We've entered a new era in which cars, like our personal computers, can have improvements and upgrades made by remote downloads. The convenience factor is wonderful (and replacement improved external charger modules are being shipped to customers). I just hope that hackers don't start disrupting vehicle performance and safety using that same channel that is supposed to be used to improve them. I also hope that nobody is making custom modifications to their cars that will corrupt these updates. I hope the long track record with automated PC security updates had identified the necessary security features for such upgrades. (I just get a little nervous posting a comment that says "hope" three times.)
You hope in vain alas, DrQuine. Hackers will exploit these loopholes and people will mess up with manufacturer installs. We can't have our cake and eat it, we have to live with the risk, try to minimise it as much as possible, knowing it will always be there.
While I agree that the hacker angle is negative, I personally can't wait to see what modifications people start doing to their own vehicles! Obviously I don't want any safety issues to arrise, but modifications to things you own are always interesting to me.
@caleb, people have been modifying their cars for decades. People "modify" their car's mileage so that they can cheat (on the age of their car0, or modify their speed so they can go faster, etc. etc. Such practices have been hazardous -- for safety reasons, as well as for protecting the value of one's car -- and carmakers have been fighting it.
It seems to me there is a fundamental issue when people are modifying the vehicle firmware and software that controls the safety systems and that may be subject to mandated safety updates ("recalls"). A hobby programmer cannot possibly know how their modifications will interact with all future hardware, software, and firmware updates. Perhaps, the personal software will be blown away by a software update, perhaps it will interact adversely. I think I'll confine my programming efforts to other venues.
As noted at the end of the article, there is some controversy over the use of the term "recall" in this case. No vehicles were actually recalled and the fix involved a wireless software update and new chargers being sent to vehicle owners:
@rich.pell, to Elon Musk's tweet, I would say, that's a BS. It's semantics. If your car needs to be fixed, it is irrelevant whether you bring your car physically to a shop. While I do appreciate the convenience of over-the-air software patch practice, well, this was a recall.
Initially, based off your comment above I was inclined to disagree with your adamancy to label this a "recall". However, once you put the perspective on this that a car can kill someone, I immediately see what you mean.
Maybe there should be some kind of new term for potentially lethal issues requiring software updates. I see this as an issue that might reach far beyond just cars, and the term "recall" itself is getting a bit dated, but in this case seems like an accurate depiction.
@Junko: If your car needs to be fixed, it is irrelevant whether you bring your car physically to a shop.
If your car needs to be fixed, it's sure as hell relevant to you.
A friend recently bought a Tesla Model S. He is over the moon about it.
He rhapsodized about an issue, where his car died in his garage and would not start. He called Tesla. The next day, Tesla was there with a truck and techs. The diagnosed the car, loaded it on the truck and took it away. They returned it the next day, fixed, and put it back in the garage where they got it.
It was warranty service, and didn't cost him a dime. Can you imagine any other auto maker doing that?
He understands that it's part of Tesla's effort to build their brand and confidence in it, since he's a sales and marketing guy by trade, but he's still grinning with delight.
How often does your car breaking down leave you smiling at the end of the process?
While I do appreciate the convenience of over-the-air software patch practice, well, this was a recall.
Was it? A recall is normally though of as having to bring the car into the shop. Is it a recall if you don't have to?
And with the increasing level of automation in vehicles and what can only be called "smart cars", we are approaching the point where manufacturers may automatically push firmware updates to vehicles to correct problems/add functionality, so you are automatically running the lastest version. Is that process a "recall"?
@DMcCunney, I love your friend's Tesla story. I've heard similar stories as well. The point is that Tesla is doing everything it can to establish its brand and please its customers. That's admirable.
That said, here's my argument about "recall" definition.
OK. You are right. Customers did not physically have to return the car -- hence it was not "recalled" as far as the definition of the word "recall" goes.
But motor vehicle "recalls" are issued because of safety standards. To me, by saying this car was NOT technically recalled, it is as though Tesla is saying that this overheating issue does not fall into the category of "defect." I find Musk's tweet clever but also that tweet makes him sound irresponsible. Just as a reminder, NHTSA only issues a "recall" when it finds the following.
The United States Code for Motor Vehicle Safety defines motor vehicle safety as "the performance of a motor vehicle or motor vehicle equipment in a way that protects the public against unreasonable risk of accidents occurring because of the design, construction, or performance of a motor vehicle, and against unreasonable risk of death or injury in an accident, and includes nonoperational safety of a motor vehicle."
Tesla is taking a page out of the High Tech World guidebook - make it easy for customers to update and fix bugs easily through a simple download. The word "recall" in this article should be the phrase "bug fix sw download". It sounds like Tesla has been doing some superior bad press damage control.
I think there may be some confusion. From my understanding, the reason for this "recall" was not to fix a problem Teslas had but to workaround a problem owner's could have with their home wiring. If a home's electrical was installed to code, the update or "recall" was unneeded. I understood it as a safety measure to protect owners from their own home's faulty wiring. Based on that understanding, this wasn't something that Tesla was liable to address, but they did it anyway. Please let me know if I have the wrong understanding here. The press release did not attribute anything with the car itself, but with the charging environment it was being connected to.
Your understanding is absolutely correct. I would not argue with that.
As a safety measure, Tesla took an initiative to report that to NHTHA, and that "recall" notice was posted by NHTSA Monday this week.
But here is the thing.
Even if this overheating issue has everything to do with corroded wall sockets, the fact that Tesla was able to address the issue by changing software means, well, there was a way to fix this on the automotive side.
And by debating it was a physical "recall" or not, we are underestimating the very fact that there was something Tesla could have done in the first place to make this safer.
Naturally, it always goes without saying, early adopter "zealots" will always put a positive spin on anything to do with the object of their zeal. So, instead of figuring that it would be practically impossible for the established automakers, who sell 100s of thousands of cars every MONTH, to make "house calls," the zealot makes it sound like government-subsidized Tesla, selling way fewer cars than that per year, is doing something extraordinary. Come now. Let's not gush.
The fact that the Tesla software can be updated remotely is both good and bad, right? Let's be consistent. This can be done to operating software in ANY new car, in principle, but most people seemed to think that it was a bad idea. Most people, in the past, thought it best to take the car to a garage, where the update can only be entered via the OBD-II physical port. Now that we're gushing, though, it's a different story. In principle, any update to any ECU in my GM could be transmitted via OnStar. (They already give the car periodic health checks remotely, via OnStar, so why not updates too?)
Finally, the idea that charging a Tesla can burn your house down is not something to be brushed off either. Filling the tank of your standard car, while in the garage, could also burn your house down. That's why we don't have gasoline pumps in garages. Presumably, to allow anyone to install a gasoline storage tank and pump in his garage, you'd have to pass some sort of stringent inspection. High current battery charging, of the type needed to transfer enough energy to power a car for more than a really short ride, is serious business. It's hardly a surprise that, meeting code or no code, careless homeowners may not know what to watch out for. How many people look inside their 230V receptacles to make sure they are sparkling clean, for instance? Or will Tesla also make house calls to do that?
Something here doesn't pass the smell test. If there is abnormally high resistance somewhere, then then the heating will occur exactly where that high resistance is, and NOT ELSEWHERE. Thus, if the high resistance abnormality is in the user's provided circuit, the overheating should occur only there, NOT in the internal guts of what should be a perfectly high conductivity, zero resistance, simple 240v NEMA Tesla plug adaptor. So why and how can a perfect zero resistance simple Tesla adaptor melt when the high resistance problem is asserted to be in the customer;s wiring?
A Book For All Reasons Bernard Cole1 Comment Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...