<p><pre><code> The two-slit experiment.
1. Single photon still produces interference pattern!
2. Ask which slit photon passes - pattern disappears
</code></pre>
This kind of explanation is common, and it has always bothered me, because it's the wrong sort of mental model.<p>The interference pattern doesn't disappear "because we're looking", or "because we're asking", or "because we thought about finding the answer".<p>It disappears because in order to check which slit a photon passed through, we need some way of measuring that. To do so, we need some way to "see" the photon; to do that, we need to shine photons on it; and it is <i>that</i> which destroys the interference pattern.<p>The reason quantum mechanics is "weird" is because it (currently) is fundamentally impossible to invent a device to answer the question "which slit did the photon pass through?" without destroying the interference pattern.<p>However, that <i>doesn't</i> mean it disappears "because we're asking". The interference pattern is destroyed because our device, no matter what it is, will <i>always</i> interfere with the experiment (shining photon A at a photon B == "well <i>obviously</i> that would change the behavior of photon B"). Nothing more, nothing less.
I'm about as far from an expert as you can get, so hopefully someone more knowledgeable can shed some light on this:<p>"A classical computer seems to need time exponential in n to predict precisely the behavior of a general quantum mechanical system of n particles. (Yet nature manages to do it in real time.)"<p>How could we know this from inside that system? For all we know it could take a billion [whatever unit is used to measure time outside of our dimension] to calculate each step, and it would still look fluid to us.
<a href="http://scottaaronson.com/blog/?p=208" rel="nofollow">http://scottaaronson.com/blog/?p=208</a> -- possibly helpful for those who saw the Shor's algorithm bit and wanted more.