1996-05-28 - Re: holographic remailing & the scientologists (fwd)

Header Data

From: Wei Dai <weidai@eskimo.com>
To: David Vincenzetti <vince@cryptonet.it>
Message Hash: 8f2e0568b2828c8da36e3304a216944c816cbd0244bf5a14d2ac435dcf7d93d9
Message ID: <Pine.SUN.3.93.960528033123.17016A-100000@eskimo.com>
Reply To: <199605280855.KAA23960@relay.cryptonet.it>
UTC Datetime: 1996-05-28 14:56:42 UTC
Raw Date: Tue, 28 May 1996 22:56:42 +0800

Raw message

From: Wei Dai <weidai@eskimo.com>
Date: Tue, 28 May 1996 22:56:42 +0800
To: David Vincenzetti <vince@cryptonet.it>
Subject: Re: holographic remailing & the scientologists (fwd)
In-Reply-To: <199605280855.KAA23960@relay.cryptonet.it>
Message-ID: <Pine.SUN.3.93.960528033123.17016A-100000@eskimo.com>
MIME-Version: 1.0
Content-Type: text/plain


On Tue, 28 May 1996, David Vincenzetti wrote:

> A number of secret-sharing/splitting programs for Unix are available
> at idea.sec.dsi.unimi.it/pub/crypt/code, ftp.dsi.unimi.it's new location
> 
> For instance, check idea.sec.dsi.unimi.it/pub/crypt/code/secshar.tar.gz

Disperse/Collect is a information dispersal program rather than a secret
sharing program.  They are similiar have different purposes.  Here we
want to improve accessibility and reliability without regard to secrecy.
That is, for information dispersal we don't care if information about the
original file is leaked with with each share, whereas secret sharing has
to guarantee that an attacker can find out nothing about the original
secret unless he has at least k shares.

That aside, Disperse/Collect might be better suited for what Vladimir had
in mind because it was explicitly designed for broadcasting files through
Usenet.  The secret sharing programs you mention would not work well
because each share would be as big as the original secret.  (There are
secret sharing schemes with short shares, but I don't think any of those
programs implement the more efficient schemes.)

Wei Dai







Thread