YHK - yet another hacker news clone

Who killed the EV cert? A browser whodunit for Halloween

13 points roddylindsay 3 days ago 25 comments

aleclindz 3 days ago | parent

Spooky!

rdl 1 hour ago | parent

I had no idea EU had mandated this from software developers/browser authors. I hope they can kindly go fuck themselves w.r.t. non-EU software not sold in EU (but merely available online for download by anyone).

9dev 1 hour ago | parent

It’s so funny how Americans seem to assume it’s the natural order of things that the whole world has to comply with American regulation, but lo and behold if the EU dares to have laws of its own!

ameliaquining 47 minutes ago | parent

Do you have any reason to believe that your interlocutor is in favor of the U.S. government enforcing regulations beyond its borders?

9dev 35 minutes ago | parent

Yes. The entire internet is US-shaped. From social networks applying ridiculous puritan beliefs globally to the madness that is the TLD system, DMCA, the CLOUD act… all the way to the NSA spying on non-US citizens to protect themselves.

A Puerto-Rican crypto bro wouldn’t have a career without the USA exerting their will and regulations onto the rest of the world.

ameliaquining 24 minutes ago | parent

I think most American tech people are against most of those things? Why would they not have careers without them?

mieko 1 hour ago | parent

For nearly a decade, I've been occasionally tracking how many of the Cloudflare top-2K sites (earlier: Alexa) respond with an EV cert. And then also pull the SSL Labs score to see if there was any correlation.

If anyone is interested:

https://github.com/mieko/greenbar

I should probably chart out the changes throughout the years, but this was more of a curiosity hack.

sekinfo 56 minutes ago | parent

Great work, did a similar analysis recently but only for the top 50, and in fact none of the top 50 had an EV at the time of the survey.

I believe the only possible advantage of an EV is not a technical one, but rather an organizational one: it encourages the use of a unique keypair on each server because Wildcard coverage is not allowed (although multi-domain is allowed). For what appear to be non-technical reasons, I have seen many entities copying their keypairs for Wildcard certs between servers, even between business critical on-prem backends and cheap shared hosting or budget VPS, or handing the keypair over to a contractor via email.

galadran 1 hour ago | parent

Clarification: Browsers or other user-agents are not required to trust QWACs for use in TLS. We fought very hard for this.

However, there are obligations to display 2-QWACs if sites choose to use them. Their keys are not used to authenticate web domains or encrypt web traffic.

Disclosure: I was Mozilla's lead for this effort and also helped design 2-QWACs (for my sins).

roddylindsay 1 hour ago | parent

2-QWACs are awesome! Thank you for your contributions!

Though I don't understand why they would atone for any sins. IMHO, you created the website identity infrastructure for the agentic web...

And yes, I hope I didn't imply that QWACs are trusted by default for TLS. 1-QWACs would have to come from an existing trusted CA to be recognized by the browser.

galadran 53 minutes ago | parent

Ah, the sins are more to do with the design of 2-QWACs.

All the design needed was for the browser to check the TLS cert and the QWAC cert covered the same domain name. Instead, folks at ETSI insisted very stubbornly on the signature from the QWAC over the cert used in TLS - which is an annoying complication for site operators - and provides little additional benefit.

I'd also like to see the EU encode more information in these certs. Knowing a company's address is not very useful (IMO), but a machine readable label that the site is regulated by $COUNTRY_NAME financial or medical regulator would be a much more useful signal.

Good luck in your endeavours!

roddylindsay 41 minutes ago | parent

Very interesting history, thank you for sharing. I had wondered about the design choices there.

Agree that the next step for 2-QWACs is additional corporate data (professional licensure like you mentioned, incorporation date) and eventually arbitrary claims vetted by a QTSP.

For an average human user, these additional claims are nearly useless since they're so buried in the browser UI...and even if they were more prominent, they wouldn't be of much benefit, as the EV history has shown.

But for an agent that knows to look out for these signals, extensible 2-QWACs will be extremely valuable.

ameliaquining 23 minutes ago | parent

Why do agents need any of this cryptographic infrastructure? Can't they just check on the certifying authority's website whether an entity is legitimate?

roddylindsay 14 minutes ago | parent

It's not so simple:

- If by certifying authority you mean a CA/Trust Service Provider, they don't generally publish client information in a central location.

- If you mean the actual jurisdictional authority, there are thousands and public records are all over the place.

