Timothy B. Lee here. You are correct, except that I think it's a mistake to treat ADAS systems as "self-driving cars." I'm bullish on true self-driving car projects like Waymo and Cruise. I think ADAS systems have a lot of problems that have become more clear to me as I've covered them in more depth.
I think that full autonomy (level 4 and 5) is an AI-complete problem and a true solution is many, many years away. On the other hand, I'm not an expert (I'm doing an AI PhD but on a different subject) so I may be overestimating the difficulties involved.
One problem with reporting on autonomous driving is that a lot of the technology is proprietary and the research takes place behind closed doors, so it's very hard to understand exactly where the state of the art is. We're left with the announcements from the companies who actually sell it, that will inevitably tend to be overinflated.
As the technology becomes more common, I guess we'll all end up adjusting our expectations, one way or another. I'm looking forward to your future articles :)
More relevant in this situation: Mobileye's stated reason for cutting those ties.
>On Wednesday, Mobileye revealed that it ended its relationship with Tesla because "it was pushing the envelope in terms of safety." Mobileye's CTO and co-founder Amnon Shashua told Reuters that the electric vehicle maker was using his company's machine vision sensor system in applications for which it had not been designed.
"No matter how you spin it, (Autopilot) is not designed for that. It is a driver assistance system and not a driverless system," Shashua said.
> Mobileye's CTO and co-founder Amnon Shashua told Reuters that the electric vehicle maker was using his company's machine vision sensor system in applications for which it had not been designed.
Props to Sashua for deciding to put safety above profit. We need people in leadership like this.
The main point is that Waymo had a 4-6 year head start over virtually everyone else in this industry. So it's not surprising that their technology seems significantly better! Obviously we don't yet know how much better, or if it's good enough to improve on the average human driver. But they seem to be pretty close.
The downside to their head start is that they had to build their system from scratch, and there was a great deal of trial and error.
The upside is that they had the very best, most passionate and talented roboticists in the field working on it, as nobody else was hiring at the time. For every company entering late in the game, securing top tier talent is the biggest hurdle.
Not only the head start but they weren't a business, they did deep research for the sake of it. Not to sell it right away. Many said that mobileye and others were very crude systems. I think the idea was that having real mileage in Teslas and similar experience would accelerate improvement .. so far it didn't help AFAIK<
The Jaguar I-PACE deal envisioned designing a custom version of the I-PACE so that the sensors and computing hardware could be installed at the factory. I haven't seen specifics on the Pacifica deal but it seems likely they'd take the same approach.
For me, Star Trek future is mix of politics, technology and physics.
In terms of politics, I see just deterioration in my region. We get technology we want (smartphones), but not what we need (carbon sequestration, better food sources). Our current tech is ruining the planet.
And as a developer, I see a lot of new and different, but little better. But that might be also connected to my world view already.
That's the point: I'm pretty sure that somebody will create automated driving technology that's better than humans within the next decade or so. But it won't necessarily be Tesla. If I had to guess I'd say Waymo is likely to get there first. And as I say in the piece, there's good reason to think that Tesla's incremental approach—start with ADAS, evolve into full self-driving—may be an evolutionary dead end because humans are really bad at supervising a driving technology that works 99.99 percent of the time but gets in a fatal crash the other 0.01 percent.
What about Tesla's choice to design its technology around cameras rather than lidar? They seem to be the only ones going that route. I've seen quotes from a number of people who build this kind of technology who say that's a fundamentally less-safe way to build self-driving vehicles. Did you consider addressing that point in the article?
Maybe you should quote the entire paragraph instead of truncating it in a way that makes it look ridiculous. As I said in the next sentence, "Musk is persistent and a quick learner."
And then as I write later in the piece:
"Musk ignored the conventional wisdom, and he has gotten much further than anyone expected. He has sold hundreds of thousands of cars and has hundreds of thousands more people eager to buy the Model 3 as soon as it's available. Moreover, Tesla has had a huge influence on the broader car industry, forcing every major carmaker to take battery electric vehicles seriously."
I didn't think the rest of the paragraph helped your cause in any way. You appear to admit he fails at his objectives, but threw in an unsubstantiated claim that he's a quick learner. If he was a quick learner, he would have learned he makes outrageous claims, and tempered his statements.
