On Mon, 08 Dec 2014 14:33:16 +0300, Alex Peshkoff <[email protected]>
wrote:
> On 12/08/14 14:02, Mark Rotteveel wrote:
>> On Mon, 08 Dec 2014 13:44:36 +0300, Alex Peshkoff <[email protected]>
>> wrote:
>>> Yes. For password >20 bytes sooner of all exists shorter one with same
>>> hash value.
>> But that is technically irrelevant.
> 
> As long as we do not talk about brute force.

True, but then again the bruteforcing problem also applies if people only
use short passwords. Finding a 'short' password takes the same time (or
maybe less) as finding a shorter password that has the same hash as the
original longer password.
 
>> Although identical hashes are certain
>> due to pigeon holing, that doesn't mean shorter passwords (or passwords
>> near the hash length) are better than longer passwords. And even then,
>> making the assumption that most passwords only use characters between
>> 0x20
>> and 0x7E (95 characters out of potentially 256 in a byte) then a very
>> rough
>> estimate is that identical hashes might only happen after 50 characters
>> (based on 256/95 * 20 = 53.9; a calculation that a cryptologist would
>> probably kill me for because it is either totally wrong or too
>> conservative).
> 
> I've never said that passwords >20 characters are bad. Only that they 
> are not as efficient as may seem.

My original point however stands. Documenting that SRP passwords are
effective up to 20 characters is invalid.

Mark

------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
Firebird-Devel mailing list, web interface at 
https://lists.sourceforge.net/lists/listinfo/firebird-devel

Reply via email to