EFF Applauds Amazon for Pushing Back on Request for Echo Data
eff.org
eff.org
It's like... PETA complimenting a butcher on a clean kill. I don't blame them though... who knew the great battle over privacy would be lost before it even started, because people want their toaster to go online?
I really despise this sentiment, it's quite defeatist and comes across as the "resistance is futile" talking point from the "privacy is dead" camp
What we need is some kind of spousal privilege [1] or secrecy of correspondence [2] for the internet age, ie, for IoT things. Surreptitious recordings from phones, messaging apps, "amazon echo"s, xbone cameras, etc, should be impermissible in court (or some variation on this theme). We struck a balance in the past, we can do it again.
[1] https://en.wikipedia.org/wiki/Spousal_privilege [2] https://en.wikipedia.org/wiki/Secrecy_of_correspondence
Unless you're Churchill rallying a nation to fight, you're in no position to accuse anyone of defeatism for pointing out the reality we live in.
That may change, but until it does, the best you can do is educate people.
I think the issue is there are no good alternatives for the typical consumer. There needs to be a push back on companies to provide these sorts of smart home/tv/music technologies for the public in a distributed manner, i.e. components not needing to connect back to a company server but instead a local one. Sure, the average HN reader can probably set something up like that, but until a solution without tons of coding and assemblage comes along, we will be stuck in this status quo. EFF could be a starting point to rally people behind creating a market demand and supporting companies in a quest for privacy-minded solutions.
So many things. Just because you can't figure out how to use a feature, doesn't mean it's not useful. Maybe I need to see how many eggs I have left, or what temperature a corner of the fridge is, or have a way to tell when my milk is going bad?
IoT is bringing in some potentially cool stuff, but being dependent on an on-line "service", and being closed and proprietary is making it worthless.
And for the ability to do that without just... you know... opening the damned fridge you're willing to buy into the IoT. That's an insane tradeoff, for benefits the typical consumer would laugh at you for buying into.
The truth is though, if you want a really secure fridge, it either needs to be much more expensive or... you know... just a fridge.
A camera in your fridge, is a camera in your fridge. It serves one purpose, and it's the one you achieve by necessity every time you get anything from the fridge.
Not sure it can help with food from your butcher or {cheese, fish}monger.
What... is the benefit of any of this?
For you. That doesn't mean it shouldn't be for everybody else.
> What... is the benefit of any of this?
Has that not already been made clear? People have listed useful examples that you are ignoring
And no, I'm not willing to buy into the IoT. Your talking about two different things. I'd love that on my network. Where I can control everything and use APIs to check myself. I'd never install something that had to go to the public internet.
I appreciate your condescension, but just because you don't have a use for something, or can't see how it can be done safely and well, doesn't mean nobody does.
Requires machine vision (or you telling it every time you take out an egg), doesn't have much to do with IoT (unless you want to see that from your TV or phone, that i guess you do).
> what temperature a corner of the fridge is
So many things to say, with the most prominent being that your fridge would have to have a ton of sensors. Again, not much to do with IoT (can even be analog, unless you want to check it on your office computer, ofc).
> have a way to tell when my milk is going bad
This is a lot harder then one might think. Did you buy fresh milk ? Pasteurized milk ? When did you open it ? What kind of bacteria live in your mouth/kitchen ? How long was it open ?
One could put a sensor for a specific gas in the milk package but even that has problems as the thin layer of milk on the side of the package will go sour faster but the milk itself will be fine.
Ofc, the fridge could guess. But you won't know anything for sure until you open the fridge and check. And the extra complexity means not just higher cost but also lower reliability.
> IoT is bringing in some potentially cool stuff, ..
Yes. I intend to, someday, plug a RPI to the TV and lights and put speech recognition on it. Star Trek FTW. But it is no problem living without it all. For a better living i recommend moving to a wooden house (or just putting wood on the inside of rooms), as it is a much better feeling then titanium white painted minerals.
no it doesn't. rfid, weight, a camera, can easily do this.
> the most prominent being that your fridge would have to have a ton of sensors
Sensors? in a refrigerator? the horror! How could we ever accomplish such a tremendous feat?!? /sarcasm
> This is a lot harder then one might think.
Sure, but it doesn't mean it's not possible. Just because something is hard doesn't mean we shouldn't do it. I'd argue the opposite.
> But it is no problem living without it all.
And it's no problem living without a cell phone, or central heating, or indoor plumbing, but we don't see many people running to that lifestyle very often.
Just because you don't want something, has no bearing on whether other people want it or not.
> no it doesn't. rfid, weight, a camera, can easily do this.
