Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

No, it won't.

We've had this technology for decades, TLS client auth with X.509 certificates has been in browsers for a very long time. There just never was any interest in it, and never any investment into making the UI/UX usable beyond the most trivial use cases.

Passkeys are trumped-up certificates with a maybe-optional (depending on attestation status) hardware keystore. And lots of vendor lockin for Google, Apple and Microsoft. The only reason that there is a push right now is that big-vendor interest in lockin.



A solution that’s perfect except for onboarding (people usually need to pay to get a client X.509 cert!), UX, and authenticating to the completely wrong entity (the TLS terminating load balancer instead of the application or authentication server holding user public key credentials).

Surprising how that didn’t become a slam dunk replacing passwords!


Nope. It used to be that browsers even had a Javascript API to create a keypair and submit the public part to the website in question to register or sign for access privileges. Exactly what Passkeys do nowadays. You never had to pay for your client cert if you didn't need it for mail signatures or something.


UI/UX is extremely important to the impact of technology, doubly so for security technologies which often are held back by the difficulty of using them correctly.


That is correct. What I mean is that if there had been any widespread interest, then browser-makers would probably have fixed their UI/UX long ago. But since there never was any interest, nothing was fixed.


There is tremendous interest. TLS client certificates are just a categorically wrong solution to user authentication.


It is basically the same solution as Passkeys. CA involvement in TLS client auth is totally optional.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: