<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
I am sceptical that this breakthrough has any practical effects. It
is important theoretically, but not in practice. Vide Scott
Aaronson:<br>
<a class="moz-txt-link-freetext" href="http://www.scottaaronson.com/blog/?p=2521">http://www.scottaaronson.com/blog/?p=2521</a><br>
<blockquote type="cite">
<p>In practice, I think, a quasipolynomial-time algorithm is
better than an n<sup>50</sup> algorithm, but worse than an n<sup>2</sup>
algorithm. Unless the constants are absurd, it’s certainly a lot
better in practice than an exponential algorithm.</p>
<p>But then again, <i>in practice</i>, graph isomorphism has
already been “basically in P” for decades! If you have two large
graphs for which you <i>actually</i> need to know whether
they’re isomorphic, just <a
href="http://www3.cs.stonybrook.edu/%7Ealgorith/implement/nauty/implement.shtml"
rel="nofollow">download NAUTY</a> and run it.</p>
<p>This contrasts with the case of factoring, for which I’d
personally say that it remains much less clear whether it should
or shouldn’t be in P.</p>
</blockquote>
<br>
<br>
<pre class="moz-signature" cols="72">--
Anders Sandberg
Future of Humanity Institute
Oxford Martin School
Oxford University</pre>
</body>
</html>