Dirt, egg-white, temperature variations. Rfid on eggs ? I get my eggs from the market. A nice lady has chickens that lay those eggs. I get 50 of them at once, that is much more then it fits in the egg drawer. (bdw lasers are better then cameras) The best solution i can think of are actually weight sensors (your idea). But even that has immense problems in that all eggs are not equal. Meaning that the whole.. what you call the little cup they sit in.. would have to be weighted. Meaning either separate cups or rubber cups, bout of which are a reliability problem.
>And it's no problem living without a cell phone, or central heating, or indoor plumbing, but we don't see many people running to that lifestyle very often.
Hyperbole. But anyway, have you tried living in a forest ? Just because you think it's bad doesn't mean it is. (see that, i can hyperbole too)
>Just because you don't want something, has no bearing on whether other people want it or not.
I didn't say i don't want it. I explained that it is too much trouble for almost no benefit. Some such things are great. I can think of a dozen things that can be done with electronics that are nice, kind-of useful and, most importantly, actually liable (fairly cheap, reliable and fairly easy to do). An egg-counting refrigerator is not one of them.
Mind you that i love electronics. (my desk is littered with it) Would even be doing it instead of programming if it wasn't so expensive.
It's not that people don't understand the implications of the IoT [1], but they often find the choice of do without more costly than do with IoT.
[1] I'll grant most probably won't understand how a lightbulb lets some script kiddie make the network shitty for a few hours or days, but even then if the people understood the why/how they wouldn't really get it in full.
If Musk and others have their way we'll change batteries at home when demand is low well before smart electrical appliances make any difference.
The EFF can't pressure lawmakers if they can't rally people to apply that pressure. When most people just shrug and buy an IoT toaster, it's not hard to see how this goes.
And not just PocketSpys and MobileAdDeliveryTablets.
I've recently been through the experience of replacing a major home appliance for a friend. One which need not intrinsically be electronically controlled. But is. With intrinsic BlueTooth connectivity (and quite possibly WiFi, I've been afraid to ask).
After units #1 and #2 arrived DOA, with electronic faults, I went through the exercise of looking for a device without integrated electronics. So far as I can tell, they simply do not exist. Perhaps if you were willing to buy a used model from the 1990s or prior -- and even in the 90s, there were electronics, they just weren't connected to anything else.
After unit #3 arrived with exterior physical damage matching a previous shipment (and directly over a very insufficiently armoured printed circuit board), and having failed to identify an alternative, we had all three boards replaced (at manufacturer cost). That's $300 in parts, or 20% of the cost of the unit, plus labour. I'm wondering how that breaks down on a per-unit basis, though our experience suggests "poorly".
I've no confidence that the electronics are suitably protected from environmental or other damage over the life of this appliance, which I estimate will be substantially less than the 20+ years of its far more robust 1960s ancestors. Which did the fundamental job very much as well.
This, I'll add, was an experience with an expensive and major purchase, based on physical characteristics. The amount of debugging and security proofing going into the $10 - $100 price-point light-bulbs, thermostats, toasters, humidifiers, security (hah!) cameras, etc., can be at most a very small fraction of that retail figure.
I'm less than enthused.
And feel that exceedingly strict legal protections will have to be imposed on manufacturers and/or retailers.
"Ok Google"
Most people have it on. You already carry around an always on, internet connected mic, why not add another one?
Even if it's "not on", how do you know it isn't?
Edit: Just to be clear, I don't do this because I think my life is so interesting that anyone who could snoop, would snoop. I do however believe that there's value in having a consistent set of behaviors and standards around these issues, in the same way that we do with sanitation. There is the added bonus that should you ever find the need to keep a secret, you won't give the change away with a change in your behavior.
They eventually handed over the data (at the owner's consent), which does two things: 1) confirms they do have data and the ability to turn it over 2) makes anyone who doesn't hand over the data seems guilty (hey, if you have nothing to hide...)
Privacy is eroded one step at a time.
Not necessarily. We don't know what they handed over. For all we know right now, it literally could be a single request to start a timer or check the weather.
The use of the data seems dubious assuming that Alexa has to hear the trigger word before sending data to Amazon. Judging by the traffic on my network, I'd say that's the case. I don't see a persistent stream coming from my Alexa device beyond periodic pings.
"Why did you use 140 gallons of water around 2AM? Long shower bud?"
“In comparison, while all four [men] were together earlier that evening, they never used more than 10 gallons of water in an hour,” police reports said. “The amount of water used between 0100-0300 hours was consistent with spraying down the back patio area, which may have resulted in the wet concrete patterns observed on the morning of November 22nd.”
The utility department's source? Each home in Bentonville was on a smart meter, police were told, to measure and record the exact consumption of electricity and water every hour. -Washington Post
The homeowner wasn't the one who died.
I guess your point is closer to 'free software makes the cloud cheaper', which in turn makes it easier to ensnare people..