Wouldn't presence of the key address most of this?

Sure, they can steal the car AND the key, but more often, it's the car alone.

This is something you add support for as an optional convenience feature because you can do it entirely in software, not something you require people to use. You still need a card reader on the charger.

This is the fundamental problem with all of this stuff -- nobody wants your app. If you can't make it better than sticking a card in the machine, go away. If you can, prove it, by making it an alternative to sticking a card in the machine instead of the only option.

Can't read. Does it say USB-C? ;)

Another reason for a card reader is that it easily handles the case of a borrowed EV.

It is customary when borrowing someone's ICE car to pay for the gas used by returning the car with as much gas as it had when you borrowed it, or even with a full tank. I'd expect people will want to continue that custom when the car is an EV.

EVs could potentially handle that scenario even better than how it works with ICE cars. With ICE there’s always the error associated with how much gas was in the car to start as well as how much should you fill up when returning it. (Yes I know it’s customary to always just return a friend’s car filled to 100% regardless of the original tank state).

Nonetheless with an EV you could just hit a button saying “friend has taken possession” or “loaner mode” similar to valet mode. Upon returning the car and disabling that mode the car could present you with an accurate dollar amount of exactly what to pay your friend.

Particularly in a world where every new phone and smartwatch integrates tap-to-pay, seems like a no-brainer to just integrate at that point.

Cars don’t need to have credit card details onboard.

Hypothetical standard: Charger lets car open a low bandwidth internet connection. Car uses it (or cellular) to open an encrypted channel to a 3rd part service (Tesla/Visa/whatever) saying it’s VIN, password, and the charger’s ID. Charger gets payment from 3rd party.

PS: I would rather chargers just had working credit card readers, but individual chargers don’t do many transactions per day so apps are a cost saving thing.

You're getting at the root of the problem here, which is that Visa doesn't actually support anything like that and "whatever" is going to add another grubby middle man to the transaction, which is the thing that most needs to be prevented.

You're fundamentally trying to transfer your money to the operator of the charger, but they aren't standing around to give cash to, so you want something digital. Alright, so your bank should do this -- you authorize your car's private key to make withdrawals from your bank account at charging points and we're all set. Except that banks suck at this sort of thing, which is why we have the credit/debit network to create some kind of standard. Except the standard is credit cards, not public key cryptography.

You wanted to give your money to the charging point operator but instead you have to give it to your bank to give it to Visa to give it to the charging point operator. Now we're going to add "Tesla/whatever" to the chain? This needs to stop. You stick with the card reader until you figure out how to get banks, using some kind of standard, to support the public/private key thing and we can remove a middle man instead of adding one.

Visa et al are going to charge a minimum transaction fee for small-value transactions, which covers the typical charging session. Third party services are an advantage, because they can bundle small transactions together and take advantage of their economies of scale and their own fraud models. There is a world where Visa and the payment card industry looked forward in the year 2005 and said “IoT devices like EVs will need payment capability, let’s think about how to be essential to this ecosystem in a way that removes the need for third party services and saves everyone money” but that kind of technical innovation seems to be outside of their core competency.

The transactions are of higher value than many public transport tickets, and those are already common — either as normal purchases from a ticket machine, or with the special transaction type used when the payment card is touched to the reader on the ticket barrier.

Following a recent trip I had a bunch of ~1€ transactions from buying single bus/metro tickets from machines.

Visa charges a 5-10 cent fixed fee plus a percentage for those 1 Euro transactions. Presumably public transport just eats those costs, but in a commodity business like electricity I imagine it’s a noticeable chunk of the profit margin.

1 cent fixed fee + percentage seems to be the standard where I am. It might be difficult in other EU countries but IIRC the fixed fee is capped at 5cents.

The only public information I can find is this, which is from 2013 in London:

Visa credit 0.65%, debit 1 pence. Mastercard credit 0.80%, debit 1 pence + 0.5%.

https://www.whatdotheyknow.com/request/costs_to_tfl_for_cont...

> of ~1€ transactions

In Europe, sure because the fees are capped. In the US you pay a fixed fee + x% which makes small transactions like this prohibitively expensive

I agree, except Visa and Mastercard are willing to innovate. Apple Pay/Google Wallet for cars isn’t an unrealistic goal.

All the car needs to store is some private key corresponding to yourself, to prove it's that car. On modern hardware you can make these really hard to steal, short of taking the entire chip with you.

Worse case, if someone steals that key (or the hardware that stores it), they can steal electricity with it. Which also means they'll be caught approximately tomorrow.

If you have that level of access/capability, there's better things to do than using it to get 50 dollars worth of electricity and probably being picked up by police as you siphon it.

Besides, there's easier and stealthier ways to get free electricity.