494 \noindent then just a lookup in the dictionary will reveal

   494 \noindent then just a lookup in the dictionary will reveal that the

   495 that the plain-text password was \pcode{password}. What is

   495 plain-text password was \pcode{password}. What is good about this

   496 good about this attack is that the dictionary can be

   496 attack is that the dictionary can be precompiled in the ``comfort of

   497 precompiled in the ``comfort of the hacker's home'' before an

   497 the hacker's home'' before an actual attack is launched. It just needs

   498 actual attack is launched. It just needs sufficient storage

   498 sufficient storage space, which nowadays is pretty cheap. A hacker

   499 space, which nowadays is pretty cheap. A hacker might in this

   499 might in this way not be able to crack all passwords in our database,

   500 way not be able to crack all passwords in our database, but

   500 but even being able to crack 50\% can be serious damage for a large

   501 even being able to crack 50\% can be serious damage for a

   501 company (because then you have to think about how to make users to

   502 large company (because then you have to think about how to

   502 change their old passwords---a major hassle).  And hackers are very

   503 make users to change their old passwords---a major hassle).

   503 industrious in compiling these dictionaries: for example they

   504 And hackers are very industrious in compiling these

   504 definitely include variations like \pcode{passw0rd} and also include

   505 dictionaries: for example they definitely include variations

   505 rules that cover cases like \pcode{passwordpassword} or

   506 like \pcode{passw0rd} and also include rules that cover cases

   506 \pcode{drowssap} (password reversed).\footnote{Some entertaining rules

   507 like \pcode{passwordpassword} or \pcode{drowssap} (password

   507   for creating effective dictionaries are described in the book

   508 reversed). Historically, compiling a list for a dictionary

   508   ``Applied Cryptography'' by Bruce Schneier (in case you can find it

   509 attack is not as simple as it might seem. At the beginning

   509   in the library), and also in the original research literature which

   510 only ``real'' dictionaries were available (like the Oxford

   510   can be accessed for free from

   511 English Dictionary), but such dictionaries are not

   511   \url{http://www.klein.com/dvk/publications/passwd.pdf}.}

   512 ``optimised'' for the purpose of passwords. The first real

   512 Historically, compiling a list for a dictionary attack is not as

   513 hard data about actually used passwords was obtained when a

   513 simple as it might seem. At the beginning only ``real'' dictionaries

   514 company called RockYou ``lost'' 32 Million plain-text

   514 were available (like the Oxford English Dictionary), but such

   515 passwords. With this data of real-life passwords, dictionary

   515 dictionaries are not ``optimised'' for the purpose of passwords. The

   516 attacks took off. Compiling such dictionaries is nowadays very

   516 first real hard data about actually used passwords was obtained when a

   517 easy with the help of off-the-shelf tools.

   517 company called RockYou ``lost'' 32 Million plain-text passwords. With

   518 this data of real-life passwords, dictionary attacks took

   519 off. Compiling such dictionaries is nowadays very easy with the help

   520 of off-the-shelf tools.

   561 Note another interesting point. The web-application from the

   564 Note another interesting point. The web-application from the previous

   562 previous section was only secure when the salt was secret. In

   565 section was only secure when the salt was secret. In the password

   563 the password case, this is not needed. The salt can be public

   566 case, this is not needed. The salt can be public as shown above in the

   564 as shown above in the Unix password file where is actually

   567 Unix password file where is actually stored as part of the password

   565 stored as part of the password entry. Knowing the salt does

   568 entry. Knowing the salt does not give the attacker any advantage, but

   566 not give the attacker any advantage, but prevents that

   569 prevents that dictionaries can be precompiled. While salts do not

   567 dictionaries can be precompiled. The moral is that you should

   570 solve every problem, they help with protecting against dictionary

   568 never store passwords in plain text. Never ever.

   571 attacks on password files. It protects people who have the same

   569 

   572 passwords on multiple machines. But it does not protect against a

   573 focused attack against a single password and also does not make poorly

   574 chosen passwords any better. Still the moral is that you should never

   575 store passwords in plain text. Never ever.\medskip

   576 

   577 \noindent

   578 If you want to know more about passwords I recommend viewing some

   579 youtube videos from the PasswordCon(ference) which takes place each

   580 year. The book by Bruce Schneier about Applied Cryptography is also

   581 recommendable, though quite expensive.  Clearly, passwords are a

   582 technology that comes to the end of its usefulness, because brute

   583 force attacks become more and more powerful and it is unlikely that

   584 humans get any better in remembering (securely) longer and longer

   585 passwords. The big question is which technology can replace

   586 passwords\ldots