Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

There are plenty of rational reasons for not wanting a smart meter. Don't let the irrational people detract from the fact that there are many real problems with smart meters. Especially lots of privacy issues.

In the UK, I report my own meter readings and the electrical company probably only really ever goes out once every few years when tenancies change. So I actually don't see what money it saves them asides from the money lost from chasing up issues where people are trying to cheat the system.

In this example it really makes me wonder if replacing all the meters in the country with non-intercompatible smart meters really saves that much money. So you have to start asking what else is in there for them to do this. Probably money for the data I would have to imagine.

Also, given how absolutely atrociously shite the security of these smart meters is (and you'll have to trust me on this, I don't know how public this information is) I wouldn't want that crap anywhere near my house in the eventuality that someone hijacks it to make it look like I'm using more electricity when they're using less (while keeping the overall books balanced so to speak) or some other nefarious purpose.

Certainly these meters won't give you 5G cancer, but they're really a horrible idea as they stand and I don't recommend anyone install them, at least not in the UK.



Sure, but those not-irrational reasons are locale-specific. I've never heard of someone self-reporting the meter reading here in the US. The meter still keeps a local log of how much energy was used, so worrying about being framed for using too much energy still feels a bit irrational to me.

OTOH, smart meters allow the utility to charge TOU rates, which helps even out the load on the grid. It benefits the utility, of course, but also customers. For example, it is minimally inconvenient to set my car to charge or my dish washer to run at night instead of day, but I might not bother to do so unless the utility charges me below average rates to do so. I calculate that I am earning several hundred dollars per hour for the time spent taking advantage of TOU rates.

As for selling the data... the solution to that is banning such sales, not banning smart meters.


If a smart meter gets hacked it's not unreasonable to imagine the local logs are compromised. This is a bit like all the arguments against voting machines but in a less concerning setting.

TOU rates are a thing you can get with pre-programmed meters. They may not benefit the utility company as much as tailored rates but they probably have 90% of the benefit while having 0% of the privacy implications.

The companies don't even have to sell the data, they can just mine it for information, such as which rate to automatically put you on once your contract finishes to make the most money out of you etc.


I just don't want the timing of my electricity consumption to end up being used as evidence against me for growing cannabis. Feels pretty rational from my perspective.


Or, have them raid your house on the suspicion of growing MJ, when you’re doing a crypto coin (or something similar; folding proteins?)


In Australia, in the state of Victoria all meters are smart meter for few years now. They are also growing in number in other states. There has been no hacking incident. in fact it makes peoples life easier by allowing them to track energy usage at every 15 mins interval with historical data using an app from the utility company. This data cannot be sold either. In fact under the Govt. Open API scheme very shortly you will be able to give access to your own data for comparison to select the best plan for you (just like open banking).


So sounds like Australian smart meter companies are better at security than British ones.

The energy usage tracking can be done without smart meters, in this country electricity companies (and lots of private companies) offered induction clamp based electrical usage logging devices. These may not have been quite as accurate as onboard measuring but this could have easily been solved with some kind of serial protocol exposed on the meter which a third party datalogger could attach to. The ability to track energy usage is not a feature of a smart meter, it's just a feature of having access to the meter's data, this data could always have been made available even if the meter wasn't networked.

Open banking is a complete disaster that I seriously don't think deserves the name "open". I still don't understand how an API which requires you to be a BANK to be able to interact with can remotely claim to be open but having tested some of the implementation for banks it's some horrific over-engineered mess.

Let's hope the data access API for your meter doesn't require you to be an electricity company to access it. As it stands, in the UK, meters are not intercompatible between utility companies so if you switch providers (which I do annually) the old smart meter just becomes a dumb meter again.


Arch-TK I absolutely agree with everything you are saying, and I don’t intend to get a smart meter myself for as long as possible.

However, you are incorrect that meters are incompatible between utility companies. You are right that SMETS1 meters _are_ incompatible. However, all new meter installations are SMETS2 and these are fully compatible between energy companies.

SMETS2 has been the standard for a number of years now. There are still old SMETS1 installations still active though.


I'd like to point out that ever since SMETS2 new (and some firmware updated) smart meters are compatible in the UK, although I do acknowledge they didn't used to be.


The big benefit imo is the load smoothing; statewide, power is cheaper and cleaner than it would otherwise have been.

Right now it’s factories and a few early adopters like me, but anyone can sign up for it and it’s substantially cheaper assuming you don’t mind turning things off at peak times.


I don't trust private companies doing anything "smart".

I say this as an electronics and software engineer. Companies doing have our best interests in mind.

Want to refute that claim? Show me the source code then


And, even if they do show you "the source code," how can you be sure it's the code that's actually running on the device?


What sort of horrible privacy issues do you suppose your smart meter has? You already tell your utility how much power you use.


Data aggregated per month is very different from data aggregated per hour or per minute. You can infer far more personal information from the latter.


No you can’t. They don’t know if you have generation or battery capabilities, even if they detect generation capabilities they don’t know how much.

With that in mind what could they “infer”?

I go weeks without triggering a single bit of usage on my meter. I bet you they aren’t thinking: this guy is mining heaps of crypto.


