while we're waiting for Utopia to arrive perhaps some decent segregated infrastructure may help more people out of their cars and onto their bikes ? We'd ideally have both but one of them can be directly under our control, now, and the other... not so much.

fukawitribe wrote:

LastBoyScout wrote:

Segregated cycles lanes are not the answer - being given the space and consideration that cyclists are legally entitled to is.

..and while we're waiting for Utopia to arrive perhaps some decent segregated infrastructure may help more people out of their cars and onto their bikes ? We'd ideally have both but one of them can be directly under our control, now, and the other... not so much.

I agree that both should be the aim.

However, given the effort in political will and campaigning necessary to get decent infrastructure built, I wonder whether similar effort could result in something meaningful for respect on the roads.

Something like presumed liability.