Should we be re-thinking this e-Bike thing?

[Kim]

Nor are there outcries for mobility scooter registration, etc.

There are occasional demands for some sort of training, usually after an inexperienced elderly user vs ankle incident.  On the odd occasion I've had reason to operate one, I've found that mobility scooters can be surprisingly unintuitive to steer, and I say that as someone who's probably had a lot more experience of random small vehicles than most newbie scooter users.

[citoyen]

I still go back to my original question though: is there any hard evidence to show how much of a genuine menace these things are? Or are we firmly in the realm of anecdata here?

The RNIB did some research on scooters, I think.  But a lot of their members would happily ban bicycles if given the chance.  The problem with visual impairment is that you're literally only aware of the law-breaking users.

The irony of your last line is not lost on me - the mother of an old girlfriend used to do a lot of work helping older people with visual impairments. According to her, many of them were still regularly driving cars on the road despite being officially registered blind and having had their licence taken away.

I don’t have any evidence to support this, I admit, but I strongly suspect there are far more KSIs caused on the road by visually impaired motorists than by cyclists. (And I don’t mean SMIDSYs, which are more down to inattention than eyesight problems.)

My own anecdata point is the time I had to jump out of the way to avoid being run down by a mobility scooter that was doing substantially more than 4mph on the pavement - and I’m pretty sure the problem was that the driver just couldn’t see where the fuck she was going.

Anyway, this is a distraction.

There may well be elements of the law that need clarifying/updating given the rapid advances in E-contraption technology in recent years, but I certainly don’t agree that licensing or chipping users is the answer - I mean, I’ve yet to hear a convincing case for it.

[Kim]

There may well be elements of the law that need clarifying/updating given the rapid advances in E-contraption technology in recent years, but I certainly don’t agree that licensing or chipping users is the answer - I mean, I’ve yet to hear a convincing case for it.

I can imagine it being effective in reducing the number of law-abiding users.  Much like the decline in motorcycling due to (amongst other things) systematic toughening-up of TEH RULEZ over the years, it becomes easier to use another form of transport, or at least practical to use a car for the same amount of legal hoop-jumping.

The cynic in me suggests that a helmet law would be a more cost-effective way to discourage e-contraption use.

[rogerzilla]

I nearly got mown down by a cnut on a fast e-scooter the other night.  Felt the wind as he passed.  Narrow pavement.

[Jaded]

How close to an RFID do you have to get?

That's where I'm not clear about the reading technology, but actually toll gates handle remote ID at speed so it should be possible.

Ah, I know this:

The RFID system we use for timing BHPC races has a day job tracking boxes in warehouses and vehicles through ports and similar.  Effective range is about 15 metres from a high-power antenna, and it will happily read at speeds in the 35mph Steve Slade going full pelt range.  Higher speeds would eventually limit the number of tags that could be simultaneously read, but would reduce the vehicle throughput to the point where that isn't a problem.

The problem that we have with HPV races is that the tag needs to be fixed to to a non-conductive surface.  If you stick them to metal or carbon fibre, it simply doesn't work, and the signal can't pass through metal or carbon fibre either.  You'd need a plastic part a good 20mm from any metal, which would be impractical on an ebike motor, or most cycle frames.  (We mostly stick them to cycle helmets, card/plastic number boards, glass fibre fairings and plastic windscreens, with the read aerials mounted horizontally at the side of the track.  For tracking cars they tend to be stuck to the windscreen or dangled from the rear-view mirror, and the aerials mounted overhead.)

And of course they could trivially be defeated by a bit of foil.

Thank you for this fact sense.

Much needed here...

[Kim]

I should add that it's also fairly expensive^[1], as you'd expect from a niche industrial product.  Cameras suitable for ANPR are probably cheaper, especially if you can stick them on a pole rather than erecting a gantry over the road.

(Reading tags from the side as we do for racing is a non-starter in mixed traffic, where cars could occlude the target.  We occasionally get missed reads from otherwise reliable tags because a carbon velomobile is passing in front of the antenna.)


