[Cryptography] new wiretap resistance in iOS 8?

John Denker jsd at av8n.com
Sat Sep 20 16:18:35 EDT 2014


On 09/19/2014 09:16 PM, John Gilmore wrote:
> There must be some other reason, I'm just having trouble thinking of it.

1) As the proverb says, don't let the perfect be the 
 enemy of the good.

 There will never be perfect security.  The measure 
 of good security is that it imposes a cost on the 
 attacker, out of proportion to the cost borne by
 the user.

 The new practice of /not/ escrowing the keys to iOS
 user data does not make the device attack-proof, 
 but it does raise the cost of the attack.

   Forsooth, if this initiative fails, it will not 
   be because it didn't sufficiently raise the cost
   to the attackers, but rather because it imposed
   too much burden on the rightful users.

2) Another proverb goes even farther in the same
 general direction:  A journey of 100 miles begins
 with a single step.

 Suppose there is a weakest-link situation, e.g.
 where locking the front door has no measurable 
 benefit until you also lock the back door, side
 door, windows, et cetera.  You still ought to lock 
 the front door!  Even if you can't do everything
 at once, take the first step and then proceed 
 from there.

3) It is a mistake to focus too directly on the
 threat from the NSA.

 Not escrowing the keys makes Apple somewhat less 
 of a target for the FSB, Third Directorate, etc.
 etc. etc. etc.  Not zero target, but less of a
 target.

 If you're worried about Apple Headquarters being 
 compelled to subvert your phone, you should also 
 be worried about a Clipper-like back door in the
 hardware, which is made in China.  Ditto for HTC
 and other brands.

 Probably the biggest threat from the NSA is more
 /indirect/.  I am referring to weakening crypto 
 standards and products, again and again over the 
 years, thereby creating conditions for a Hobbesian 
 war of all against all.  For example, IMHO it was 
 both arrogant and stupid for the NSA to think they 
 would be the only ones who could break 56-bit DES.

 Tangential remark: Interesting reference:
   Michael Schwartzbeck
   "The Evolution of US Government Restrictions on
    Using and Exporting Encryption Technologies"
   From "Studies in Intelligence"  (the secret internal CIA magazine)
   (date not obvious;  circa 1998)
   http://www.foia.cia.gov/sites/default/files/DOC_0006122418.pdf (prettier)
   http://www.foia.cia.gov/sites/default/files/DOC_0006231614.pdf (same, but uglier)

 Also:  CIA FOIA homepage (with search feature)
  http://www.foia.cia.gov/
 Hundreds of "Studies in Intelligence" articles
 were released last week.

4) It's bad practice to support a strong argument
 with a weak one, but since the topic has already
 been brought up, let me address it.

 For the /subset/ of the problem that concerns
 NSA versus Apple, laws matter ... somewhat.  Yes, 
 there is a long track record of violations, but
 in the spirit of item (1) above, forcing the NSA
 to resort to lawless and unconstitutional methods
 raises the cost to them.

 In particular, if I have information about you, 
 I can be subpoenaed to produce it.  However, if
 I don't have the information, I cannot easily
 be compelled to break into your house to collect 
 it.  If somebody wants to break into your house 
 badly enough they can do it, but we can take 
 steps to raise the cost.

5) We agree that illusory security is worse than
 none.  Tom Mitchell pointed out yesterday that
 Apple does not want to be "directly" complicit 
 in pillaging your data.  However ... if pillage
 is still going on, a big pretense of security
 would be worse than nothing.  It would reflect 
 a "Not My Job" attitude:
   https://www.av8n.com/physics/not-my-job.htm

 So I say let's take a step in the right direction
 today ... and then take whatever additional steps
 are necessary.



More information about the cryptography mailing list