It's inherently worse in that model particularly if the server isn't a slave of the phone that drops the key based on phone use.
-
-
Replying to @CopperheadOS @RichFelker
It loses nice properties an impl can have that's on-device and the server could be in attacker control when the user unlocks.
2 replies 0 retweets 0 likes -
Replying to @CopperheadOS
Your threat model seems really warped & out of touch with the reality of how vulnerable Android devices are.
1 reply 0 retweets 0 likes -
Replying to @RichFelker
You're the one with warped perceptions, bias, lack of experience, and lack of threat modelling.
2 replies 0 retweets 0 likes -
Replying to @CopperheadOS
Attacker breaking into home/office & successfully tampering w/phys tamper-resistent server != normal person's threat model.
2 replies 0 retweets 0 likes -
Replying to @RichFelker
Far more likely than anything you're mitigating, and your model is incompatible with proper physical security.
1 reply 0 retweets 0 likes -
Replying to @CopperheadOS @RichFelker
It introduces all kinds of new attack vectors, and doesn't actually remove any or provide additional security properties.
1 reply 0 retweets 0 likes -
Replying to @CopperheadOS @RichFelker
Plus same vendors are just going to be churning out this server hardware and doing the same stuff they do on client hardware.
1 reply 0 retweets 0 likes -
Replying to @CopperheadOS
No. The whole idea is to displace the vendors selling the crap they sell now.
1 reply 0 retweets 0 likes -
Replying to @RichFelker
So instead there will be vendors selling a poor implementation of that instead. Good stuff exists today, people mostly buy bad.
2 replies 0 retweets 0 likes
Good stuff costs too much for people who need it to afford, especially when you have to trash it after potential compromise.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.