Update wording of idea and FAQ sections

This commit is contained in:
Quantum 2018-12-05 00:52:18 -05:00
parent b474b17c49
commit 1ad63b528f

View file

@ -136,9 +136,9 @@
<span class="mark">500 billion attempts every second</span>. Even then, this scheme will resist the <span class="mark">500 billion attempts every second</span>. Even then, this scheme will resist the
cracking attempt for <em>over a year</em>.</p> cracking attempt for <em>over a year</em>.</p>
<p>You don't have to worry about this hypothetical scenario. Running this sort of attack is prohibitively expensive. <p>Running this sort of attack is <em>prohibitively expensive</em>. Unless you possess highly valuable secrets that
Unless you possess highly valuable secrets that makes this sort of effort worthwhile, in which case you should be makes this sort of effort worthwhile (you should use something more secure than passwords in such a case), you don't
using stuff more complex than a password to protect it. If you have to, you can simply add more words.</p> have to worry about this hypothetical scenario. If you have to, you can simply add more words.</p>
<p class="lead">In practice, most real life systems use secure password hashing algorithms, captchas, and other <p class="lead">In practice, most real life systems use secure password hashing algorithms, captchas, and other
mechanisms to stop password cracking. It is unlikely that your attacker can reach anywhere close to a million mechanisms to stop password cracking. It is unlikely that your attacker can reach anywhere close to a million
@ -159,8 +159,8 @@
<p class="card-text"> <p class="card-text">
Another major part of the reason is that a lot of the similar websites fail to generate passwords securely: for Another major part of the reason is that a lot of the similar websites fail to generate passwords securely: for
example, they used <code>Math.random()</code>, which is not a cryptographically secure random number generator, example, they used <code>Math.random()</code>, which is not a cryptographically secure random number generator,
and should never be used to passwords. This website aims to avoid such pitfalls, and provide a truly secure and should never be used to generate passwords. This website aims to avoid such pitfalls, and provide a truly
experience. secure experience.
</p> </p>
</div> </div>
</div> </div>
@ -169,12 +169,16 @@
<h5 class="card-title">Why is this site more secure than alternatives?</h5> <h5 class="card-title">Why is this site more secure than alternatives?</h5>
<ol> <ol>
<li>I have a large list of 8192 words.</li> <li>I have a large list of 8192 words.</li>
<li>All passwords are generated on your computer. No one else will see it.</li>
<li>I use <a href="https://developer.mozilla.org/en-US/docs/Web/API/Crypto/getRandomValues"> <li>I use <a href="https://developer.mozilla.org/en-US/docs/Web/API/Crypto/getRandomValues">
<code>window.crypto.getRandomValues()</code></a> &mdash; the only cryptographically-secure random number <code>window.crypto.getRandomValues()</code></a> &mdash; the only cryptographically-secure random number
generator in JavaScript, instead of generator in JavaScript, instead of
<a href="https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Math/random"> <a href="https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Math/random">
<code>Math.random()</code></a>, whose use is explicitly discouraged for this purporse. <code>Math.random()</code></a>, whose use is explicitly discouraged for this purporse.
</li> </li>
<li>The source code is available on <a href="https://github.com/quantum5/correcthorsebatterystaple">GitHub</a>.
You can audit the code all you want, and build your own trusted version locally.
</li>
</ol> </ol>
</div> </div>
</div> </div>
@ -206,9 +210,10 @@
<strong>shamelessly promoting his own password generation scheme</strong>. <strong>shamelessly promoting his own password generation scheme</strong>.
</p> </p>
<p class="card-text"> <p class="card-text">
If you don't believe me, that okay. Lots of other people agree with me. Check out If you don&rsquo;t believe me, that&rsquo;s okay. Check out
<a href="https://security.stackexchange.com/q/62832/103593">this post on the Information Security Stack <a href="https://security.stackexchange.com/q/62832/103593">this post on the Information Security Stack
Exchange</a>. The top four answers (by a huge margin) agree with me.</p> Exchange</a>. The top four answers (with vast majority of the votes) agree with me, and they provide more
details and additional material to support xkcd-style passwords.</p>
</div> </div>
</div> </div>
</div> </div>