Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[Kevin W. Wall]

On Thu, Jun 16, 2011 at 5:27 PM, James A. Donald  wrote:

> On 2011-06-17 4:02 AM, Nico Williams wrote:
>
>  Crypto is no more than an equivalent of doors, locks, keys, safes, and
>> hiding.
>>
>
> The state can break locks, but it cannot break crypto.
>
> Hiding *is* effectual against the state - and long has been even before
> crypto.
>

The key word here being *effectual*. Crypto is effective, but some of your
posts
make it seem to be a panacea, similar to how Bruce Schneier originally
thought
(see preface of *Applied Cryptography*) that cryptography was going to be
the
salvation of information security. Crypto certainly has a major role to play
in ensuring confidentiality and integrity, but it is not an be-all
andend-all.
The point is, the state doesn't always *need *to *break *crypto to get your
secrets.

To that end, I think you are misinterpreting what Nico was trying to say,
which
was, crypto is no guarantee that you can hide things from the state, at
least
as it is practiced by the general populace.

Specifically, if that "state" is some corrupt regime, crypto *may*[1] help,
but it
will not ensure with 100% certainty that your secrets will remain
confidential
from the state.

For that to be true, everything would have to be secure, from the OS all the
way
down to all the firmware. (See Ken Thompson's ACM Turing Award lecture,
*On Trusting Trust*.)  You'd also have to eliminate all possible side
channel
attacks such as EMF leaks. And even if you are secure from attacks coming
from
all those threat sources, an unscrupulous state will have no compunctions
about using a rubber hose attack on you or ones you care about to get your
secrets or get you to divulge your crypto keys. (Someone in an earlier post
mentioned how it is already getting close to that in certain criminal cases
in
England. How much worse would it be with a corrupt regime not following
principled rule-of-law at all?)

While I don't want to put words into Nico's mouth, I think he was merely
trying to point out the difference between the use of crypto in theory and
crypto in practice.
_
[1] Using crypto in a fascist or otherwise corrupt state where crypto is not
the norm
may have the opposite affect of drawing attention to yourself and arousing
the
suspicion of the state. So in such cases, one at least needs to account for
plausible deniability, otherwise you'd be better off keeping your head low
so as not to be noticed in the first place.

-kevin
--
Blog: http://off-the-wall-security.blogspot.com/
"The most likely way for the world to be destroyed, most experts agree,
is by accident. That's where we come in; we're computer professionals.
We *cause* accidents."-- Nathaniel Borenstein
___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[James A. Donald]

On 2011-06-17 4:02 AM, Nico Williams wrote:
> Osama, for

example, was found in part by traffic analysis.  Maybe he should have
been using Tor instead of USB sticks and couriers, but I bet usage of
Tor from a sleepy Islamabad suburb would have led to his being found
sooner.


That was not a sleepy Islamabad suburb, but the center of military elite 
- ground zero for conspiracy as Stanford is the center of venture 
capital.  Had he been using Tor, would have fitted right in.  Plus, if 
it had been a sleepy Islamabad suburb, Tor would have been even better 
because of all the suburbanites using Tor to download porn.



Crypto is no more than an equivalent of doors, locks, keys, safes, and
hiding.


The state can break locks, but it cannot break crypto.

Hiding *is* effectual against the state - and long has been even before 
crypto.  The collapse of tax revenues in Greece is largely due to 
methods of tax evasion and tax resistance against the predatory state 
that would be familiar to any third worlder.  That tax evaders can now 
do business over skype video just makes it easier.  They don't notice 
that they are using crypto to resist the state.


Indeed it is a general user interface principle, that for people to use 
crypto effectively to resist the state, they should not need to notice 
that they are using crypto.  Tools that require conscious deployment of 
crypto will never be widely deployed, and when deployed, the users will 
foul up.


> At the end of the day the rubber has to meet the road.

People need to interact with other people in order to obtain the
things they need for mere survival.


They interact through a banking system that is largely a branch of the 
state, and an accounting system that is transparent to the state - both 
of which are collapsing due to incompetent, corrupt, and ham fisted 
state intervention.


In the financial crisis, it not only became apparent that political 
favor is more important to bankers than profits and losses, it also 
became obvious that Sarbannes Oxley compliant accounting fails to give a 
true and fair account of a business's financial status.


