On Mon, Sep 19, 2005 at 02:16:10PM +0100, Michael Rogers wrote:
> Yongqian Li wrote:
> 
> >SHA1 has recently been broken (see
> >http://www.virusbtn.com/news/virus_news/2005/08_29.xml). Will support
> >for a stronger hash be added in Freenet 0.7?
> 
> As far as I know the collision-resistant property of SHA-1 has been 
> broken, but the second-preimage-resistant property hasn't. In other 
> words it's possible to generate two messages with the same hash, but 
> it's impossible to find a second message with the same hash as a given 
> message.

Attacks always get better. The correct response is to upgrade to
SHA-256, and when the dust settles hopefully there will be a consensus
on a better hash.
> 
> I think Freenet requires second-preimage-resistance rather than 
> collision-resistance, although I could well be wrong. For example, if 
> someone wants to insert two new files under the same CHK, is it really a 
> problem? I can see how it would be a problem if someone could insert a 
> second file under an existing CHK, but that isn't possible with the 
> attacks that have been published so far.

Well.. there are some obscure attacks maybe. But the point is that once
you have a practical birthday paradox attack, the hash is broken, and a
second preimage attack is very likely to be found in the future.
> 
> On the other hand perhaps there are social attacks where, for example, 
> you create two files with the same hash, get someone to publish the 
> first file and then claim that they published the second? I suppose this 
> could be used to undermine the credibility of public-key-based pseudonyms.

Right, even the birthday paradox attack can be of *some* use, especially
if combined with social engineering. But it's not a particularly serious
concern.
> 
> Cheers,
> Michael
-- 
Matthew J Toseland - toad at amphibian.dyndns.org
Freenet Project Official Codemonkey - http://freenetproject.org/
ICTHUS - Nothing is impossible. Our Boss says so.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: 
<https://emu.freenetproject.org/pipermail/devl/attachments/20050919/1f6b7649/attachment.pgp>

Reply via email to