_{[1] Well, the reader equipment is.  The tags themselves are cheap enough as to be considered disposable.  Good for applications where you're tracking a lot of objects in a small number of places.}

[geraldc]

If they want to stop the illegal ebikes, just make the delivery companies Deliveroo, just eat, stuart etc responsible for ensuring that their riders are riding road legal bikes.  The majority of illegal e bikes are doing food delivery.

Round here e bikes are used by parents taking their kids to school, or food delivery.

[Ham]

I should add that it's also fairly expensive^[1], as you'd expect from a niche industrial product.  Cameras suitable for ANPR are probably cheaper, especially if you can stick them on a pole rather than erecting a gantry over the road.

(Reading tags from the side as we do for racing is a non-starter in mixed traffic, where cars could occlude the target.  We occasionally get missed reads from otherwise reliable tags because a carbon velomobile is passing in front of the antenna.)


_{[1] Well, the reader equipment is.  The tags themselves are cheap enough as to be considered disposable.  Good for applications where you're tracking a lot of objects in a small number of places.}

There's clearly a difference between a toll application - where the owner of the tag wants it to be read - and a tracking application, where the reverse is likely to be true. But, I'm not convinced that the tech you are describing is the only way to do this. Chances are that an effective system will combine resilient technologies, much in the way Amazon Fresh combines camera, face recognition, AI and probably RFID, and it isn't difficult to see how any such system would have privacy concerns. But, if at the heart of it the tracked item is an NFT, maybe it is possible, especially if it is sold to people as a security feature.

[quixoticgeek]

. But, if at the heart of it the tracked item is an NFT, maybe it is possible, especially if it is sold to people as a security feature.

I see this thread is going for full buzzword compliance?

J

[Ham]

. But, if at the heart of it the tracked item is an NFT, maybe it is possible, especially if it is sold to people as a security feature.

I see this thread is going for full buzzword compliance?

J

If you like. Thing is, while I'm very firmly in the  cryptocurrency and NFT hype sceptic camp, I'm fascinated by blockchain technology and can see it's potential for real world use. That is, firmly and absolutely linking a <%thing> with a <%owner>, providing a level of trust that has never been possible to date. The fact that the primary use currently seems to be endless ponzi schemes or other means of extracting money from the gullible is irrelevant. Supply chain is the obvious application, and it is being used for that, but it doesn't make headlines.

There is an absolute need to link the driver of a car with the responsibility for what that car does, it is also the single most expensive item an individual buys and the potential to prove ownership would be an instant theft deterrent, it is difficult to see how it could be bypassed (aside from the thief selling the car in the Wild West)

[quixoticgeek]

If you like. Thing is, while I'm very firmly in the  cryptocurrency and NFT hype sceptic camp, I'm fascinated by blockchain technology and can see it's potential for real world use. That is, firmly and absolutely linking a <%thing> with a <%owner>, providing a level of trust that has never been possible to date. The fact that the primary use currently seems to be endless ponzi schemes or other means of extracting money from the gullible is irrelevant. Supply chain is the obvious application, and it is being used for that, but it doesn't make headlines.

Fair.

There is an absolute need to link the driver of a car with the responsibility for what that car does, it is also the single most expensive item an individual buys and the potential to prove ownership would be an instant theft deterrent, it is difficult to see how it could be bypassed (aside from the thief selling the car in the Wild West)

The flaw there is that I have driven dozens of vehicles in my life. I've never owned any of them. They've either been owned by my employer, a friend, or a partner.

The idea of One owner == one driver insurance model we have in the UK is relatively unusual. Lots of countries you insure the vehicle, not the driver. So anyone can drive it, with owners permission.

J

[Kim]

The flaw there is that I have driven dozens of vehicles in my life. I've never owned any of them. They've either been owned by my employer, a friend, or a partner.

The idea of One owner == one driver insurance model we have in the UK is relatively unusual. Lots of countries you insure the vehicle, not the driver. So anyone can drive it, with owners permission.

