Tesla won’t give drivers their own crash data without a court order
consumeraffairs.com
consumeraffairs.com
> When asked why Tesla has an apparent policy requiring drivers who want their own data to file for a subpoena, a spokesperson framed it as a privacy issue.
> “We handle all customer data in accordance with our Owner’s Manual and privacy policy, which clearly outlines the type of data we collect and the lengths that we go to protect a customers’ privacy in the process,” Tesla said via email, referencing back to a statement on the Tesla website about the company’s data collection practices.
You can pay a third party $1000 to access the data, or Telsa will publicly shame your driving if they think you are in the wrong during a crash, but they really need to fight you to protect your privacy from yourself?
EDIT: A nasty surprise that might be, too, if it turns out there was important data via that service that you wanted to access abroad but couldn't.
It could comply the normal way.
(i) might be enforible against a US corp if they have something in the EU to do business
* and there are European citizen dual passport shenanigans
If the company is established in the Union, then GDPR applies to all personal data processing it does. See Article 3(1): "This Regulation applies to the processing of personal data in the context of the activities of an establishment of a controller or a processor in the Union, regardless of whether the processing takes place in the Union or not."
The question, then, is Tesla merely a company that is offering goods and services to people in the Union, but not itself established in the Union, or is Tesla established in the Union?
Point: how is the GDPR ever going to be enforced between US citizens and US companies-with-an-EU-presence?
No, they'd sue to get it from the EU one. As it's part of the same company. The rest of your questions are predicated on the core misunderstanding you have of this so don't apply.
> how is the GDPR ever going to be enforced between US citizens and US companies-with-an-EU-presence?
4 percent of the organization's annual global revenue will work quite well I'd imagine.
It also gives the right to to sue. See Articles 79, 80, and 82.
There's a nice summary here: https://privacylawblog.fieldfisher.com/2016/getting-to-know-...
Here's an interesting law review article examining these developments: The Arc and Architecture of Private Enforcement Regimes in the United States and Europe: A View Across the Atlantic, https://law.unh.edu/sites/default/files/media/rathod_-_final...
Personally in terms of public policy I prefer private action, but maybe that's because I'm American. Private actions are usually predicated on actual, individualized harm, whereas regulators can strong-arm companies without any evidence of actual harm. Theoretically it makes for a better business environment, especially for startups, legal anxiety about nuisance lawsuits notwithstanding[1]. But private action doesn't scale; if you don't permit aggregation of claims (i.e. class actions) then private enforcement can't redress systemic behaviors.[2]
[1] The anxiety is invariably overblown. I think it's because most people's eyes will glaze over when hearing stories about regulator enforcement actions. But stories about slip & fall lawsuits are both legion and relatable; like with terrorism, people develop a false sense of the legal risks and costs.
[2] I mean, it could work without aggregation if you changed the rules of litigation to really streamline individual claims, such as by shifting the burden of proof onto the defendant for certain categories of behavior. Then it would sort of act like a dynamic tax that responded to business practices. Keep your customers happy and you pay a minimum, baseline tax--the cost-of-doing-business that is whatever amount you want to pay out for fraudulent claims, not unlike what businesses already do with accounts payable and shrinkage. Do something that might upset your customers and you'll have to face an onslaught of cases that either require a quick payout or costly litigation. I imagine such a regime would look much like that with class actions, with specialized law firms that identify and aggregate plaintiffs to benefit from economies of scale, and take a cut of the proceeds. But it would still make it much easier for people to prosecute small claims themselves.
Airbag controllers have a circular buffer of the last 15 seconds or so of the vehicle basics - speed, steering angle, acceleration, braking. No GPS. That data can usually be retrieved from the wreckage after a crash. This requires physical access to the vehicle. The data is used to analyze airbag failures, especially unwanted airbag deployments, and for crash reconstruction.
Airbag controller data is much less extensive than what Tesla seems to sending back to the mothership.
More likely it’s “we want to be the only one who can see it.”
Semi-related: Reminds me of the time Amazon promoted a “recommend books to your friends” feature, and someone used it to make a recommendation to me with a vaguely threatening message attached. I asked them who it was and they screamed bloody murder about protecting customer privacy and I’m like “really? Privacy of your identity to a friend you’re recommending a book to? As a feature that you’re allowing to be repurposed into an anonymous messaging service?” (Eventually they relented.)
However, if you want the data, you can't have it… you have to go to court to get it.