Lightning's core innovation is the use of the Bitcoin blockchain as a cryptographic backstop for payment channels. If the other party in a payment channel stops cooperating, you can broadcast the current commitment transaction to the blockchain, which effectively refunds the current balance back to each party. There's a similar mechanism for enforcing the hashed time lock contracts that make Lightning payment chains possible. I don't know how you could do anything similar with conventional bank transfers.
Well, it can be solved the way current banks deal with fraud and chargebacks: rely on a very small number of parties misbehaving and set off a small percentage of money in the system to offset fraud. Another (complementary) way is a reputation system for "nodes" (as is the case in hawala). Both imply some sort of centralization, but so do Lightning incentives (see the discussion around "payment hubs"), so not much difference there.
(1) is false. If you're not online, counterparties might decide to close open channels. But that just means you get your share of the funds in that channel refunded to you. You would need to pay a transaction fee to the bitcoin network for the on-chain transaction, but the other party doesn't get any of your money.
> 9.5 Forgetting to Broadcast the Transaction in Time
> If one does not broadcast a transaction at the correct time, the counterparty
may steal funds.
You need to be online all the time (or at least frequently enough) so that you can publish a penalty transaction if your counter party tries to steal coins from you.
you don't. you're not handing over the private keys. you sign the transaction locally, and give them to the third party to broadcast in case you're not online. it can even be multiple third parties so you're not trusting a single one. this provides better reliability than broadcasting it yourself on your home/vps/colo connection.
If you're dealing with amounts so large that you can't trust [large number] of independent third parties, you can always fall back to on-chain. but you've got to be pretty paranoid to think that starbucks is going to collude with [large number] of independent third parties to steal your $100 coffee deposit.
* The time lock on commitment transactions is typically set to be a few days, so you don't need to be online "constantly." At most you need to check in about once a day.
* As the white paper says, you can delegate this function to a third party, without giving the third party the ability to steal your funds.
* If the other party tries to steal your funds and you catch him, you get to steal all of his funds. So in practice I don't expect this to be a common situation. If the other guy believes there's at least a 90 percent chance that you'll check the blockchain on any given day, then the expected value of broadcasting a revoked transaction is going to be heavily negative.
One interesting comment about the third point is that once a channel gets sufficiently depleted (say down to 10% of the balance on one side), then there is an increased incentive to cheat (because the potential losses are lower). So you will usually have to keep channels at above a certain balance for security reasons.
Sure, but there’s services you can outsource that to. Furthermore, if you only use lightning for sending payments (the majority of us) then publishing an earlier version of the channel benefits you!
You could call them that if you want, but they are significantly different than traditional banks.
These services won't actually hold your money, just a transaction that sends all the money to you if someone tries to cheat. It's more of an escrow service than a bank, but the escrow service again doesn't actually hold anything valuable.
Not exactly banks, but there's a huge chance they will be big corps with a money transmitter license, KYC, AML. What if they refuse your channel or transaction?
I think we are talking about different things. This thread is about services that will watch the blockchain for "cheaters" and will punish them on your behalf so you don't need to be always online.
You are talking about LN nodes, which work a bit differently than you seem to think it does.
For starters, anyone can "refuse" to open a channel with you, but they can't "refuse" to pass along a transaction (well they can, but because of the routing system used, you don't know where a transaction came from, or where it's going to, so they have no reason to refuse it). So if coinbase refuses to open a channel with me, I can still send money using coinbase as a middle "hop", and they would never know.
And KYC/AML laws aren't really applicable at this level in the stack. KYC laws apply regardless of what i'm buying/selling (in most cases). If I'm buying from best buy, they need to know who I am. That has nothing to do with how I'm paying, and the "middle hops" are like routers more than banks. Even if an internet router is sending a financial transaction, that doesn't mean they need to follow KYC laws. It's the same with LN, if your node is acting as a "hop" for a transaction, you don't know who or where the tx is coming from, or who or where it's going to, and it's counterparty risk free. It's more like an internet router than any kind of financial institution, so KYC laws don't apply.
