Hacker News Reader

Comment by coldpie | Hacker News Reader

As has been demonstrated many, many (many, many (many many many many many...)) times: there is no such thing as computer security. If you have data on a computer that is connected to the Internet, you should consider that data semi-public. If you put data on someone else's computer, you should consider that data fully public.

Our computer security analogies are modeled around securing a home from burglars, but the actual threat model is the ocean surging 30 feet onto our beachfront community. The ocean will find the holes, no matter how small. We are not prepared for this.

View on HN

> As has been demonstrated many, many (many, many (many many many many many...)) times: there is no such thing as computer security.

Of course there is, and things are only getting more secure. Just because a lot of insecurity exists doesn't mean computer security isn't possible.

It's a matter of opinion, but no, I disagree. People are building new software all the time. It all has bugs. It will always have bugs. The only way to build secure software is to increase its cost by a factor of 100 or more (think medical and aviation software). No one is going to accept that.

Computer security is impossible at the prices we can afford. That doesn't mean we can't use computers, but it does mean we need to assess the threats appropriately. I don't think most people do.

It's not a matter of opinion at all. You can disagree but you can disagree with the earth being a sphere also.

> People are building new software all the time. It all has bugs. It will always have bugs.

No. Most bugs these days are due to legacy decisions where security was not an issue. We are making advances in both chip and software security. Things are already vastly more secure than they were 20 years ago.

20 years from now, security will be a lot closer to being a solved problem.

> The only way to build secure software is to increase its cost by a factor of 100 or more (think medical and aviation software). No one is going to accept that.

What are you basing that cost on?

> Computer security is impossible at the prices we can afford.

No, it really isn't. There's a reason some organizations have never been hacked and likely never will be. Largely because they have competent people implementing security that very much exists.

> Our computer security analogies are modeled around securing a home from burglars

Well, no home is burglar-proof either. Just like with computer security, we define , often just implicitly, a threat model and then we decide which kind of security measures we use to protect our homes. But a determined burglar could still find a way in. And here we get to a classic security consideration: if the effort required to break your security is greater than the benefit obtained from doing so, you're adequately protected from most threats.

I agree, my point is we need to be using the correct threat model when thinking about those risks. You might feel comfortable storing your unreplaceable valuables in a house that is reasonably secure against burglars, even if it's not perfectly secure. But you'd feel otherwise about an oceanfront property regularly facing 30 foot storm surges. I'm saying the latter is the correct frame of mind to be in when thinking about whether to put data onto an Internet-connected computer.

It's no huge loss if the sea takes all the cat photos off my phone. But if you're a hospital or civil services admin hooking up your operation to the Internet, you gotta be prepared for it all to go out to sea one day, because it will. Is that worth the gains?

And I think there's some cognitive problem that prevents people from understanding that "the effort required to break your security" has been rapidly trending towards zero. This makes the equation effectively useless.

(Possibly even negative, when people go out and deliberately install apps that, by backdoor or by design, hoover up their data, etc. And when the mainstream OSes are disincentivized to prevent this because it's their business model too.)

There was a time, not very long ago, when I could just tcpdump my cable-modem interface and know what every single packet was. The occasional scan or probe stuck out like a sore thumb. Today I'd be drinking from such a firehose of scans I don't even have words for it. It's not even beachfront property, we live in a damn submarine.

by this logic, every picture you'll ever take with your phone would be considered semi-public as phones are Internet connected.

While I wouldn't have too much of an issue with that, I'm pretty sure I'm a minority with that

> every picture you'll ever take with your phone would be considered semi-public as phones are Internet connected

Correct.

https://en.wikipedia.org/wiki/2014_celebrity_nude_photo_leak

https://www.cybersecurity-insiders.com/glitch-makes-data-fro...

https://arstechnica.com/gadgets/2023/09/apple-patches-clickl...

Do you use a bank account? Or do you still trade using only the shells you can carry in your arms? Perhaps networked computers are secure enough to be useful after all.

I never claimed the Internet isn't useful. I just think people don't recognize how vulnerable computers are to attack. Search this very incomplete list for "bank": https://en.wikipedia.org/wiki/List_of_data_breaches

Reply on news.ycombinator.com