You know what sounds nice?
Having the choice to not buy a Tesla.
I don’t need Government™ to tell me what kind of contracts I can agree to.
This, plus the fact that the increasingly monopolistic reality of ISP in the United States is a direct result of overregulation.
We don’t need more laws. We need to enforce some of the most basic laws that have existed fore more than a century.
If Tesla's business model proves successful, all other car makers will copy it, and then it doesn't matter which brand you buy.
Look at the TV market. You technically have the choice not to buy a "Smart TV," which sells your info to god-knows how many third-parties and spies on you 24/7. But good luck finding one. Sony, Samsung, LG? They all have the same spyware features. Your only option is to buy brandless Chinese TVs with terrible picture and worse sound quality.
How long until all manufacturers start acting like Tesla? What choice will we have then? Start walking everywhere like our ancient ancestors?
I don’t have any problem with your union; in fact, I might join such a union. Just don’t force everyone to join it, which is what happens when laws are made for everyone as a result of the bad decisions of some.
I don’t know if such a union is legal to form… which I guess illustrates the point that @gonational is trying to make; too many laws can bite you.
Second and more importantly, the companies have found a way to circumvent one of the most important and fundamental pillars of our civilization, access to judicial system, through forced arbitration and class action waivers. If they had the power to get rid of something this fundamental, I am sure they will find a way to defeat any attempt to form a union like the one you advocate though laws, contracts, EULAs, etc.
Pertaining a purchasers’ union, I don’t mean something that would bureaucratically bogged down purchasing. I mean something like a club where you only purchase from those companies approved by the club, and everybody in the club gets to vote on which companies will be allowed to be purchased from.
https://www.tesla.com/about/legal
> Tesla may transfer and disclose information, including information that may or may not personally identify you, to third parties to comply with a legal obligation (including, but not limited to, subpoenas);...to verify or enforce our policies and procedures;...to prevent or stop activity we may consider to be, or to pose a risk of being, illegal, unethical or legally actionable; or to protect the rights, property, safety, or security of our products and services, Tesla, third parties, visitors, or the public, as determined by us in our sole discretion.
Is only the owner allowed to drive a Tesla? Do any other drivers have to sign a document on the dashboard before starting?
https://gas2.org/2015/01/06/this-is-what-30000-of-damage-loo... and lots of forum posts out there on this
Indeed. It is disturbing that Tesla is not in the slightest bit concerned by how obvious it is that it is lying about its motives, as that indicates that it does not see any downside from doing so.
Plus Tesla never releases the crash data, they release select snippets that make them look good. For example, the famous "six seconds" claim in the barrier incident but absolutely zero data on their automatic emergency braking's actions, what autopilot was doing, what sensors detected, or any of their other systems.
https://www.reddit.com/r/teslamotors/comments/8jz5tl/tesla_r...
> Ok so here's what bugs the shit out of me: The overwhelming majority of the time that my car nags me to put my hand on the wheel, my hand's already on the wheel, I'm just not applying enough torque for the car to "see" that because my left elbow is resting on the window sill and my hand is resting on the 9oclock position. So I give it a quick wiggle and then we're friends again. This means the car logged a "no hands" event.
In the case of the fire truck, the driver pretty much admitted to fault. But what about the California man (i.e. the "six seconds" incident) who died by running into a median? We don't know for sure if his hands were on or off, or when, or if the car properly detected it. Tesla was kicked off the NTSB investigation for releasing data. Tesla's argument was that it had the right to make a defense, but I think it's reasonable to argue that NTSB's rule is to prevent such selective release and framing of data:
https://www.ntsb.gov/news/press-releases/Pages/NR20180412.as...
And that is something that I support the GDPR for.
Cars are a significant property which, unlike a lot of other items, even comes with a proper title of ownership, it's hard to imagine any of Tesla's behavior regarding the car's data, activation practices, etc. would hold up in a court of law.
Given that most people who can afford Teslas can afford lawyers, I'm kinda surprised nobody's gone for the payday.
Like visiting websites while not logged into anything in generating tons of data, but it can't be easily linked back to you.
It would be possible to link some of it back to session IDs or other things you could be linked back to your device, but that doesn't show it was you using the device, so they may be releasing data belonging to someone else using your device to you. And that is before we get into possibilities like cookie hijacking or other methods of falsely tying your device to that generated data.
