The devices we own should be acting in our own best interest; we shouldn't need to treat them as adversaries.
A large part of trust is auditability. And a large part of auditing a device is looking at its communication. What we are kind of running into is the security implications of a debug-interface.
Your point about dissidents needing secure access is a strong one. Any way to MitM a connection for audit purposes can be repurposed into an MitM for surveillance given enough coercion. However, I don't think this applies to a network connection between my TV and Samsung.
Once a surveillor coerces me to give them access, I can disconnect my TV from the internet. The same goes for an amazon echo, or a juice press with WiFi. It is different with whatsapp, facebook and internet banking. These are much more essential, so it is more important that it is hard to MitM them even if the user gives consent.
Sure, but you can also look at the device's communication if you have administrative access to it yourself, rather than trying to MITM it. I suppose this could be a vulnerability itself if done poorly, but users have administrative access to their PCs; why should other devices they own be any different?
>These are much more essential, so it is more important that it is hard to MitM them even if the user gives consent.
If the user gives consent, they should be allowed to inspect the connection. The device's UI can make it _very clear_ that they're about to do something dangerous, but it's their device and should be their choice.
This reminds me of Programming Satan's Computer [1] (PDF link), the first time I understood that while programming is hard, security programming is absolutely insane.
I don't agree. If you let a guest use your WiFi network for instance, there's no inherent moral right for you to intercept their emails.
However I generally agree with the principle that if you own a device, you have the right to learn what it's doing, and the trend towards black boxes is concerning.
You might have purchased a particular device, but you generally don't have the rights to the software or firmware that runs it.
If you're truly concerned, you should support manufacturers that leave their system open by design, or support open source devices, and contact manufacturers describing your concerns.
But the tradeoff wars of freedom v. security were lost long ago. The idea that one should shoulder any blame for choices is lost on a generation that needs YouTube videos to cut an onion. Likewise, a significant number of individuals can't be bothered to change their router password from 123456, and an even greater number of folks think the govt needs to regulate the internet to keep it fair and safe for everyone.. Forced encryption with no choice by design is a feature, not a bug.
Use open source or
True, I don't have a right to see their source code. But decent laws shouldn't prevent me from reverse-engineering the code (DMCA is not a decent law).
Moreover, access to code is orthogonal to control over devices one owns. If I own a device, I should be the one who has ultimate control over it.
Legally this is correct in certain jurisdictions. But I'd say this is similar to arguing that you have no right to know the ingredients in food you buy.
Which is part of why the more paranoid of us steadfastly refuse to own such devices.
My smart TV (I am ashamed to admit I have one) is really useful. Very little about the idea of a TV with build-in Plex support requires it be totally locked down. Hence it is a buisness decision that could be competed or regulated away.
(I was going to make the same argument about a TV with build-in Netflix support, but I believe the DRM requirements of Netflix require quite a bit of lock-down)
Possible items for its agenda:
- Right to Repair
- Right to Tinker
- Right to "Pwn Ur Own"
- Hardware sellers required to deliver firmware source to buyers
--- including build scripts
--- including device-specific signing keys
--- even for cars and tractors
- Public APIs required for public-facing services
- Right to non-backdoored strong encryption
- Formal requirements for custodianship of PII
- Formal requirements for security in IoT devices
- Ban on local monopolies on wired telecoms
- Definitions for terms used in advertising (i.e. "5G")
- Establish a separate USPTO regime specifically for computer programs
The lawyers of the Boomer generation that seem to be running the show now are not very well-suited to understand the near-future problems in the technology field. A lot of these problems are better solved by strongarming companies into behaving more eusocially than by attempting a technical solution or starting disruptive competitor businesses.For awhile I've been wondering if people would like me to be their spokes person for such issues. I have a lot of experience and I think I have a mind for politics as well. I think its really critical we get it right at this point in history. I would love to work with others for such a cause.
If you have any advice to get started I would humbly accept. I think its really critical we get it right at this point in history.
That, or you can buy a commercial display instead of a TV.
Might need to invest in a large Faraday cage
What's more important, is the possibility of having a camera-less and microphone-less smart-tv. Or to have the possibility of practicing a microphonectomy and a camerectomyon your smart-tv.
The last carrier I worked for was running trials of a "virtual CPE" that replaced your home gateway with a much dumber and cheaper device that effectively extended your home LAN to the local exchange (at least), where the actual isolation, filtering and NAT were performed.
