Ian Grigg wrote:
> 
> Not that it is a huge deal, security wise (any hash is better
> than no hash, which is where we are now), but you should
> consider using SHA-256 for this.
> 
> As of a couple of days ago, SHA-0 was cracked (at least once,
> see below), which pretty much means MD5 is deprecated, and
> SHA-1 is nervous.
> 
> Anyway, I stress that for normal purposes, MD5 is fine, it's
> just that if this feature got added, then it would be smart
> to use the better hashes from the start.  And yes, it would
> be a grand feature to have as then we could do secure
> end-to-end downloading without the infrastructure cost
> imposed by the public key signing methods.

Cracked or not, MD5 is still quite useful for detecting corruption
of a file while downloading it.  I'm not talking about intentional
hacking of a file.  I'm talking about lost packets or inverted bits
while the file is transiting the Internet or being written to the
user's hard drive or a LAN file server.  

However, I do agree with comment #14 in bug #101743 that there
should be options for different hashes.  This would reflect the
possibility that different file sources might use different
hashes.  

-- 

David E. Ross
<http://www.rossde.com/>  

I use Mozilla as my Web browser because I want a browser that 
complies with Web standards.  See <http://www.mozilla.org/>.
_______________________________________________
mozilla-crypto mailing list
[EMAIL PROTECTED]
http://mail.mozilla.org/listinfo/mozilla-crypto

Reply via email to