The argument here is that the company knows who the data belongs to, and the person above is saying that they shouldn't be able to abstract "you" away from the data and claim it as their own property.. especially because, as we know, most anonymization techniques have serious flaws that in fact do allow individuals to be identified after close scrutinization.
I agree that your proposal would be a good extension. Or at least that it should be the default, and you'd need to sign for being okay with not owning the data. Or that those devices contain warning stickers.
Sounds like PII!
I did this by accident when I "bought" an iPhone a few years ago and was told I couldn't keep it after I water damaged it and got a replacement. I started buying devices outright after that.
Got to amazed at chutzpah some companies have, my favorite whipping boy is Onstar. Here my nice little car has the ability to notify authorities in a crash BUT only if I pay for that service. Now on one hand I would not mind a nice small fee per month, say five bucks or less for something that would infrequently used if ever, but with Onstar the lowest cost is 24.99 a month!!
fwiw, when I had my car new they would even send driving reports, like when the system decided i accelerated to hard, braked too hard, and such. Real joy in what they collect AND sell to third parties
Classic work-for-hire includes ghostwritten novels and autobiographies, corporate reports, web pages, and most likely every line of code you've ever written while employed by a corporation.
There are jurisdictions where moral rights cannot be sold, and others where they can, but work-for-hire nearly always sidesteps the issue.
I agree that preventing a person to choose to sell something they own is a restriction on their freedom. But problems arise when the negotiation of that exchange is unilateral, relegated to EULAs and contracts which are incomprehensible, non-negotiable, and mandatory to purchase the product, and performed in a less-than-perfectly-efficient market.
It would be one thing if I could buy a Tesla, a Nest, or log onto a site with ads, and the process included a step where I could elect to share my data, to not share my data, or to give them the data and promise not to ask for it back, with different price points for each.
It's another thing when giving people the freedom to give up certain rights results in all manufacturers requiring people to exercise that freedom in order to purchase products.
The market is far from efficient. There is no car that's just like a Tesla but costs more by the value of this data exchange. There is no negotiation of the contract with Tesla, it's built into the purchase price and required by the legal department.
Observing that the market is pervasively inefficient in a way that leads to a bad outcome, and devising regulations that correct the inefficiency, is precisely why we have laws and governments.
In prior times, such covenants were also used to try to racially segregate property, by forbidding sale to a person of a race the covenant's creator disliked (such covenants are technically still valid, but unenforceable by US courts).
And they're sometimes used for other things, too; I recall there was a house near my college's campus that the college wanted to purchase and use, but the owner had set up a restrictive covenant that would forbid any later purchaser/inheritor from selling it to the college.
If I'm allowed to take the 5th to not incriminate myself and not reveal my memory (data in my head), I should be similarly allowed to delete my data when it is my data in my device.
It'd be an interesting case to take to the Supreme Court.
The next step in your scenario is "that security camera has data on me, I want it deleted".
The link provided by wang_li applies and it seems such questions have come up in the past in front of courts, and honestly I won't be surprised if they come up again.
http://lawdigitalcommons.bc.edu/cgi/viewcontent.cgi?article=...
Once money exchanged hands, it's not their product, it's mine, as is _all_ of its output, including logs, etc.
Otherwise, I'd call it 'spyware'.
Here's an example: You go to the library (or a privately owned web cafe), and write your stories there. It doesn't matter that you don't own the device. It's still your data.
That's BS. The real justification for anyone having that right is that 1) the data is about them and 2) they own the machine that collected/created the data.
> The device manufacturer/website owner/etc does the work of creating the data, yet you think it's yours?
They did the work of creating a device for sale. After the sale, their rights to the device end and the data should be owned by the new owner.
But the post you're replying to wasn't about copyright at all.
I'm with you and 'adrianN on this. I hate this direction and wish we could turn it back somewhat, or at least cover for the negative aspects of servicization of products. Regulation might be a necessary tool for that.
In the future, everything will have sensors in it. If we leave things as they are now, your Samsung refrigerator will be reporting back what you eat, and when. A good place to sell that data is to grocery stores and General Mills.
That's great if you want your doctor to know that information, or if you need an alibi; "your honor, I was eating yoghurt at the time".
Oh wait - bad example, because you own the refrigerator. My point was that in the future, when you go outside, everything will have recording devices in it. So, the public spaces and government rituals you are subjected to need to be dealt with, as well.
Property rights no longer suffice for these situations.
