I think absolute entropy can be easily calculated.

Relative entropy - which is a logical as opposed to a physical concept - is 
much much more difficult.

-----Original Message-----
From: Ben Scott [mailto:[email protected]] 
Sent: Friday, February 8, 2013 2:09 PM
To: NT System Admin Issues
Subject: Re: Passsword Meter

On Fri, Feb 8, 2013 at 12:57 PM, Michael B. Smith <[email protected]> wrote:
>>   Ideally, password meters should measure entropy, but that's hard to deduce.
>
> Not so much, actually, as long as you control what you mean by entropy.
>
> http://xkcd.com/936/ :)

  I *almost* posted that link.

> See the response here by Akton:
>
> http://programmers.stackexchange.com/questions/167235/how-can-i-estima
> te-the-entropy-of-a-password

  That still seems to be mostly concerned with (1) permutations of character 
sets and (2) dictionary attacks, neither of which directly address entropy.  
Now, a comprehensive dictionary, combined with a comprehensive set of 
substitution rules, can at least tell you if your password can be found using 
such, which is prolly just as good for practical purposes.

  But to measure password strength derived from entropy, I think you'd need 
heuristics taking into account things like patterns and frequencies of both 
letters (spelling) and words (syntax, semantics).

  Disclaimer: Everything I just wrote could be a TOTAL LIE.

-- Ben

~ Finally, powerful endpoint security that ISN'T a resource hog! ~ ~ 
<http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/>  ~

---
To manage subscriptions click here: 
http://lyris.sunbelt-software.com/read/my_forums/
or send an email to [email protected]
with the body: unsubscribe ntsysadmin

~ Finally, powerful endpoint security that ISN'T a resource hog! ~
~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/>  ~

---
To manage subscriptions click here: 
http://lyris.sunbelt-software.com/read/my_forums/
or send an email to [email protected]
with the body: unsubscribe ntsysadmin

Reply via email to