- Websites aren't generally part of corporate filing records. So the agent would still need a way to connect the website to the corporation even after checking the legitimacy of the corporation.

- Serving the structured info in an HTTP header that is cryptographically verifiable by the browser itself (via its trusted roots) essentially serves as a distributed database for website identity, which eliminates the need for each agent to come up with its own ad-hoc verification system on the fly (and all the additional fetches that come along with it)

ameliaquining 6 minutes ago | parent

Can you give a concrete example of the kind of claim you think QWACs would be useful for verifying, that can't be addressed just as well with existing protocols?

michaelt 1 hour ago | parent

> The Stripe we are all familiar was incorporated in Delaware, so there was not a collision. But both companies were described by a browser as “Stripe, Inc.”

As I recall, EV certificates also had the 'problem' that a lot of companies trade under names that aren't their full legal name, making the EV cert confusing. To use modern examples, twitter.com having an EV cert for "Space Exploration Technologies Corp", google.com having an EV cert for "Alphabet Inc" and suchlike.

sekinfo 1 hour ago | parent

Question for HN regarding EV certs:

Have seen some entities request a large number of EV certs (more than 10, sometimes up to 40) for subdomains under a single domain. At the same time, they are requesting OV Wildcard certificates for the main domain, which have overlapping coverage with the EV certs. These entities do not have any regulatory requirements to use EV as far as I know.

Is there any point in doing this?

The overlap from the Wildcard would seem to nullify any theoretical advantage of the EV cert, given that both will be recognized as valid certs that can cover the subdomains. This in addition to the fact that browsers do not treat EV, OV and DV certs differently in any significant user-facing way since many years now.

Frequently these entities also use Let's Encrypt DV certs on some subdomains alongside their EV certs, so of course a CAA record cannot be used to limit emissions to the CA which emitted the EVs.

Can anyone think of a legitimate technical motivation to use EV certs alongside an overlapping Wildcard OV and various DVs?

orf 53 minutes ago | parent

anyone who has ever worked at a large non-technical company can tell you exactly where these requests came from

rgbrenner 50 minutes ago | parent

I remember when these rolled out.. it was just an obvious cash grab by cert vendors, trying to create a trust-badge in collusion with browsers.

Back then, cert vendors would try to sell you on "premium" certificates.. telling you that cheap ssl certificates weren't good enough, so you needed to pay for $100-300+/yr certificate, and you could show to your visitors by displaying silly trust seals like the "Thwarte Seal".. as if customers would know who the hell Thwarte is.

Then this EV seal was launched, displaying what was essentially a trust seal in the browser.. and these same vendors would charge $500+/yr for it.

I can't express how great it is that EV certs died.

ameliaquining 42 minutes ago | parent

The browser vendors, I think, were at least making a good-faith attempt to address a real security problem; it's not like they made any money from EV certificates. Even if it was inevitably going to be marketed somewhat predatorily, and ultimately didn't work out.

logartis 49 minutes ago | parent

Keeping identity separate from TLS makes a lot of sense to me!

ipdashc 30 minutes ago | parent

I recognize that they had a lot of issues and I'm glad the implementation of EV certs that we had died out, but I've always thought the concept itself wasn't a bad one. For stuff like banking or government websites, it would be nice to have them tied to a legal entity instead of purely just a domain name.

In theory we could even limit it so that only a given country's CAs can sign entities under that country (somewhat mitigating the whole eMudhra thing...)

It does feel like kind of a shame sometimes, there's a lot of stuff that would be possible with TLS/certs/webpki that we don't really explore. Mutual TLS, for instance. I get why, and I'm as happy as anyone that Let's Encrypt exists, but sometimes it feels unfortunate that we've settled into the comfort zone of "get a domain name, pull an LE cert for it" and don't often do much else with the technology and all the CA infrastructure.

ameliaquining 8 minutes ago | parent

Do you have any particular reason to think that the Stripe problem is solvable?

ccTLD name constraints are an old idea, back in the news again lately for obvious reasons, and the same arguments against them still apply: https://groups.google.com/g/mozilla.dev.security.policy/c/pF...

Are there other use cases besides those that you think the Web PKI should support?

Ayesh 6 minutes ago | parent

There are discussion in CA/B forums for root and intermediate CAs with technically scoped issue policies.

For example, a government could be an intermediate CA, and have it technically limited only to issue certificates for `*.gov.CCTLD` domains.