And remember this story:
https://arstechnica.com/cars/2016/03/man-hacks-tesla-firmwar...
https://www.gov.uk/respond-data-protection-request
> Under data protection law, anyone can ask if your organisation holds personal information about them - you must respond to their request as soon as possible, and within one month at most.
But, then again, I can see how you'd think Tesla was pure evil too if you buy into all the endless nonsense published about them without doing any research.
Which begs the question, why do I comment in these discussions at all? I do not know.
More “journalistic” nonsense. I agree they have a lot of legitimate issues (cash flow, scaling, retention, working conditions, process maturity, overpromising) to deal with, but can’t journalists focus on those instead of clickbait?
Sidenote: if you take issue with this, sponsor a federal law to fix the problem through your legislator
But this? Asking me to take personal responsibility with no legal recourse? This is the line they can't ask me to cross. If this is fake news, fine, it'll come out. If it's not, then sorry Tesla, can't do it.
I consider them holding themselves to a higher standard, but as in politics, you can’t find someone who agrees with you 100% in office. Have to compromise, and this isn’t a hill I care to die on.
As I stated above, it’s a public policy issue, not Tesla specific.
> if you take issue with this, sponsor a federal law to fix the problem through your legislator
More to the point, Tesla's wish to be seen as a market disrupter is only going to hold as much water for me as their willingness to do it when it counts. This is one of those "when it counts" moments for me.
I mean, it's a bit silly, no? Everyone complaining in this thread while Tesla churns out thousands of vehicles a week to a backlog of consumers willing to wait at least another year for their vehicle. To you, it might be an issue. To those who matter (Tesla investors, employees, and customers), it isn't. ¯\_(ツ)_/¯ The market has spoken.
The market also speaks by shorting Tesla, and with TSLA volatility on par with a cryptocurrency.
Not that Tesla owners should be the only ones interested in the policies of Tesla, one of the most high-valued car companies and potential industry leaders. I'm not an Android user but I pay attention to how they deal with privacy and security issues.
No, it was the act of going to the cross that made it possible for Rome to not crucify Christians in the future. Or for you non-Christians, an act of symbolism through self-sacrifice. If Tesla wants to be Jesus, it's going to have to get up on the cross. Otherwise nothing changes.
If Tesla can't shoulder that responsibility, I'm certainly not going to.
They're not. They restrict access to the service manuals behind a ridiculously expensive subscription that's only provided in states where it's legally required. It costs $30 a day or $3000 a year.
Service manuals for other cars cost ~$50 for a hardcopy.
https://electrek.co/2017/01/30/tesla-opening-up-service-repl...
Tesla's attitude is almost like you don't own your car, you're just leasing it in perpetuity from Tesla for a one-time fee.
This affects me personally.
I have no real dog in this fight though. Any reasons you have to choose one car over the other are your own, and it's not my intent to criticize one choice or the other. The reasoning here just seems slightly difficult to follow.
From a consumer standpoint you're buying a black box controlled by a company that frequently behaves in a overtly hostile manner towards the "owners".
Imagine Apple not letting you use the accelerometer or GPS sensors in your phone. Instead they'd just collect telemetry from it. Now imagine you claim a "spontaneous shattering of the screen". To counter this, Apple publishes select pieces to show that the phone experienced acceleration consistent with a crash after a drop. On Twitter.
Now would you expect Apple to do that? I'm not finished. Would Apple go on the record to say that you personally have not been taking proper care of your phone? That's what Tesla is doing.
> Tesla's internal policies regarding the release of crash data to consumers doesn't say anything about the 'awesomeness' of the car itself.
Much like food by a cook that doesn't wash hands can still be delicious.
>"The suspension ball joint experienced very abnormal rust,” Tesla said of his car, adding that Cordaro “lives down such a long dirt road that it required two tow trucks to retrieve the car.” The car, Tesla also added, was “caked in dirt."
I'll be one of the first people to tell you that Pennsylvanians are a bunch of loud mouth complainers who don't really have it as bad as they say when it comes to vehicle rust.
However, the idea that a low volume, relatively new (there are no 20yo Teslas let alone Teslas that have spend those 20yr in salt states) car company, primarily located in California (!!!), that primarily recruits in California (!!!), is going to be able to determine "normal corrosion" looks like is laughable. Even established car companies can't accurately and consistently predict corrosion 10yr out. Tesla doesn't even have a big enough data-set to make that prediction, let alone the expertise.