If you use a car-club car, hire bike, Voi scooter, or similar, the vehicle has a registration number for members of the public to trace.  The owner of the vehicle has a database which shows who was responsible for operating it at the time (be it by smartcard, app, signing a piece of paper, or whatever - the technology is irrelevant).  Which seems to work just fine for speeding offences, parking fines or whatever.

You can imagine automating this in a more universal way, so that Alice can allow Bob to use her vehicle in a way that creates a cryptographically secure audit trail.  You can imagine this involving lots of Blockchain, because blockchain is cool, or being functionally deficient, because it's a government IT project.  It might end up as a high-tech way for Alice to claim that Carol was the one who was driving, so that Bob doesn't get too many points on his licence.  It will, undoubtedly, be late and over-budget.

[quixoticgeek]

Maybe if we just got rid of the cars, so the electric bikes had more space, and weren't sharing with pedestrians, it might solve a lot of these problems?

J

[Cudzoziemiec]

Maybe if we just got rid of the cars, so the electric bikes had more space, and weren't sharing with pedestrians, it might solve a lot of these problems?

J

Words of truth.

[DuncanM]

Blockchain is fundamentally just a different sort of data storing process. The DVLA might be rubbish, but blockchain won't solve that, because the data in the system is the crucial part, and that needs users to fill in the forms correctly. Whether it's some magic beans based distributed ledger system, or a mainframe running Oracle, the tech is an irrelevance to the problem at hand.

Private e-scooters are illegal right now. They are ridden all over the country by all sorts of people for all sorts of purposes. It's pretty straightforward to see which are illegal and which aren't, but it's way down the list of priorities for the police (and for very good reason). Creating new laws, which make some legal and some illegal is going to make it much harder to police, and anyone who wants to break them will be able to do so with ease and impunity.

[Polar Bear]

Maybe if we just got rid of the cars, so the electric bikes had more space, and weren't sharing with pedestrians, it might solve a lot of these problems?

J

Quite.

In the middle of Rugby is a pedestrian zone.  Pick any day and time you like and there will be cars and delivery vans, e-scooters, mobility scooters and the occasional bike or e-bike for pedestrians to vie for priority with.

It's totally out of control.

[Ham]

Blockchain is fundamentally just a different sort of data storing process.

In the sense that it is data that is stored, you could be considered correct, but I'm sorry, that's as accurate and helpful as saying a swiss watch and a horseshoe are the same as they are both made of iron.

Blockchain and its distributed ledger is a way of ensuring the honesty of the data you are looking at, be it a series of numbers or NFT or cryptocurrency. With today's technology (and anything in our forward view), there is no way of faking it. That's game changing. Nothing in human history has been incapable of being forged. That cryptocurrency has jumped on board and made a nonsense out of the virtues of blockchain is almost irrelevant.

[Cudzoziemiec]

But the data is only "honest" if it was input by honest people. If someone buys a bike off Amazon, Ali-Express or Halfords and then doesn't register it, or registers it under a false name and address, the data might remain uncorrupted and traceable to a point of sale but it's not "honest".

[quixoticgeek]

In the sense that it is data that is stored, you could be considered correct, but I'm sorry, that's as accurate and helpful as saying a swiss watch and a horseshoe are the same as they are both made of iron.

Blockchain and its distributed ledger is a way of ensuring the honesty of the data you are looking at, be it a series of numbers or NFT or cryptocurrency. With today's technology (and anything in our forward view), there is no way of faking it. That's game changing. Nothing in human history has been incapable of being forged. That cryptocurrency has jumped on board and made a nonsense out of the virtues of blockchain is almost irrelevant.

Only upto a point. If you have a block chain that forks and now you have two chains both claiming to be The One True Record™, and both holding different state. Then it doesn't achieve anything.

Ultimately for a lot of the tasks people think a block chain can solve, you can achieve the same thing by just using cryptographically signed individual records. With the associated web of trust.

J

[DuncanM]

Blockchain is fundamentally just a different sort of data storing process.

