Date: Tue, 20 Mar 2018 17:08:22 +0100 From: Matlink <matlink@...link.fr> To: john-users@...ts.openwall.com Subject: Re: Is john really using Narayanan and Shmatikov whitepaper? Le 16/03/2018 à 20:07, Solar Designer a écrit : > I'm not aware of any documentation of incremental mode's internals. > Since I still have an unimplemented idea for improving it, maybe I > should revisit this if I have time in a few years from now (I doubt I'd > have time for this sooner than that), and give a talk on it at that > point (as I'd need to recall a few things anyway) - although it'd be a > bit weird to do that on something that is already 20 years old, except > for those improvements. > The thing is that if researchers have no explanation on how this mode works, they are barely constrained not to use it because they won't be able to interpret or understand their results with it. Moreover, an implementation is always a subjective manner to resolve a problem. Therefore, shortcuts are often applied due to the used language, where they cannot be used in another language. If we have no generic (language independent) algorithm (i.e. pseudocode) it might be complicated to understand what the coder wanted to do. Having such an explanation/documentation can benefit since many contributors could question the way it has been implemented. Currently I don't see any difference between incremental mode and a raw order 2 Markov-chains-based model (or maybe letters positions are also considered). Mat -- Matlink - Sysadmin matlink.fr Sortez couverts, chiffrez vos mails : https://café-vie-privée.fr/ XMPP/Jabber : matlink@...link.fr Clé publique PGP : 0x186BB3CA Empreinte Off-the-record : 572174BF 6983EA74 91417CA7 705ED899 DE9D05B2
Powered by blists - more mailing lists
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.