Really?
Even I don't press my brake like that when parking.
Will autopilot even let you floor the accelerator with a car right in front of you?
Yes. Autopilot does not override any user input. This is a major reason why L2 system failure will always be the responsibility of the driver.
You might if you thought it was the brake and the car was accelerating.
https://www.caranddriver.com/features/its-all-your-fault-the...
And this is normal in a mistaken accelerator pedal application. You press down on the 'brake', the car speeds up, you panic stomp down on the 'brake' as hard as you can.
without unfettered access to the full data that Tesla has we will never know. Unlike Toyota Tesla does not seem willing to cooperate.
I am all for logging and accessibility to it but logging is only as valuable when you know the how and why of each event.
Such data loggers are commonly used in racecars, using either their own sensors, or in stock versions, also tied into the OBDII ports. It looks at everyghing from driver inputs, to engine performance, 6-axis G loads, etc.
Seems it'd be pretty nice to get the data anyway to help manage your own driving (or check the habits of someone borrowing your car, like your kid), and in case of an incident, your own log of the events so you don't have to pester/subpoena Tesla.
I called Netflix customer support and asked if I could get location/IP logs. The Netflix support person was very sympathetic but said that it was not company policy to release that info. I didn't ask for the issue to be elevated, or ask about a subpoena (I assume Netflix would respond to a subpoena).
But IMO, the difference between my situation with Netflix, and Tesla is:
1. Netflix does not seem to have a practice of divulging my user data in a public forum.
2. The circumstances for my data request were unusual -- it's not as if every time a Netflix-equipped device gets stolen, Netflix is expected to respond, or even be aware of it. But presumably, Tesla has a data process ready in the event of every kind of accident, no matter the actual cause or circumstances.
In other words, what I was asking from Netflix was refused because they don't have a standard operating procedure for responding to info about IP/location requests. If such a thing were needed, it's not hard to imagine Netflix making it a service, just as FB and Google data takeouts include login location history.
But Tesla does have standard procedure for retrieving, storing, and transmitting vehicle logs, including divulging the data publicly without being compelled by the government. Which is why users seem to have a greater expectation that Tesla should be able to send them their own data.
(Note, this was separate from the real problem of floor mats interfering with the pedals, for which there was a recall).
The NHTSA apparently agreed.
Because proof that the ETCS-i caused the reported UAs was not found does not mean it could not occur. However, the testing and analysis described in this report did not find that TMC ETCS-i electronics are a likely cause of large throttle openings as described in the VOQs
Acronyms and details in the report:
https://one.nhtsa.gov/About-NHTSA/Press-Releases/ci.NHTSA%E2...
2) If so, can the customer actually get a court order in arbitration?
[+] https://www.dailykos.com/stories/2018/5/29/1767826/-The-War-...
EDIT: below content moved up to this comment from danso reply in order to be more concise as well as to keep signal to post ratio high:
I do my best to provide non-biased sources/citations, as well as provide arguments without any intellectual disingenuity. If you can find something that is inaccurate or subjective, please point it out.
It's fairly obvious that the data used by Reveal was manipulated in their sensational Tesla piece, and I'm happy to continue to aggregate additional sources that bear that out. You left out the next six paragraphs that explain why there is no basis in fact for what Reveal wrote.
Sounds like an article committed to defending Tesla. Not that there's anything wrong with that, but "with as little bias as possible" is debatable.
For example:
> The first of their “personal stories” was about how a person involved in developing the factory was told that they can’t use yellow caution tape or beeping forklifts because they offend Musk’s sensibilities. The lack of these things, according to Reveal, could be to blame for the “high” rate of injuries.
The assertion that Tesla lacked yellow because of Musk's preference is attributed to the factory's former safety lead, Justine White, and documented in her 2017 resignation letter: https://www.documentcloud.org/documents/4437759-Resignation-...
You can argue about whether Reveal should have talked to Tesla's former safety lead, but the story doesn't assign blame to lack of yellow or whatever for the rate of injuries. In fact, the story is not about Tesla's "high" rate of injuries. The story notes that Tesla's injury rate in 2017 fell steeply. The point of the story is that, according to Tesla's own internal log, injuries that are mandated to be reported were not listed on the official report.
According to a Reveal followup a month later, Tesla's official injuries report has been amended with the injuries that Reveal accused Tesla of hiding:
https://www.revealnews.org/article/tesla-left-injuries-off-t...