They also interact through vpns and encrypted video conferencing - the 
Chinese economy is booming in substantial part through a high tech 
sector that largely bypasses or ignores the regular banking and regular 
accounting system for higher tech alternatives, and the entire Greek 
economy is going dark, disappearing from sight of the state, though I 
would guess that in Greece this largely through third world methods of 
tax evasion and tax resistance, rather than high tech methods.  They 
keep raising taxes, and, mysteriously, tax revenues fall every time.


___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[Nico Williams]

On Wed, Jun 15, 2011 at 5:08 PM, James A. Donald  wrote:
> On 2011-06-15 7:05 PM, Nico Williams wrote:
>>
>> It's only when push comes to shove that crypto
>> doesn't help.
>
> In the conflict with Al Quaeda, as in any war, push has come to shove, and
> yet encryption does help.

Encryption, assuming they use it, hasn't helped them all that much.  A
great many of their rank-and-file as well as their leadership are
dead.  At least some of them were found via intelligence processes
that, from what we know, required no code breaking.  Osama, for
example, was found in part by traffic analysis.  Maybe he should have
been using Tor instead of USB sticks and couriers, but I bet usage of
Tor from a sleepy Islamabad suburb would have led to his being found
sooner.

In other words, you've proved my point.

> [...]
>
> The state can escalate - but so can we.

Sheer lunacy.  There's a real world analogy for most online
situations, and if you're willing to stretch a bit, for all online
situations.

Crypto is no more than an equivalent of doors, locks, keys, safes, and
hiding.  At the end of the day the rubber has to meet the road.
People need to interact with other people in order to obtain the
things they need for mere survival.  There are not eight billion
deserted islands in which each of us can survive alone, and we'd not
want to anyways because we're social beings.  And when the rubber
meets the road, the crypto stops helping.

Anyone can build a fortress to protect themselves from the surrounding
state.  There have been many such fortresses in the U.S., yet when
push came to shove in the case of the Branch Davidians in Waco, TX,
the fortress didn't help.  The reason: the state has more resources
because they represent a larger society.  Now, maybe the Davidians
should have made bio/chem/nuclear weapons with which to blackmail the
state into leaving them alone, but that would probably have brought
them to the attention of the state sooner, soon enough, one hopes,
that they never managed to get the weapons and delivery mechanisms
that would make them capable of pulling off the blackmail.

Nico
--
___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[Jeffrey Walton]

On Wed, Jun 15, 2011 at 5:51 PM, James A. Donald  wrote:
> On 2011-06-15 6:22 PM, Adam Back wrote:
>>
>> Then there are countries where crypto is officially or effectively already
>> banned - there being caught with privacy tech on your laptop, cell phone
>> etc
>> would be dangerous.
>
> Which, however, tend to be the countries where there is lots of privacy tech
> on people's lap tops.
>
> Consider Al Quaeda.  The US, rightly, does not allow anything to limit its
> pursuit of Al Qaeda.  The US does not bother with arrests nor charges, nor
> worry overmuch about innocent bystanders.  All is fair in love and war.
Debatable, but off topic.

> Do al Qaeda members say "Oh dear, I guess encryption is out for us.  It
> might look suspicious."
http://www.theregister.co.uk/2011/03/22/ba_jihadist_trial_sentencing/

Jeff
___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[James A. Donald]

On 2011-06-16 4:47 AM, Nico Williams wrote:

That's nice, but not scalable.  Scale that up enough and you have
anarchy, which is just a temporary situation until a strongman takes
over.


Firstly:
Anarchy always exist.  The state is an island in a sea of anarchy, and 
that island increases or diminishes from decade to decade.  In 
particular, our banking system is now collapsing as the state pervades 
it, for the state lacks both the competence and the moral integrity 
needed to operate banking.  Basel was an expansion of that island, and 
when the banking system collapses and is bypassed, this will reduction 
in the state:


Secondly:
States, a monopoly of force, are not that easy to establish.  It took 
the Taliban nine years, using the most horrifying and desperate means, 
to establish state authority in Afghanistan after communism fell.


For a single authority to make itself supreme, to attain a broadly 
accepted monopoly of force, is invariably long, costly and terrible, 
requiring at best the means the Sherman applied in his march to the sea, 
at worst the methods that Taliban applied.


___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[James A. Donald]

On 2011-06-15 7:05 PM, Nico Williams wrote:

It's only when push comes to shove that crypto
doesn't help.


In the conflict with Al Quaeda, as in any war, push has come to shove, 
and yet encryption does help.



Long before push comes to shove you have to deal with the fact that
your crypto is only a small part of the big picture: do you know if
your peers are malicious?