I switched home providers when I left there, and if my current provider ever goes the same route I'll drop my own firewall in front of my LAN.
(Theoretically there is already relatively little isolation between your LAN and parts of the carrier network if you have VoIP or IPTV, but in this case I happen to know who tests that equipment and have a very good idea of what it really does, because I used to work in that team a long time ago...)
I'm with you 100%. The reality is, though, your only choice is to not run those devices with access to the internet. A TV should not require internet access to be usable. I won't use an Echo, and IoT devices are isolated to their own internal network without WAN access.
Without those we might as well go back to storing content locally which let's face it - is mostly retrieved quasi-legally through BitTorrent.
But at least it's always available and will not disappear between the moment you first watch it and the moment you want to show it to your spouse.
I really expect (and hope) to see a resurgence of torrenting in the wake of ongoing balkanization of streaming platforms.
Agree. A lot of people think it's still the torrent dark-ages where you need to search for individual torrents like a cave-person. If so google sonarr, radarr and get a vpn!
Still have to trust the manufacture for that to work, but at lease you're not just left wide open there. (I mean you still could be, but it seems less likely with this model). And, if it doesn't stay up todate, you can toss the box and plug in a new one.
Honestly, who cares? I mean, it's a TV. It should not need Internet to work, and it definitely shouldn't need software updates. Displaying video on a screen is a solved problem.
The very fact you can use "software update" and "TV" in the same sentence signals a pretty big problem - a problem of companies selling you TV as a service, and customers accepting half-done pseudo-products.
Don't make the mistake of thinking you are a representative of all customers. YOU may not care about connectivity, but others, many others, do.
As long as the tv has import ports I can use other devices than the TV itself to get content so I'm not really concerned about that.
I bought the TV because it was good at being a TV - that is, it is OLED, high res, good upscaling, etc. The "smart" stuff I could take or leave, but I'm enjoying the convenience of it while it works.
Security isn't about being perfect, so stop pushing that false narrative. It's about being good enough, and plenty of companies making smart TVs can certainly become "good enough".
I prefer to just have a screen and a box though. I cut the wifi out of my TV and only plug it in to update firmware a few times/year.
My whole point in my post is that I don't want devices that I cannot look into and control. Unfortunately that's become incredibly difficult if not impossible. Personally, the line I've drawn does not include my TV. it's not that important to me to have an integrated TV system that I can't control.
If you own a device, and said device is transmitting data from your environment, you should be able to know what information this device is communicating. It is not enough to trust a company privacy policy.
Are there enterprise versions of all these devices that we can buy instead?
You may found the evidence of malicious act after it has happened, but often it's too late.
If Amazon is directly responsible for Amazon Echo's malicious act, you can blame Amazon. But if the attacker was government spy agency, you're out of lack.
Just don't use these devices which cannot be examined thus cannot be trusted.
What worries me is that the full chain (Client-software -- Network -- Server-software) is totally opaque and immutable at the discretion of a vendor. That means we are left at the mercy of the vendors.
I'm hoping for laws like 'Right to Repair' to help with this. The alternative would be to go full Richard Stallman, and I still think that is too radical.
Years ago sticking these types of devices on a trusted home LAN would have been unthinkable.
The best solution to the privacy problem is to do research and exercise consumer choice about what sorts of devices you purchase.
One the one hand, ubiquitous encryption is simply required for security on the internet. Things like lets encrypt and warning on http are great improvements.
On the other hand, the owner of a network has some right to look into the packets on that network. Especially if the owner of the network also owns the end-points of that traffic. My main use-case here isn't corporate networks, snooping there makes me uncomfortable.
Really, my issue is stuff on my own network. I want to see what my TV sends home. Same with an amazon-echo, or really any IoT thing. Yet, if they all use SSL and don't allow me to add a root CA, I can't look at what they run.
A user has no control over an amazon echo. You can't modify the software because the bootloader is locked down. You can't inspect the traffic because it is SSL cert-pinned. Amazon can push updates to it at any time. All a user gets to do is decide whether it is turned on, and whether it gets a network connection.
Really, what I would want to see is the option to install a CA cert on any device I own. At the same time, that is a terrible idea. Every 14 year old with google is going to find some stack-overflow answer that'll tell them to MitM their TV to do some simple thing.