In the sense that it is data that is stored, you could be considered correct, but I'm sorry, that's as accurate and helpful as saying a swiss watch and a horseshoe are the same as they are both made of iron.

Blockchain and its distributed ledger is a way of ensuring the honesty of the data you are looking at, be it a series of numbers or NFT or cryptocurrency. With today's technology (and anything in our forward view), there is no way of faking it. That's game changing. Nothing in human history has been incapable of being forged. That cryptocurrency has jumped on board and made a nonsense out of the virtues of blockchain is almost irrelevant.

To a certain extent that's correct, and it's super useful in applications such as anti money-laundering and anti tax-evasion, but it doesn't matter in this application. To be sure of who is in charge of a certain vehicle, at a certain point in time, then (absent tagging both vehicles and people) you have to trust the starting point and have a reliable chain of custody. One of the main problems with the DVLA database is that it thinks that vehicleA is owned by person1 (who lives in LocationX), and one or more of those things is no longer true.  Blockchain doesn't fix that.
Garbage in, garbage out.

[grams]

The thing that makes blockchain records unforgeable is public key signing, same as your boring chip and pin card, same as the HTTPS connection you’re viewing this on now. The blockchain is an irrelevance.

[Cudzoziemiec]

However the technology works, the basic problem with this idea is that, like so many ideas which claim or intend to improve safety etc for cycling, it does so by putting additional difficulties in cyclists' way.

[quixoticgeek]

However the technology works, the basic problem with this idea is that, like so many ideas which claim or intend to improve safety etc for cycling, it does so by putting additional difficulties in cyclists' way.

And does absolutely nothing to improve safety for cyclists and vulnerable road users...

J

[Ham]

The thing that makes blockchain records unforgeable is public key signing, same as your boring chip and pin card, same as the HTTPS connection you’re viewing this on now. The blockchain is an irrelevance.

There is a huge difference between public key technology which,  while secure in its essence, is wide open to fraud: stealing your chip and pin, hijacking an https session etc etc etc and blockchain which uniquely proves ownership in its shared open ledger. Crucially, it is not a single key pair, but a distributed sequence which defeats forgery under current, and current view of future, technology. Blockchain gives you a token. Either a fungible (eg bitcoin, every bitcoin is the same as the other)  where holding it is proof of ownership or non-fungible "NFT" where each token is different and crucially links to ownership. The transaction is indeed validated by a private/public key pair (as established, this is secure in itself) but the underlying ownership is the pointer to the blockchain ledger, unique and unforgeable.

Given the value of cars, the amount of technology inbuilt into new ones and their propensity for being stolen, the application of blockchain technology to supplement and replace the current paper/number plate system is an obvious move that would be likely be welcomed by owners, I was floating the concept that the other benefit could be that owners could also be held to account, which would likely be less popular.

Of course any system with meatware is subject to exploitation, and people will be fraudulently tricked out of ownership I am sure, but the big difference is that just stealing the item will no longer be sufficient. Once you have the system established, its application is much wider to anything of value, especially with inbuilt technology and that brings into question the problems that exist with any monoculture, but that's a can we can kick for the moment.

[Paul H]

The EU regulation is specifically intended to to classify these pedelecs the same as non assisted cycles. 
The choices are:
a) Keep those regulations
b) Introduce new regulations that apply to all cycles.
c) Create a new class of transport with it's own regulation.

I'd choose a. It'd be hard to justly a 250W, 25kph cycle needing something that didn't apply to all other cycles.  I'd like to see some enforcement, maybe something could be done at the point of sale. It wouldn't be my priority for use of police time.
It might also help if the UK didn't make running a Speed Pedelec so costly and complicated.  The EU created this class but left the implementation to member states and the UK chose the easiest option and lumped them in with mopeds.

I don't know what the danger this thread intends to prevent is, are there any figures or is it just the perception?  I know that when walking a shared use path, being close passed by a cyclist feels just as unpleasant as being close passed by a speedy illegal E-bike while cycling.  The former happens a lot more regularly than the latter, often by people who look like they ought to know better.