If some of your peers are not malicious, mixing will work.  If all of 
your peers are malicious, steganography will work.


Because menacing state demands for the contents of your files are 
common, steganography is common, for example truecrypt.  Because state 
attacks on encrypted communication are uncommon, people do not stego 
their communications.  But if the state attacked encrypted 
communications, then we would also see plenty of encrypted communication 
stego.


The state can escalate - but so can we.

___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[James A. Donald]

On 2011-06-15 6:22 PM, Adam Back wrote:

Then there are countries where crypto is officially or effectively already
banned - there being caught with privacy tech on your laptop, cell phone
etc
would be dangerous.


Which, however, tend to be the countries where there is lots of privacy 
tech on people's lap tops.


Consider Al Quaeda.  The US, rightly, does not allow anything to limit 
its pursuit of Al Qaeda.  The US does not bother with arrests nor 
charges, nor worry overmuch about innocent bystanders.  All is fair in 
love and war.


Do al Qaeda members say "Oh dear, I guess encryption is out for us.  It 
might look suspicious."

___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[Nico Williams]

On Wed, Jun 15, 2011 at 1:36 PM, StealthMonger
 wrote:
> Some folks do not choose to have a state.  For them, all states are
> foreign powers.

That's nice, but not scalable.  Scale that up enough and you have
anarchy, which is just a temporary situation until a strongman takes
over.  And even for a few individuals it's a problem.  How do you make
a living?  Even if you're independently wealthy, where do you get your
supplies?  What are you doing online?  How will you get online in a
world where most people necessarily live in states, thus your access
to _them_ depends on those states.  Push comes to shove you could live
as a hermit, armed to the teeth to defend yourself from those foreign
powers.

You're not convincing me that crypto is the answer.  It's part of the
answer, sure, but not _the_ answer.

Even though this is about crypto, it's still OT.  Have the last word,

Nico
--
___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[Nico Williams]

On Wed, Jun 15, 2011 at 1:36 PM, StealthMonger
 wrote:
> Some folks do not choose to have a state.  For them, all states are
> foreign powers.

That's nice, but not scalable.  Scale that up enough and you have
anarchy, which is just a temporary situation until a strongman takes
over.  And even for a few individuals it's a problem.  How do you make
a living?  Even if you're independently wealthy, where do you get your
supplies?  What are you doing online?  How will you get online in a
world where most people necessarily live in states, thus your access
to _them_ depends on those states.  Push comes to shove you could live
as a hermit, armed to the teeth to defend yourself from those foreign
powers.

You're not convincing me that crypto is the answer.  It's part of the
answer, sure, but not _the_ answer.

Even though this is about crypto, it's still OT.  Have the last word,

Nico
--
___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[Jeffrey Walton]

On Wed, Jun 15, 2011 at 2:36 PM, StealthMonger
 wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> Nico Williams  writes:
>
>> crypto has a place ... to protect us ... from foreign powers, and
>> from casual inspection by one's state 
>
> Some folks do not choose to have a state.  For them, all states are
> foreign powers.
>
>> You must participate in ... politics   you must change [your
>> society's] culture 
>
> No, you may simply go your own way in peace.
Ask anyone who lived under the Khmer Rouge or Democratic Republic of
Vietnam about this.

Jeff
___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[StealthMonger]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Nico Williams  writes:

> crypto has a place ... to protect us ... from foreign powers, and
> from casual inspection by one's state 

Some folks do not choose to have a state.  For them, all states are
foreign powers.

> You must participate in ... politics   you must change [your
> society's] culture 

No, you may simply go your own way in peace.

- -- 


 -- StealthMonger 
Long, random latency is part of the price of Internet anonymity.

   uinmyn: Is this anonymous browsing, or what?
   
http://groups.google.com/group/alt.privacy.anon-server/msg/59a1d785aaa19de5?dmode=source&output=gplain

   stealthmail: Hide whether you're doing email, or when, or with whom.
   mailto:stealthsu...@nym.mixmin.net?subject=send%20index.html


Key: mailto:stealthsu...@nym.mixmin.net?subject=send%20stealthmonger-key

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iEYEARECAAYFAk342yoACgkQDkU5rhlDCl7j0wCghWZg3XK75QnouNT3Lup8KSCx
ohsAn1jqEm3amszVrUElcmfoTMG/lFDi
=ZPhI
-END PGP SIGNATURE-

___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

Re: [cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[Nico Williams]

On Wed, Jun 15, 2011 at 3:22 AM, Adam Back  wrote:
> Well said StealthMonger, I suspect Nico is in the minority on this list with
> that type of view.
>
> I read Nico's later reply also.  Short of banning crypto privacy and
> security rights stand a better chance of being balanced by more deployment
> of crypto.  (In terms of warrantless wiretaps etc which seem to just keeping
> going and getting worse in many supposedly civilized western democracies.)
> There are still plenty of things government security people can usefully do
> towards security - spend the money on inflitration of groups who are real
> security threats.

