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

10 Years of Let's Encrypt

letsencrypt.org

Post reply on HN
Print view View original

121–130 of 361 posts

Re: 10 Years of Let's Encrypt

#121

Post by 12_throw_away »

ok123456 wrote: post #66

Snowden was the other big reason that TLS became the de facto standard for every site. Prior to that, the consensus was that you only really needed TLS if you were dealing with money and wasn't worth the hassle otherwise. You could sniff traffic from Facebook and Twitter easily. I remember listening to a talk given by an IRS investigator in around 2008 about how they were able to do a sting and shutdown illegal inter…

I think it was a lot earlier than 2013 - SSL was inevitable by the late 2000's, as soon as major ISPs decided they could make more money by injecting ads into http connections (e.g., [1]). It obviously took a while for the infrastructure to scale up ... but I'd imagine that concerns about stolen ad impressions drove a lot more HTTPS adoption than concerns about the NSA.

Re: 10 Years of Let's Encrypt

#122

Post by letsgetreal »

tptacek wrote: post #112
letsgetreal wrote: post #108

Let's Encrypt allows anyone to have secure https communication, sure, but it doesn't address the question of website authenticity. I groan when I'm on an e-commerce site and I click on the browser URL lock icon and see a Let's Encrypt certificate because frankly anyone can create one for no cost and I don't know if it's the real website or if I made a URL typo. Say what you will about the expensive cert providers, bu…

It was never a reasonable goal of the WebPKI to authenticate entities; only to help establish end-to-end encryption between unrelated parties on the Internet. The WebPKI can ensure you're talking to whoever controls `ycombinator.com`, but it has to be up to some other layer of the security stack to decide whether you want to be talking to `ycombinator.com`. (This is in fact part of the logic behind FIDO2 and phishing…

FIDO2 doesn't solve the first website contact trust problem - only the HTTPS certificate does that.

Re: 10 Years of Let's Encrypt

#123

Post by schoen »

tptacek wrote: post #112
letsgetreal wrote: post #108

Let's Encrypt allows anyone to have secure https communication, sure, but it doesn't address the question of website authenticity. I groan when I'm on an e-commerce site and I click on the browser URL lock icon and see a Let's Encrypt certificate because frankly anyone can create one for no cost and I don't know if it's the real website or if I made a URL typo. Say what you will about the expensive cert providers, bu…

It was never a reasonable goal of the WebPKI to authenticate entities; only to help establish end-to-end encryption between unrelated parties on the Internet. The WebPKI can ensure you're talking to whoever controls `ycombinator.com`, but it has to be up to some other layer of the security stack to decide whether you want to be talking to `ycombinator.com`. (This is in fact part of the logic behind FIDO2 and phishing…

> It was never a reasonable goal of the WebPKI to authenticate entities

The confusing thing is that this goal nonetheless appeared in some original marketing and explanations about the web PKI from the late 1990s when it was first introduced. There was another smaller burst of this when people were arguing over the formalization of DV certificates and of Google's UI changes that stopped treating EV specially (as some people found both of those changes objectionable).

I agree with you that the goal of authenticating entities was impractical, but the mental association and expectation around it still hasn't been completely dispelled. (I think I saw some form of this when doing support on the Let's Encrypt Community Forum, as people would sometimes complain that a site shouldn't have been allowed to have a certificate, either because it wasn't the organization they expected, or because it was malicious somehow.)

Re: 10 Years of Let's Encrypt

#124

Post by xandrius »

mikeyouse wrote: post #99
t1234s wrote: post #87

Lets hope they stay independent and never get acquired by Google or any other large tech company. You can imagine a web where SSL issuance is used as a tool to censor websites. I think most browsers have been made to make standard http sites look malicious to normal users.

They're a nonprofit - so they can't be acquired like a typical for-profit company. They could in theory sell some assets but it'd be very convoluted if they were the core assets -- per US tax law, nonprofit assets must remain in the nonprofit world, so there's no risk of any tech company ruining them.

I heard similar things about another American nonprofit and now I'm not so sure about it. When money and will comes along, loopholes come as well.

So, I wouldn't be so sure, unfortunately.

Re: 10 Years of Let's Encrypt

#125

Post by schoen »

sam_lowry_ wrote: post #34

[flagged]

You might want to be more specific about the meaning of "between" here. It's not a cryptographic MITM attack, and if it ever facilitated someone else in performing one, that should be detectable.

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

(It's also true that the level of active monitoring of CT logs has never gotten very high.)

Re: 10 Years of Let's Encrypt

#126

Post by xandrius »

letsgetreal wrote: post #108

Let's Encrypt allows anyone to have secure https communication, sure, but it doesn't address the question of website authenticity. I groan when I'm on an e-commerce site and I click on the browser URL lock icon and see a Let's Encrypt certificate because frankly anyone can create one for no cost and I don't know if it's the real website or if I made a URL typo. Say what you will about the expensive cert providers, bu…

Not really the point of ssl certs though. And I'm pretty sure those limitations are the smallest hurdle, most people wouldn't even care checking.

Re: 10 Years of Let's Encrypt

#127

Post by pedrozieg »

It’s easy to forget how awful TLS was before Let’s Encrypt: you’d pay per-hostname, file tickets, manually validate domains, and then babysit a 1-year cert renewal calendar. Today it’s basically “install an ACME client once and forget it” and the web quietly shifted from The impressive bit isn’t just the crypto, it’s that they attacked the operational problem: automation (ACME), good client ecosystem, and a nonprofit CA that’s fine with being invisible infrastructure. A boring, free cert became the default.

The next 10 years feel harder: shrinking lifetimes (45-day certs are coming) means “click to install cert” can’t exist anymore, and there’s still a huge long tail of internal dashboards, random appliances, and IoT gear that don’t have good automation hooks. We’ve solved “public websites on Linux boxes,” but not “everything else on the network.”

Re: 10 Years of Let's Encrypt

#129

Post by letsgetreal »

xandrius wrote: post #126
letsgetreal wrote: post #108

Let's Encrypt allows anyone to have secure https communication, sure, but it doesn't address the question of website authenticity. I groan when I'm on an e-commerce site and I click on the browser URL lock icon and see a Let's Encrypt certificate because frankly anyone can create one for no cost and I don't know if it's the real website or if I made a URL typo. Say what you will about the expensive cert providers, bu…

Not really the point of ssl certs though. And I'm pretty sure those limitations are the smallest hurdle, most people wouldn't even care checking.

The "most people won't care argument" doesn't inspire confidence in the authenticity of the website.

It's essentially a self-signed cert that anyone could make with the false security of a root certificate authority.

Re: 10 Years of Let's Encrypt

#130

Post by duskwuff »

realityking wrote: post #89
RonanSoleste wrote: post #59

Earlier quoted context omitted.

When you request an EV. They call you by the phone number that you give to ask if you requested a certificate. That was the complete extend of the validation. I could be a scammer with a specificity designed domain name and they would just accept it, no questions asked.

EV certs also showed the legal name of the company that requested the certificate - that was an advantage.

Which would have made sense if company names were unique - which they aren't. See e.g. https://groups.google.com/g/mozilla.dev.security.policy/c/Nj... for an example of how this was abused.
Post reply on HN