Hacker News Reader

Comment by noonespecial | Hacker News Reader

The implications are troubling. Your TV collects and broadcasts for the permanent record of anyone who can snoop the cleartext (your neighbors, your ISP, whatever the NSA looks like in your country, etc) all the media it can find on your network.

We used to need firewalls at the edges of our home networks to keep bad actors out. Now we need firewalls that point the other direction to keep the bad actors on our networks in.

View on HN

A good friend of mine did exactly that for a network some 15 years ago. He called it his "waterwall".

It was for some kind of internet-cafe. But that we seem to need this nowadays for our private homes, troubles me.

What is this with the attitude, that it is ok, to view into the innermost privacy-sphere of your customers? Sorry, but LG is dead and gone for me. A company, that converts its paying customers into a product, ready to be sold to some advertiser just sickens me.

What comes next? Automatically detecting, when people no longer look at the screen, via a camera, to stop the showing of ads and resume, when people return in front of the screen? Or to detect how many people are watching (as Microsoft had patented)?

What happened to good old: We build one thing, sell it and that is fine. Why does everybody have to be an advertising-company, trying to maximize this revenue stream on the back of the live(data) of its paying customers?

I suspect that this is the result of incompetence on their part rather than an actual intent to log private data. Either way though, obviously not good.

I do understand, that most seemingly malicious intend has its roots in incompetence.

In an answer to German press-inquiries, LG states, that the feature will stay. The only thing, they will change is, that disabling the feature really disables it. They stated, that this was a bug. And I actually believe this. I did not allege, it was malicious intend.

My real argument was, that I am more then happy, to pay good money for a good TV. But that I see no reason, why I should have to pay with my data as well (not being informed on that by the way, when buying this TV) and with this feature being enabled by default.

Sidenote: They told the German press, that the receiving server dropped the information, when the feature was disabled. They never told anyone, how the server should have been able to do that. As far, as I could see, the requests did not change, when deactivating that feature.

Hi! Friendly native English speaker here. Just a tip, your posts read as very disjointed because you are using the comma so often! You can think of the comma as representing a pause in speech. Re-read your post, taking a pause every time you encounter a comma and you'll see how disjointed it feels. You are making good points, but I found your posts difficult to read because of this issue.

Thanks a lot for your tip. I know that the comma is some form of tick for me. Will try to better myself.

Greetings from Germany.

"Any sufficiently advanced incompetence is indistinguishable from malice"

I'm definitely going to have to remember that one for later.

> why I should have to pay with my data as well (not being informed on that by the way, when buying this TV)

You were probably informed, somewhere between page 1 and 514 in the EULA.

Nah. Consumers are meat. Product to be packaged and sold in bulk. Nothing personal, but the corporate masters just don't care that much.

Good firewalls should always block unwanted traffic whether it's inbound or outbound. This is not anything new, but you're right that most people trust the devices they own. Not anymore.

Good luck making that easy for the end-user to configure.

I think this is something of a chicken and egg problem. The basic firewall UI is allow <computer> to use <port> for <incoming|outgoing>. This interface can be made very easy to use. The problem is that if this was the default then programs would simply not work, and users would not know why.

However, if it became standard, then programs would tell users to check their firewall, and can even tell users the information that they need to fill into the firewall (eg. name of computer, port, ...).

Conceivably, you could define an protocol for devices to request an opening. This protocol could include a way of sending the password.

We have to assume that our lives are public; and that we have no secrets from those who want to get them the most (i.e. bad actors). We are vulnerable and weak, and had better get used to the idea that we can be f*cked over at will by whosoever takes a fancy to the idea.

Yep you've basically refuted the idea of personal security due to a vague insinuation of how "most people" trust their electronics. Or did you have anything constructive to add?

Sorry, I should have said:

"Historically it's been very difficult to make detailed firewall configuration user-friendly. As a result, we've lived with the simplistic compromise of 'allow outgoing connections, don't allow incoming connections'. It seems like it's going to require some incredible new firewall configuration interface if we hope to make it possible for the ordinary clueless computer user to properly configure outgoing firewall rules."

I want to try an experiment that goes like this:

1. Block all network traffic, inbound and outbound.

2. When I decide that I need a specific network resource, enable it in a whitelist (i.e. enable traffic to/from news.ycombinator.com on port 80)

I wonder what that Internet would look like.

RFC 3514 makes this easy:

http://www.ietf.org/rfc/rfc3514.txt

They'll just switch to port 80. No one is going to be willing to block that outgoing.

You know firewalls work based on port AND host, right? Block all connections coming from the TV.

If you have a smart TV you probably want to browse the web on it.

Or boycott LG and any businesses that acts in this manor.

Hey man; it's manner, not manor. Manor is more akin to a large house or mansion.

I am often hesitant to offer this kind of correction because I feel like it sometimes sounds conceited, or as if I am trying to be superior to you somehow. I honestly don't feel that way and am just offering the correction because I feel like I would like to be corrected myself. I respect you for having learned a second language, it's more than I can say for myself.

You'll find that people that have english as their first language tend to make that kind of mistakes more often than people who have it as a second language. In my case, for example, my first language is spanish, where everything sounds as it's written. When I see mistakes like writing "manor" instead of "manner", or "should of" instead of "should have" I can't help but think that I would never make that kind of mistake, because of the way people learn english as a second language in comparison to a first language.

This might be also because you learned English when you already knew how to read, which isn't the case for people learning it as a first language.

You know "manor" is an apropos typo, signifying LG being haughty, privileged and out-of-touch living in this really expensive house they built with the money from paying customers that they are treating like cyber-peasants.

I wish I knew a name for a homophonic typo which suggests criticism in that way, I used to know a guy who made typos like that all the time, emails from him had a kind of surreal meta-level quality to them. It is kind of a mondegreen, but not quite.

https://en.wikipedia.org/wiki/Mondegreen

eggcorn? http://eggcorns.lascribe.net/

Example

> When all is set and done

Unless he lives in a manor, then it works actually.

But my witty jokes aside, that reminds me of Little Snitch for Mac OS, which keeps installed programs from opening unauthorized connections to the outside. Is there a firewall setup that works this way for an entire home network? Possibly something that could be run on a small device/router?

That's just called a firewall. You will probably enjoy http://www.pfsense.org/

How do you know it wasn't manure what he meant? :-)

I think it's just a marvelous pun!

Although manor is oddly apt, don't dare act up in my manor!

Or if you're in the EU, where this sort of thing is quite likely to be illegal, complain to your local Data Protection Commissioner.

> Now we need firewalls that point the other direction to keep the bad actors on our networks in.

That's not recent, most firewall do outbound traffic as well, and there are software solely dedicated to outbound traffic management, e.g. http://www.obdev.at/products/littlesnitch/index.html

The real difference is not between outbound and inbound. Classical firewalls have been blocking outbound connections depending on ports since forever.

Firewalls for Windows (I first saw it in ZoneAlarm) introduced application specific filtering. So you can allow one application to connect to HTTP servers and disallow another application from doing the same.

Edit: I was just reading this article an hour ago, about how to programmatically add rules to Windows built-in firewall using Delphi:

http://theroadtodelphi.wordpress.com/2013/11/21/using-the-wi...

Reply on news.ycombinator.com