Podcast
Root Causes 321: CABF Moratorium on New Certificate Consumer Members


Hosted by
Tim Callan
Chief Compliance Officer
Jason Soroko
Fellow
Original broadcast date
July 27, 2023
The CA/Browser Forum recently passed a temporary moratorium on new members of the Certificate Consumer class. We explain how Certificate Consumers have been admitted in the past and the pros and cons of creating stricter rules for Certificate Consumers.
Podcast Transcript
Lightly edited for flow and brevity.
So therefore, this is a temporary moratorium basically on new members that would meet the consumer class. There are two classes of voting members in the CA/Browser Forum. There are CAs. Those are organizations that are qualified to issue publicly trusted certificates that are trusted by the ubiquity of consuming devices in the world. And then there are consumers, which are people who have the devices that needs to read and trust those certificates. Those are the two groups inside the CA/Browser Forum, and this was created way back in the beginning, when CA/Browser Forum really started to build some rules around itself, the recognition of the fact that there's these two sets of parties, and they depend on each other to have an ecosystem that works.
This ballot, which was passed on - I'm going to get the date, it was pretty recently - which passed in late June or in mid-June of 2023. What it does is it sets that new applications of Certificate consumers to the server certificate working group, which is the working group that deals with TLS, new applications will not be accepted, either until one of two things occurs.
Number one is that a ballot is passed with rules around who a Certificate consumer can be or the end of this year, December 31, 2023, whichever of those two things occurs first, that's when the moratorium passes. This is unusual. Like we've never had this before in CA/Browser Forum.
There’s two arguments. Argument number one and the motivator behind this ballot goes the bar for getting into the CA/Browser Forum as a Certificate consumer is incredibly low, you can take a public open-source project like Chromium, you can branch it, you can put your own name on it and now you're a browser. You can call into the calls, you can show up at the meetings, you can take up everybody's time, you can grab the mic, you can vote as a browser, and you may not necessarily have a useful contribution to the web PKI. Like if we wanted, if we were just about power, we could create, we at Sectigo, could create our own browser pretty easily and nobody would have to use it. It wouldn't matter if a single human being in the world didn't use it. Once we had a browser, we can now vote as a browser in CA/Browser Forum and try to skew votes our way and things along those lines. Now, we didn't do that, but somebody could.
If you look at the CA side of the of the CA/Browser Forum, there are very clear guidelines about how you become a member. You have to meet certain criteria, you have to have your roots in trusted root stores. It's an objective thing, it's codified, and anybody can read it. It's available to public.. You can go look at it. You know what you need to do. We can look at a CA and we can say they qualify, or they don't. There are clear rules. If they qualify, they're in and if they don't, they're out.
On the consumer side, there aren't any defined rules at all. There's just a set of people that are consumers, pretty much because they were there at the beginning, pretty much. They're accepted to be consumers, and everybody just accepts it. The impetus behind the ballot, and the worry is, look, this is too loose, and this could be abused. It hasn't been, but it could be, and it's just too loose and this isn't what we're about. We're governing the webs, the web's trust model in a fundamental way. We're governing the web PKI. We got to hold ourselves a little stricter on this. So that's the argument on the one side.
The argument on the other side goes, well, look, what are these rules going to be? I'm not sure I'm comfortable with that. What if they're overly strict? So I'll make something up. Let's suppose you say, can't be a member of CA/Browser Forum if you don't have at least 2% global market share? Okay. Well, perhaps if I'm not a member of CA/Browser Forum, that's what prevents me from getting to 2% global market share. Perhaps this is monopolistic behavior. So you start to get all of those worries. Now, again, none of this has happened. So it's not like anybody has done anything wrong or malfeasant, but we can all see the potential. On the one hand, we get the potential problem. I do.
On the other hand, this idea of collecting power in a small group whose membership supports and reinforces itself is fundamentally uncomfortable, especially for technology people who are supposed to be controlling the web PKI trust infrastructure. And so when you look at it that way, there was a lot to be to be discussed on both sides. The ballot did pass, but the voting was very low, and the abstentions was very high. In general, I think that's reflective of the general discomfort that a lot of people felt about these ideas, even if we could see the reasoning behind them. Does that make sense?