The dailykos article doesn't have to agree that Reveal was overall in the right to investigate Tesla, but an objective analysis would note Tesla's actions in response to the investigation.
In fact, if you read the Reveal followup, you'll see that the added injuries don't even make Tesla a particular outlier:
> The additions raise Tesla’s 2017 injury rate to 6.3 injuries per 100 workers, just above the 2016 industry average of 6.2.
So noting the followup, and the adjusted stat, is well within the comfort zone of an article that attempts to objectively defend Tesla. Yet the dailykos writer seems to have completely missed reading the followup and can only throw insults at Reveal. That does not seem like an "unbiased" analysis to me.
That said, I'm not making an a priori argument that Reveal is right. My objection was to you claiming the DK post has "as little bias as possible", when it seems to be a vigorous defense of Tesla, by a Tesla owner. Again, nothing wrong with stated bias -- I'm objecting to your characterization.
For example, the author has time for rhetoric like this:
> Of course, Reveal has shown no interest whatsoever in fact checking. If you have anything bad to say about Tesla, by all means, give them a call. They'll write an article about whatever you tell them.
Yet this objective article you tout fails to note that Tesla -- about a month after Reveal's initial investigation that Tesla "left injuries off the books" -- belatedly added the injuries that Reveal called out:
https://www.revealnews.org/article/tesla-left-injuries-off-t...
> Tesla Inc. recently added more names to its list of injured employees after Reveal from The Center for Investigative Reporting raised questions about whether the company was counting all of its work injuries, records show.
> The electric car company added 13 injuries from 2017 that had been missing when Tesla certified its legally mandated injury report earlier this year.
Again, feel free to trash Reveal's overall aims and ulterior motives. But an objective critique of their investigation should probably note that Tesla moved to fix the errors that Tesla had previously denied.
If it's true that Tesla is hiding workplace industries, this is of course something that I will want to find a way to get answers to at the annual meeting. That is unacceptable regardless of organization.
In my opinion, the company needs what Uber got: a radical leadership change. It may tank the stock for a few quarters, but it'll make the company stronger in the long-run.
They really should just cut their losses on autopilot and focus on figuring out their production line.
but Model 3 pricing still being higher than promised, and not being produced fast enough, and Model S running into things while on autopilot was also helping before that.
At the end of March Tesla had $2.6bn of cash on its balance sheet [1]. Against that they have (a) $1.8bn of current debt (i.e. due within one year), $1.2bn of which is recourse (page 24) and (b) between investments and operations a cash burn of $1.1bn in Q1.
Tesla needs to raise new capital. That is as obvious to investors as it is to bankers. So the latter are getting ready. I've received a call every week for the past months from Morgan Blergh or Goldman Go-away sounding me out for exotic hybrid debt products. After those calls, I dutifully searched for and read a few articles. Publishers saw the views. So we get this.
From a utilitarian perspective, the financial press is informing investors about something they will need to dig into soon. From a stupid perspective, we have the bandwagon.
[1] https://www.sec.gov/Archives/edgar/data/1318605/000156459018... page 4
It appears from the comments that the Cult of Elon can’t handle this. Elon is perfect, anything negative about him or his interests are conspiracies to destroy him.
In order for a media conspiracy to work, you would need to convince dozens if not hundreds of individual, competing journalists (who are not exactly known for their conformity) to go along with a scheme. I don't believe that should pass Occam's Razor for anyone.
In addition, you'll often see more negative reporting come about after one story as people who have inside knowledge choose to share their stories after initial reporting. I had people reach out with additional information, unsolicited, to both my former colleagues and me following the publication of a piece. It's generally along the lines of "oh, you think that's bad, have a look at this."
All that happens because it's very rare to see isolated incidents of bad behavior inside an organization. When there's smoke, there's usually fire.
Part of why I find the Elon Musk story so fascinating is because we're seeing in real time how the 'great man' and 'great forces' narratives of history can both be applied as filters to understand what's going on. I would think the flood of bad press is better explained by the 'great forces' filter, but I don't understand that filter intuitively enough to usefully speculate what's changed.
I think this is just how he operates.
Edit: I'm one of those people that believe in what Elon is doing and while I don't consider myself a fan-boy, I get plenty frustrated by all of the biased negativity I somehow feel the need to inject some balance by saying something. That may come across as only ever defending, but the amount of negative fluff articles is astonishing.
https://www.theguardian.com/technology/2017/apr/03/the-custo...
