<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p><br>
    </p>
    <div class="moz-forward-container"><br>
      <br>
      -------- Forwarded Message --------
      <table class="moz-email-headers-table" cellspacing="0"
        cellpadding="0" border="0">
        <tbody>
          <tr>
            <th valign="BASELINE" nowrap="nowrap" align="RIGHT">Subject:
            </th>
            <td>Re: MinEntropy Implications for Passphrase Strength</td>
          </tr>
          <tr>
            <th valign="BASELINE" nowrap="nowrap" align="RIGHT">Date: </th>
            <td>Thu, 12 Dec 2019 15:03:38 -0500</td>
          </tr>
          <tr>
            <th valign="BASELINE" nowrap="nowrap" align="RIGHT">From: </th>
            <td>Arnold Reinhold <a class="moz-txt-link-rfc2396E" href="mailto:agr@me.com"><agr@me.com></a></td>
          </tr>
          <tr>
            <th valign="BASELINE" nowrap="nowrap" align="RIGHT">To: </th>
            <td>procmem <a class="moz-txt-link-rfc2396E" href="mailto:procmem@riseup.net"><procmem@riseup.net></a></td>
          </tr>
        </tbody>
      </table>
      <br>
      <br>
      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
      It’s not an easy question to answer.  Here is a somewhat more
      legible discussion:
      <div class=""><br class="">
      </div>
      <div class=""><a
href="https://crypto.stackexchange.com/questions/66097/why-is-min-entropy-significant-in-cryptography"
          class="" moz-do-not-send="true">https://crypto.stackexchange.com/questions/66097/why-is-min-entropy-significant-in-cryptography</a></div>
      <div class=""><br class="">
      </div>
      <div class="">At the simplest level, if you think of the Diceware
        word list as a set of symbols, and you are picking each symbol
        with a uniform random process, which physical dice approximate
        very well, then min entropy equals Shannon entropy. On the other
        hand, if you look at the resulting pass phrase as a string of
        characters, the distribution will not be uniformly random, and
        the min entropy will be less than the Shannon entropy. The
        Diceware word list can occasionally generate passphrases so
        short that are subject to brute force searches, that’s why I
        recommend requiring a minimum length. 
        <div class=""><br class="">
        </div>
        <div class="">Min entropy attempts to bound the worst case
          behavior, but that is not necessarily realistic. The words
          have semantic meaning and it is possible to randomly generate
          a passphrase like “Four score and seven years ago” which might
          be in a list of, say, the top 1000 English phrases. That could
          be considered a min entropy of less than 10 bits. But such
          occurrences are rare and are fairly easy for humans to spot.
          This does not only apply to Diceware. A string of random
          characters could spell a word. A random hex string could be
          3243F6A8885A3, aka Pi.  One solution would be to check a
          generated password or phrase against a collection of cracker
          lists, but any given password could be added to such lists at
          a later date, so that won’t completely eliminate the problem.
          What Shannon entropy does do for a password or phrase
          generation scheme is measure the likelihood that a weak
          password will be generated, which in the case of Diceware is
          extremely low.</div>
        <div class=""><br class="">
        </div>
        <div class="">Best,</div>
        <div class=""><br class="">
        </div>
        <div class="">agr</div>
        <div class=""><br class="">
        </div>
        <div class=""><br class="">
          <div><br class="">
            <blockquote type="cite" class="">
              <div class="">On Nov 19, 2019, at 6:20 PM, <a
                  href="mailto:procmem@riseup.net" class=""
                  moz-do-not-send="true">procmem@riseup.net</a> wrote:</div>
              <br class="Apple-interchange-newline">
              <div class="">
                <div class="">Hi Arnold. I came across a publication
                  that claims minentropy is a more<br class="">
                  accurate measure for passphrase strength than Shannon
                  Entropy. The<br class="">
                  Wikipedia article on the topic is complex and not
                  really accessible for<br class="">
                  people who want to learn about it.<br class="">
                  <br class="">
                  Questions:<br class="">
                  <br class="">
                  * What is Minentropy and how does it impact Diceware
                  passphrase strength?<br class="">
                  <br class="">
                  * How do I calculate it?<br class="">
                  <br class="">
                  I would appreciate a plain English explanation I can
                  add to our<br class="">
                  documentation. TIA.<br class="">
                  <br class="">
                  <br class="">
                  <a
href="https://www.cs.bu.edu/~reyzin/papers/entropy-survey-ICITS-2011-no-animations.pdf"
                    class="" moz-do-not-send="true">https://www.cs.bu.edu/~reyzin/papers/entropy-survey-ICITS-2011-no-animations.pdf</a><br
                    class="">
                  <br class="">
                  <a class="moz-txt-link-freetext" href="https://en.wikipedia.org/wiki/Min-entropy">https://en.wikipedia.org/wiki/Min-entropy</a><br class="">
                  <br class="">
                  PS. Before sending I found this link that somewhat
                  helps:<br class="">
                  <br class="">
<a class="moz-txt-link-freetext" href="https://crypto.stackexchange.com/questions/63786/relation-between-entropy-and-min-entropy">https://crypto.stackexchange.com/questions/63786/relation-between-entropy-and-min-entropy</a><br
                    class="">
                  <br class="">
                  Does this imply minentropy is only relevant in cases
                  where passphrases<br class="">
                  are formed from sources with non uniform
                  distributions?<br class="">
                  <br class="">
                  I have CC'd our ML so your reply can benefit our
                  users.<br class="">
                  <br class="">
                  <br class="">
                </div>
              </div>
            </blockquote>
          </div>
          <br class="">
        </div>
      </div>
    </div>
  </body>
</html>