News for hackers. Stuff that matters.
Re:Memoirs from the old web: The KEYGEN element by jeroenhd on Monday May 01, 2023 @06:55AM (#35771145)
It's annoying that this technology never made it, but there's also a downside to this tech. Last time I've set up a client cert, I got constantly prompted to authenticate on random websites, and it turns out that ad tracking/fingerprinters/other cyberstalkingware abuse this process to track information about you. Another major problem is that middleboxes freak the fuck out when you don't follow TLS 1.2 to the point crappy middlebox vendors have bothered to implement the spec. Personally, I'd just accept that and let people stuck behind these crappy middleboxes get their ISPs/workplaces to get their shit together, but that's not a very business-friendly approach.

To fix the tracking issue, you'd need to employ strict limitations about what service can or cannot prompt for authentication when you're on a web page, and I'm sure restricting that stuff will break the few remaining big sites that rely on client certificates today.

Luckily, we have an alternative today. WebAuthn uses similar cryptographic principles to set up secure tokens and the UX is a lot better.

Re:Memoirs from the old web: The KEYGEN element by dcow on Monday May 01, 2023 @06:20PM (#35780029)
Meh. Is it really though? Middle boxes can't MITM the signed requests so that's still a "problem". I agree signed requests are better than bearer tokens, but webauthn only signs the auth request and then you typically are back at square 1 receiving a cookie from the server with a bearer token anyway... so no improvement for 99.999% of the requests you actually make. The tracking issues are solvable with browser UX, they're not inherent to client certs.

The only way I see out of this mess is for everyone to start replacing bearer tokens with signed requests. And possibly some extension to webauthn so the browser will use your webauthn creds to sign every requests using a standard signing protocol.

Re:Memoirs from the old web: The KEYGEN element by bandrami on Tuesday May 02, 2023 @06:23AM (#35785485)
There's two different problem domains here that people keep stubbornly insisting on solving with the same technology.

1. I care about the identity of the other station. This does require a full PKI (or something like it)

2. I don't care about the identity of the other station I just don't want some third-party rando listening in. This is the majority of my web traffic, personally: I don't trust ycombinator.com any more than I would trust someone pretending to be ycombinator.com, so the verification that they are ycombinator.com doesn't actually do anything. Just encrypt opportunistically, everywhere, and leave the PKI for situations where it actually matters (like, if I were applying to ycombinator or something).