Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What If We Put Warnings on IoT Devices?

troyhunt.com

Post reply on HN
Print view View original

111–120 of 159 posts

Re: What If We Put Warnings on IoT Devices?

#111

Post by thinkling »

TremendousJudge wrote: post #54
npsimons wrote: post #17

Earlier quoted context omitted.

> This feels kind of like that. It might feel like that, but it isn't. Everything[1] causes cancer. Not everything needs to have an Internet connection to a service to function, not essentially at least. It's only these people looking to monetize or "h0ok all teh th7ngs up!!1!eleven!" that are shoveling crap that no one asked for, but everyone is too lazy to object to. I've got a pair of wireless earbuds, completely…

Right now I'm writing my bachelor's thesis on IoT platforms. I think the real problem is that there's no real established pattern for authentication (the way there is for things like signing up for a web app). You ship devices and there isn't a 'one true way' for authenticating them against your platform, authenticating the user, partnering the user with the device, and then encrypting all traffic from the device to…

Another thing about IoT gadgets and "smart"stuff in general, is the guaranteed obsolescence. It's not literally planned obsolescense, but when you install that wi-fi smartlock in your door, you know that it 5 years the company will stop updating the app that goes with it, and it won't work anymore under iOS 16, and so you'll have to throw it out and get a new lock, a new dimmer, a new garage door opener, new audio system, you name it.

Re: What If We Put Warnings on IoT Devices?

#112

Post by thinkling »

teekert wrote: post #98
FabHK wrote: post #34

Earlier quoted context omitted.

Yes. There's this doorbell (400-1500 USD) that connects via internet to a central host, and then notifies you on your smartphone. Seems sort of a neat idea, but a house lasts several decades - is that startup and its server going to be around that long? http://www.doorbird.com

Well, a doorbell is not exactly a part of the foundation and it is easily replaced. My house from the 1930's has seen several doorbells already.

This isn't limited to the doorbell. It's the doorbell, the stereo system (Sonos), the connected garage door, the alarm, the security cameras, the thermostat, ... they're all dependent on their dedicated app.

When the vendor stops updating the app that goes with a 5-yo product that they no longer sell, at some point that app will no longer work under the new version of your mobile OS (iOS 16? 17?), and you'll have to give up on it.

There's a huge amount of obsolescence coming down the pike in 5-10 years.

Re: What If We Put Warnings on IoT Devices?

#113

Post by taneq »

freehunter wrote: post #61
seanmcdirmid wrote: post #44

Earlier quoted context omitted.

Even if they did, would it matter? It would be some chemical term, like the one you see in an ingedient list, that is meaningless to most people. And these are places, not products...with gardens. Even if they use organic pesticides, they probably still have to put up the warning.

>And these are places, not products Those places have products that contain the toxic chemical, An apartment building is not poisonous by itself, they contain products that contain the harmful chemical. That's what they should be warning you about.

Depends what it's made of. An asbestos apartment building is kinda toxic.

Re: What If We Put Warnings on IoT Devices?

#114

Post by kaybe »

seanmcdirmid wrote: post #75
freehunter wrote: post #66

Earlier quoted context omitted.

How about a warning that says "The battery in this cell phone contains an amount of theomacrotasium known to cause cancer if ingested." That's a completely actionable warning. Don't eat the contents of the battery of your cell phone. Or how about "This carpet contains formeldahyde, which is known to cause birth defects in pregnant women. Limit exposure to this carpet for a few days after installation." That's a good…

Hey, if it's just one thing, great! Now, how do you do that for a hundred things? Again, warnings work great for very small numbers of N. We have to find a middle ground between "has chemicals that may cause cancer" and "a hundred specific chemicals in X quantities that may cause cancer." Abstraction is very necessary as a pragmatic solution, even if full disclosure makes sense as an ideal. Heck, most places are just…

Well, information on how to migitate the cancer risk is already a step forward.

If it is only harmful if ingested you know not to give it to your toddler. If it gives off toxic fumes while burning, well, stay away if you made a fire mistake. etc. The warning itself is pretty useless as it is.

Re: What If We Put Warnings on IoT Devices?

#115

Post by DarronWyke »

0xfffff wrote: post #43

I read about proposal to prominently put expiration dates IoT devices, to show how long it will be servered with security updates. That could be interesting and also raise awareness.