Power consumption correlated with commercial breaks tells you what show somebody is watching. Power consumption correlated with 9-5 tells you if somebody is working from home. Power consumption correlated with a specific time in the morning tells you when somebody wakes up and subsequently turns the heat on. Lower power consumption over several days tells you when somebody is on vacation.

Are these relatively minor invasions of privacy compared to what advertising companies perform? Yes. But that's no reason to pretend that they aren't privacy-hostile moves on their own.


This is conspiracy theory territory. You <100wh tv isn’t going to show a anything on your power meter during a commercial break.

Thermostat heating ruin your wake up time theories.

And you avoided my actual point: a battery and solar/wind hides all of this,


Yes, they absolutely can. Like with any surveillance technology, there are things you can do to obfuscate your patterns, but that doesn't mean that a broad rollout of the technology won't have a negative privacy impact on most customers.


It’s not obfuscation, if you are concerned about privacy. You can completely hide your usage patterns with a battery or solar/wind generation.

Are you honesty making the point that: knowing you used 5 units of power in 1 month (where someone has to walk onto your property and read a number, as is the case in most dumb metered scenarios) is less of a privacy concern than knowing you used .3 of unit of power in the last 15 minutes (without needing to walk onto your property).

What am I missing?


> knowing you used 5 units of power in 1 month (where someone has to walk onto your property and read a number, as is the case in most dumb metered scenarios) is less of a privacy concern than knowing you used .3 of unit of power in the last 15 minutes (without needing to walk onto your property).

Personally, I think so for most people. But that depends on how much privacy your property provides from pedestrians and where your meter is located.

However, that is orthogonal to the debate since there are other options. Some places allow self reporting and Automated Meter Reading can be done without a smart meter that reports live power usage.

I am unsure why you are so vested in arguing that nobody has a legitimate reason to be concerned about this. It is fine if it doesn't bother you, but it is really necessary to paint those with different concerns as irrational?


no, it’s always worth pointing out irrational reasoning though.

There are clear benefits to real-time monitoring of power. So far I’ve only heard made up, theoretical, what if, privacy concerns.


> no, it’s always worth pointing out irrational reasoning though.

You haven't done that though. You've made accusations of irrationality without ever once backing them up.

> So far I’ve only heard made up, theoretical, what if, privacy concerns.

I think engineers have am ethical duty to consider "theoretical" privacy concerns when they are working on new technology.

Given the history of data mining and brokerage in the US, I would be willing to make significant bets that, (in the absence of a new law being passed to prohibit it,) consumer power usage patterns from smart meters will be sold to data brokers. Pretending this isn't going to happen belies either your knowledge of the existing markets for data or your own rationality.


Irrational premise: smart meters have privacy concerns over normal meters.

Me: how so, there is way too much noise (generation and storage) to get signal (usage correlation). In real life smart meters allow for all sorts of benefits/innovation (time of use charges, time off user feed in)

You: you haven’t said anything, just wild accusations of irrationality. Privacy is a concern, data harvesting!!! Pretending this isn’t going to happen is crazy.

Wtf? Smart meters exist, and this hasn’t happened. Show me one instance. Time of use charges are a thing, look them up. Using electrical usage data to gather private information is not a thing. Look that up too.


I think it's only a matter of degree. And at every level someone can complain. So where do you draw the line?

Watching a meter spin or reading it once a month you can tell if someone is on vacation. Isn't that equally private and personal information?


It is a matter of degree, but that degree is not small. Anytime you decrease the interval, you need to justify the commensurate loss of privacy. You can't just handwave away these concerns like posters in this thread are doing.

A rough inference of which months might involve vacations (data about which is probably already being sold from other sources) is far less invasive than a daily record of your sleep cycle.

With the lack of privacy laws in the US, it is pretty much a given that this data will be sold as soon as the private utility companies in the US start collecting it.


...You say from your personal wiretapping device.


Which I have the ability to choose when/if I am survieled by by leaving it at home or throwing it in a river.


You can turn your power off too whenever you like.


You can also bill users for peak usage times when electricity is expensive or requires falling back on non-renewable resources for production with hourly data.


I can think of a number of ways to do that that preserve privacy far better than real-time reporting of power usage.


How can you charge time of use without collecting time of use data?


Do you carry a mobile phone. If so, you have bigger privacy problems than how much power you use per minute.


That is a trade off that consumers should have the ability to evaluate and decide for themselves.


When it comes to smart meter security and privacy (and perhaps in this day and age diversity and equity) concerns let's discuss them when there's an evidence that they have caused problems.


There's not that many things about the smart meter that are much worse than the vulnerabilities of the plain old spinning disk meter. There were many problems with old meters too. And the benefits far outweigh those issues. Smart meters are not being hacked left and right.

And your privacy concerns are just a matter of granularity of time. You report your usage monthly -- that is also private information. Smart meters just do it on a finer timescale. Not a fundamental difference.

Anyway, back to the main topic.


The vulnerabilities of the plain old spinning disk meter may have been bad but they couldn't be exploited remotely from someone else's house.

Yes, granularity of the measurements IS a problem. If these things only reported the readings when I pressed a button, I would not be so concerned about privacy (that is if the companies could prove to me that the meters did not report the readings outside of these times).




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: