Fighting cookie theft using device bound sessions
(blog.chromium.org)
from neme@lemm.ee to technology@lemmy.world on 03 Apr 2024 00:29
https://lemm.ee/post/28533245
from neme@lemm.ee to technology@lemmy.world on 03 Apr 2024 00:29
https://lemm.ee/post/28533245
threaded - newest
This is an incredibly important step forward but I have to wonder why its taken this long to come up with.
Isn’t this what WebAuthn already does? Why introduce a new protocol when another one does the job well?
I don’t think WebAuthn protects against cookie theft. WebAuthn better protects the login process. But if the result of the login process is still a session/auth cookie, that can be stolen like any other cookie.
That took forever to implement, almost as if someone had stake in not doing this
Be careful since it is a double edged sword. Device bound session means the browser has the capabilities to differentiate devices, and thus can be used for more accurate tracking information. Of course I’m not saying it is not useful, having created a fair share of websites myself, I know the pain of authentication on the web and how it can be challenging to secure from tons of possible attack vectors. And in my experience, the weakest link is always the user.
The browser can do it whether this exists or not. The only information the website gets is that the browser supports this feature or not, and nothing else.
My bad, I worded that badly. What I meant is that the website now has access to those features via the browser (js or some other mechanism). Now suddenly fingerprinting a device can be made easier.
That’s a valid concern, but according to the article all the website can access is the random public key, or the fact that the feature is unsupported in this browser (for an unspecified reason).
Yeah, I’ve also read the article. I am just being cautious on how it can be used for other things that cause privacy concern. And so far, I’ve come up blank too.