Re: [GNUnet-developers] Plausible deniability challenge?

[Christian Grothoff]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Saturday 28 June 2003 10:58 am, address@hidden wrote:
> A question from a Freenet user:
>
> One of the ways GNUnet improves on Freenet is that it allows persistent
> storage. However, since storage space is finite, it requires people to
> delete files from time to time.

GNUnet does not require file deletion. 

> If gnunet-delete uses a secret key to allow only the inserter of a file to
> remove it, can't the presence of that secret key on a user's hard drive be
> used to prove the user is the originator of that file?

Well, the "secret" that gnunet-delete currently requires is the entire file. 
And note that gnunet-delete is intended for indexed files (not inserted 
files, read docs on insert vs. delete; maybe I should put some more on that 
in the FAQ). Anyway, you can only delete a file that you indexed and you can 
only delete it from your local data store. You have no control over pieces 
that have been distributed to other peers in the meantime. And you don't have 
to delete it. 

> I haven't been successful at finding much documentation on plausible
> deniability in GNUnet and the way insertions and deletions work. If I'm
> grossly misunderstanding something, please let me know.

Read the "Efficient Sharing of Encrypted Data" paper, that should be the best 
we have on that topic. It has some similarities with Freenet, so take
"Freenet CHK" as a reference point. There is also the ESED II page 
(http://www.ovmj.org/GNUnet/encoding.php3) which describes the current 
implementation in GNUnet (essentially, what is different from the original 
paper).

Christian
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE+/4GU9tNtMeXQLkIRAtjSAJ9vKJ6UfrEIaYHW08ZerMKl1LGPiACeJqHQ
XeKD5MZTPDKzKMH2vsCcE8Q=
=TwDy
-----END PGP SIGNATURE-----