[Cryptography] The role of the IETF in security of the Internet: for or against the NSA? for or against the security of users of the net?

Phillip Hallam-Baker phill at hallambaker.com
Sat Jul 19 12:11:55 EDT 2014


On Fri, Jul 18, 2014 at 3:47 PM, Paul Wouters <paul at cypherpunks.ca> wrote:
> On Tue, 25 Mar 2014, tpb-crypto at laposte.net wrote:
>
>> Security will not be improved with "trusted proxies" that is for sure. But
>> it seems such ideas are making their way into the protocol while being
>> pushed by the monarch.
>
>
> They are not. Anyone can submit a draft to IETF. That does not mean
> anything. Rest assured, I've talked to enough people to know that
> no "trusted proxy" protocol change will happen.
>
> If you fear one is being sneaked in somehow, contact me one or the
> Security Area Directors at IETF.

The thing about categorical statements like this is that they are
almost certainly wrong.

There is really no problem with a trusted proxy, the question is
whether the proxy is trustworthy or not. Consider the following
possibilities:

1) A trusted proxy run by Vladimir Putin (or choose your favorite war criminal)
2) A trusted proxy you run yourself on a machine that you deployed
situated in a secure location.

Yes, a trusted proxy might not be trustworthy. But it is certainly
possible to deploy a trusted proxy in circumstances that are more
trustworthy than the end point device you happen to be using.

I can't see a great deal of leverage in the approach but it might have
benefits. I certainly don't see any call to attack the IETF for daring
to consider it.


Conventional wisdom very frequently gets it wrong. It is really easy
to get obsessive about one security problem to the exclusion of all
others and to ignore the problem that is the real one.

So while a trusted proxy might not be the best idea, people who try to
slap the idea down without actually thinking about it are not
contributing to the conversation. In fact I am pretty sure that sort
of behavior is how the paid NSA trolls have discouraged consideration
of ideas that would cause real difficulty for their schemes because
they could be widely deployed in favors of crypto-perfectionist
schemes that will only ever be used by crypto-geeks and then only
occasionally.


So when people slap ideas down without giving an argument please
either ignore them or seriously consider the possibility you are being
trolled.


More information about the cryptography mailing list