> The Guardian could not find a single case in which Tesla had sought the permission of a customer who had been involved in an accident before sharing detailed information from the customer’s car with the press when its self-driving software was called into question. Tesla declined to provide any such examples and disputed the description of its automation software, called Autopilot, as “self-driving”.
I've Googled around but haven't been able to find instances of other companies releasing data in the wake of an incident, or being criticized by owners for not releasing data, though obviously the average car does not have the data systems that are standard on Tesla.
When Facebook gets criticized in an article for exposing/sharing user data, do you ignore the article if it doesn't have a complete rundown of how every other tech company may or may not abuse user data?
Accusations of bias like this always confuse me, as if reporting is obliged to present both sides: "Tesla won't give customers their own crash data. But some say that's a good thing!"
No, I'm not a Tesla customer. Nor do I have any position in Tesla stocks (or oil, for that matter).
To be honest, I think the Tesla cars sound quite good, and appreciate what Musk and the company has done to make EV an attractive industry. But as you've probably seen in my Tesla-related submissions, I'm currently not a huge fan in how Tesla is currently promoting AutoPilot and responding to AutoPilot incidents. I don't know much about car manufacturing but I do have an interest in AI/machine learning.
https://news.ycombinator.com/item?id=17232991
I get that Tesla's in a somewhat tough situation, having increased pressure to prove its manufacturing and sales are long-term viable. I don't agree that Tesla gets disproportionate negative media/attention (relative to the positive media that it gets), but I can see why Tesla would feel particularly harmed at this stage in development.
But since they've sold AutoPilot as a major defining feature, I think it's fair to raise it as an issue. Particularly on HN given the general interest here in AI/machine learning.
Note: My question is in no way trying to argue against the consumer owning the data. I'm just wondering how this kind of data could be useful without access to the means to analyze it.
Edit: Not sure why all the down votes. I explicitly tried to indicate I was asking a question about the data value and avoiding the ownership question. I, personally, think the data should be the device owner's property. But my opinion on ownership and my question wasn't related to the ownership.
It's not as if there's only one objective way to analyze the data.
We can also reverse this question: why would Tesla hide this type of data from their customers so intensely?
One reason that comes to mind on both questions, is that customer's (or someone on their behalf) are able to review if the crash is due to software malfunction.
> isn't this otherwise very similar to being in a plane crash and demanding the black box data
Most people don't buy (hence own) a plane but buy a ticket with an airplane company instead. The airline company actually does get to access the black box data in case of a crash. So do the authorities.
To give them the benefit of the doubt:
* Do you consider this diagnostic data to be highly personal, worth protecting heavily? (Does it include GPS?)
* Do you trust your engineers to properly verify the identity of a person before granting this potentially highly personal data?
* Does your legal department trust your engineers that much?
* Do you have the resources/desire to lay out a foundational system of getting that identity verification and granting that data?
If you answered "No" to any of those questions, you might very well want to say "We will do it if it is required of us by a court order."
While it is unideal, it's not nonsensical.
Depending on the human "readability" of this data - how could someone without proprietary knowledge find any value in it? That's my question - which is more leading towards "Tesla won't give drivers a means to analyze their crashes" which is a much harder pill to swallow as a Tesla driver, law enforcement officer, etc.
And what do you base this speculation that Tesla does something so strange? Because Tesla has a machine-learning heavy reputation? Google does far more machine-learning and yet allows us to export data, such as geolocation, in portable JSON format:
https://blog.rakam.io/rakam-google-takeout-importer-analyze-...
I never said it was. I'm not sure why you continue to try to come up with unrelated arguments to push your agenda in a discussion completely unrelated to what you're talking about.
"And what do you base this speculation that Tesla does something so strange?"
Because educated speculation can lead to positive and constructive discourse? I have a fair amount of experience in the ML and CS field and I could imagine a situation where the data is in fact difficult to analyze.
Also, logging geolocation data is a fair amount different than logging neural weights or other types of ML data. Without knowledge of the types and composition of ML algorithms those numbers simply won't help you. Which was my point to begin with.
I'm talking about data that's incapable of being analyzed in a meaningful way without contextual / proprietary knowledge of how that data was created, is used, and is stored. In reality, I'd wager that most companies have more of that type of data than not.