Not really. Don't forget that for the longest time home routers shipped broadcasting the default SSID with no password. Now they come preconfigured with a locked-down SSID and password (though the management interface is still usually behind admin/admin or an equivalent). Security isn't on most people's minds -- it has to be built into the product from the get-go to get the largest reach.

Re: What If We Put Warnings on IoT Devices?

#117

Post by TeMPOraL »

notauser wrote: post #101
logfromblammo wrote: post #69

Earlier quoted context omitted.

It would be much more useful if it were in the form of "This property contains substances determined by the state of California to be toxic to humans. The site report may be viewed in person at 123 Maple St, Suite 200, San Mateo, or electronically at https: //sitereport.ca.us/ " Then the print report would basically be a binder full of MSDS-like information sheets, along with the history of measurements recorded on t…

This is the same problem with the cookie law in Europe. They allowed one single, generic, disclaimer which every site pops up. If they'd demanded: - a separate disclaimer for each domain (or at least company) setting a cookie - a description of the purpose of the cookies (e.g. advertising, remembering log-ins) Then the law might have actually achieved something.

TBH, you need to put a cookie warning only if you use cookies for stuff like tracking and advertising. Logins and general site functionality are extempt from that.

Basically, cookie warning on site means the site tracks you.

Re: What If We Put Warnings on IoT Devices?

#118

Post by Mister_Snuggles »

FabHK wrote: post #34
EvanAnderson wrote: post #4

I'm of the opinion that products which require a separate service to perform their advertised functions (i.e. a "cloud" service-- be it "free" with the product or subscription-based) should be clearly labeled as such. I know that I don't actually own anything that I can't self-host (or pay whoever I want to host it), but it's clear that most people don't. Public education on this front seems valuable to me (but, then…

Yes. There's this doorbell (400-1500 USD) that connects via internet to a central host, and then notifies you on your smartphone. Seems sort of a neat idea, but a house lasts several decades - is that startup and its server going to be around that long? http://www.doorbird.com

This is exactly why I haven't picked up a video doorbell doorbell yet.

Currently I have a bunch of cheap WiFi cameras, all made by D-Link. The thing I like about D-Link cameras is that when you put them on a WiFi network that isn't allowed to access the internet they still work as expected.

The cameras themselves have all of the smarts to do motion detection and email short recordings, they can also FTP images somewhere or you can use something like Zoneminder to handle things. None of this requires them to phone home or make use of some kind of cloud service.

Of course, D-Link does have a cloud service, but it's not required. To me it seems like the best of both worlds - the easy-to-use cloud service for those who want it, plus the ability to work standalone for those who don't want to use their service.

Why doesn't someone build a video doorbell that works like this? D-Link keeps making and selling these cameras, so clearly there's some kind of market for products that have a cloud service but don't require it.

Re: What If We Put Warnings on IoT Devices?

#119

Post by seanmcdirmid »

taneq wrote: post #113
freehunter wrote: post #61

Earlier quoted context omitted.

>And these are places, not products Those places have products that contain the toxic chemical, An apartment building is not poisonous by itself, they contain products that contain the harmful chemical. That's what they should be warning you about.

Depends what it's made of. An asbestos apartment building is kinda toxic.

Sure. Also, no one on china trusts each others apartment renovations for this reason, which is why you are expected to re-renovate every time you buy (and your own renovations do nothing for the resale value).

That stuff is actually easy to track down is what I meant. It is the day to day stuff that isn't.

Re: What If We Put Warnings on IoT Devices?

#120

Post by TeMPOraL »

oneweekwonder wrote: post #29
downer72 wrote: post #9

Earlier quoted context omitted.

For sure, that’s what killed the Pebble watch for me. I got it home, opened the box, thinking I had a new watch! Nope. You have to connect it to the internet and let it phone home first. It would not tell time, permit you to manually set the time, or let you explore that watch at all, until it spoke with it’s mothership. I put it back in the box, and tossed it in the closet, and it’s still there, and now it’s complet…

If it is in a mint condition with boxing, hold on to it, in a decade or so I believe it would be worth a lot.

In a decade the hardware may fail (yay today's manufacturing practices). But in 2-3 years, most Pebble fans will probably have broken their current watch, and will be desperately looking for replacement. Since among smartwatches, Pebble was the single least shitty option (by a wide margin) and no company is stepping up with a comparable alternative, people might pay good money to get a working one. I know I probably will.
Post reply on HN