Just for reference: 0.96.0 released Feb 12 0.97.0 released Mar 19 0.98.0 released April 7 The reason for the short turnaround is to simply get people to run the latest released and find out if there are any bugs so 1.0 can get released Having people running 0.97.0 for another month reporting this problem and that doesn't do any good if those have already been fixed. So it seems the only way to verify stability (and that the expected fixes do in fact really fix the problem) is to put out a new release and see what gets reported. I have no real idea how many people are running what version (I can look at the metaserver and see that information, but presumably a lot of people are running on private connections or are not reporting to the metaserver). So its really hard for me to gauge just how reliable a release is. And looking at the metaserver, with its abou 10 or so listings, show 3 running 0.96.0 or earlier. I do know that it was shortly after the release of 0.97.0 that a bunch of bugs got reported, so I do sort of base my conclusions that it takes real releases to know stability. As for time between release of 0.98 and 1.0? Ideally, just a couple weeks - it will probably depend a lot of how many bugs get reported - if there are very few bugs reported, I'll take that to mean that the release doesn't have bugs and not that people are not running it. The fact is that it seems that by and large, a lot of developers (me included) want to get a 1.0 release out soon. So the release cycle right now is sped up a bit to help this out. After 1.0 is released, I would expect it to be quite a while before there are any subsequent releases.