Don't misunderstand me: I think crypto has a place, and that place is
mostly to protect us from other private citizens, from foreign powers,
and from casual inspection by one's state (i.e., keeping the state and
its minions honest).  It's only when push comes to shove that crypto
doesn't help.

Long before push comes to shove you have to deal with the fact that
your crypto is only a small part of the big picture: do you know if
your peers are malicious? are your compute resources physically
secure? are you certain of that? are they tamper resistant? are there
unpatched, or worse, unknown-to-you vulnerabilities in your software
(or worse, firmware, or worse, hardware) that others could exploit? is
your key management secure?

Security is oh so much more than just using AES, so much more than
just using secure cryptographic protocols and algorithm suites.
Crypto does not completely change the nature of security in the online
world versus physical security in the off-line world -- there's
analogies for most situations.  Crypto alone is not a panacea.

If you want to live in a free society you must do more than hide
behind ciphers.  You must participate in its politics to keep your
society free.  If it isn't already free, then you have a very big
problem -- crypto can only be a small part of how you might address
it.

For example, if in order to free your society you conclude that you
must change its culture openly, then crypto won't help you for you
must speak publicly.  Crypto will help you, to a point, if you're
trying to organize a revolt, but don't be surprised when crypto fails
to keep you safe in that case -- you'll likely need weapons and to be
willing to use them.

BTW, I'm surprised any of what I've said on this is remotely controversial.

Nico
--
___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography

[cryptography] crypto & security/privacy balance (Re: Digital cash in the news...)

[Adam Back]

Well said StealthMonger, I suspect Nico is in the minority on this list with
that type of view.

I read Nico's later reply also.  Short of banning crypto privacy and
security rights stand a better chance of being balanced by more deployment
of crypto.  (In terms of warrantless wiretaps etc which seem to just keeping
going and getting worse in many supposedly civilized western democracies.)
There are still plenty of things government security people can usefully do
towards security - spend the money on inflitration of groups who are real
security threats.

I would say privacy tech & crypto is essential to maintaining a good point
on the security/privacy balance in a world where security policy
encroachment has gone into overdrive.  To retain electronic liberty, crypto
is the answer.  I dont think crypto can be realistically banned in western
countries at this stage, the electronic part of security encroachment is
mostly opportunistic hoovering up things that are not protected.

There are multiple privacy properties - confidentiality of communication
contents, privacy of association (cryptographic freedom of association) like
pseudonymous email (protection against traffic analysis), cryptographic
enforced member only discussion groups/chats.

Then there are countries where crypto is officially or effectively already
banned - there being caught with privacy tech on your laptop, cell phone etc
would be dangerous.

Crypto and other privacy techniques can counteract somewhat - with
steganography, that though obviously its a tough threat model.  See 


http://www.nytimes.com/2011/06/12/world/12internet.html?_r=1

Its also a kind of interesting conflict that western governments think of
themselves, or try to portray themselves as moral forces of good and yet
there are a few cases where this technology the US is helping fund really
needs to be used in western democracies, including the US.  


The UK governments right to force key disclosure is an abomination, no
civilized country should be going in that direction.

Adam

On Tue, Jun 14, 2011 at 10:30:18PM +0100, StealthMonger wrote:

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Nico Williams  writes:


Crypto will NOT protect you from the state.


Hmm?  Protection from the state is the very reason some of us are
here.  Even Philip Zimmermann wrote twenty years ago [1]

  Why Do You Need PGP? ...  you may be doing something that you feel
  shouldn't be illegal, but is.

And the very title of David Chaum's 1985 paper was "Security without
Identification: Transaction Systems to Make Big Brother Obsolete" [2]

[1] pgpdoc1.txt

[2] CACM 28(10), October 1985

___
cryptography mailing list
cryptography@randombit.net
http://lists.randombit.net/mailman/listinfo/cryptography