6 ms·

The problem is that we are not enough - you have to get the other 95 % not really caring on the bandwagon...and this without causing the slightest bit of additi

by danbruc 13y ago

The problem is that we are not enough - you have to get the other 95 % not really caring on the bandwagon...and this without causing the slightest bit of additional inconvenience.

greyman 13y ago
Or it is rather 99.995? ;-)
- danbruc 13y ago
  True. 316 million US citizens and 117 thousands of them signing the Snowden petition yields a first approximation of 99.96 %.
eru 13y ago
95% would be nice, but a few percent should be good enough, if there's enough traffic overall. (And even 100% would not be enough if there's almost no traffic, like in the beginning of the internet.) Absolute numbers matter more than shares, here.
buro9 13y ago
We own websites... make them SSL. We affect the 95%.
- malloc2x 13y ago
  You must assume that the NSA already has master keys for all domestic CA root certificates, and given how many were hacked recently, foreign ones too. In which case SSL traffic is effectively the same as plaintext to them. The solution could be a distributed CA system like http://convergence.io/ http://convergence.io/
  - jeltz 13y ago
    Perfect forward security should prevent the master key from working.
  - karlmdavis 13y ago
    Is this correct? Wouldn't they still need all of the leaf private keys to decrypt things? My understanding was that having a CA's private key just enables someone to issue new child keys for that CA. That vulnerability could be addressed with certificate pinning.
    - malloc2x 13y ago
      To decrypt after-the-handshake bytes I think you're right, they would need a leaf private key. However, they absolutely can mount a MITM with the CA root. EDIT: Further, if they can compel a master key then they can also compel a copy of all the private keys the CA generates.
      - juhanima 13y ago
        Not quite sure what you mean, but for the record, as a general rule CAs do not generate keys. They just sign the public keys coming in as Certificate Signing Requests. Without ever seeing the accompanied private key.
mhandley 13y ago
On the other hand, we do have the influence to get something like http://tcpcrypt.org http://tcpcrypt.org standardized and deployed in commodity operating systems if we really care enough. Then the 95% won't need to care.

by danbruc 13y ago

The problem is that we are not enough - you have to get the other 95 % not really caring on the bandwagon...and this without causing the slightest bit of additional inconvenience.

greyman 13y ago

Or it is rather 99.995? ;-)

danbruc 13y ago
True. 316 million US citizens and 117 thousands of them signing the Snowden petition yields a first approximation of 99.96 %.

eru 13y ago

95% would be nice, but a few percent should be good enough, if there's enough traffic overall. (And even 100% would not be enough if there's almost no traffic, like in the beginning of the internet.) Absolute numbers matter more than shares, here.

buro9 13y ago

We own websites... make them SSL. We affect the 95%.

malloc2x 13y ago
You must assume that the NSA already has master keys for all domestic CA root certificates, and given how many were hacked recently, foreign ones too. In which case SSL traffic is effectively the same as plaintext to them. The solution could be a distributed CA system like http://convergence.io/ http://convergence.io/
- jeltz 13y ago
  Perfect forward security should prevent the master key from working.
- karlmdavis 13y ago
  Is this correct? Wouldn't they still need all of the leaf private keys to decrypt things? My understanding was that having a CA's private key just enables someone to issue new child keys for that CA. That vulnerability could be addressed with certificate pinning.
  - malloc2x 13y ago
    To decrypt after-the-handshake bytes I think you're right, they would need a leaf private key. However, they absolutely can mount a MITM with the CA root. EDIT: Further, if they can compel a master key then they can also compel a copy of all the private keys the CA generates.
    - juhanima 13y ago
      Not quite sure what you mean, but for the record, as a general rule CAs do not generate keys. They just sign the public keys coming in as Certificate Signing Requests. Without ever seeing the accompanied private key.

mhandley 13y ago

On the other hand, we do have the influence to get something like http://tcpcrypt.org http://tcpcrypt.org standardized and deployed in commodity operating systems if we really care enough. Then the 95% won't need to care.