Re: chaffing and winnowing

Rickard Armiento <[email protected]>:

> clemens:
>> for chaffing and winnowing see
>> <URL:http://theory.lcs.mit.edu/~rivest/chaffing.txt>.  this is not
>> really encryption, but it will legally work even in countries were
>> encryption is outlawed.
>
> I have taken this outside our other discussion since it may be
> of more general interest.
>
> I have browsed the linked text and find myself thinking that
> "chaffing and winnowing" is a bogus "solution" to the problem of
> governments wishing to control information content much in the same
> way D. J. Bernstein argues about real and bogus "solutions" to spam,
>   http://cr.yp.to/qmail/antispam.html

> If chaffing and winnowing become widely used, any government that
> already has laws against non-government sanctioned encryption would
> easily outlaw this too, using some vague formulation:
>    "For digital communication, you are not allowed to conceal content
>    from the government on purpose, for example by using
>    'chaffing and winnowing' techniques."

if an ISP (or the user) adds mock datastreams for conceiling what
he's really communicating, you're right.  but there's no need for
that:  the ISP just appends hundreds of messages to form millions of
data packets, each with its own MAC.  there's no conceiling taking
place, it would be up to the spies to pick out the streams he's
interested in and verify their MACs.

then when users send voice audio together with important other
documents like faxes and PDF documents, who's to accuse him of
conceiling anything?

> Hence, using this in an email protocol with the purpose of going
> around govenments who have outlawed encryption seems meaningless to
> me.

yes, and it will remain meaningless.  no court rooms to visit, no
defendants to be tried.

  clemens