Fellow ChucKists,<div><br></div><div>I was poking around in the source for &quot;maybe&quot; to try to figure out how hard Joe&#39;s idea would be to implement.</div><div><br></div><div>So far it looked really quite easy to me, except that the chuck operator would have to be overloaded to deal with chucking floats to &quot;likely&quot; and &quot;unlikely&quot;. This made me wonder what kind of error this would give;</div>
<div><br></div><div>1 =&gt; maybe;</div><div><br></div><div>It turns out that doesn&#39;t give a error at all and instead seg-faults (latest miniAudicle, linux ALSA, canonical version except for the CNoise fix).</div><div>
<br></div><div>So; regardless of whether or not we go with Joe&#39;s idea this area does need a look.</div><div>I vote that we do take this idea as it would be very little work for something that would be that much fun. With a default of .75% for likely and 25% for unlikely it would have all of the advantages of the various proposed strategies and on top of that we&#39;d get a easy &quot;crossfade&quot; for chance-based techniques that may span several shreds.</div>
<div><br></div><div>Yours,</div><div>Kas.</div>