If you ask me, it was about time.
http://www.webmonkey.com/2010/08/google-bails-on-wave/
http://googleblog.blogspot.com/2010/08/update-on-google-wave.html
If you keep asking me, I say it's because of lacking integration and backwards (cross-) compatibility right off the bat.
From day #1 what I wanted to do most of all was to hook wave up with the jMonkey team's Google Group. Whoever didn't have Wave, or Gmail for that matter, would follow any mail exchange as usual, except the occasional synopsis of several waves (or were the smaller ones called blips, no blops, or…) into one mail when a bigger, nested and possibly live exchange had happened within wave.
You can't push for a major transition, an evolution of web communication, without the means to make the transition. Wave was all-or-nothing; the majority clearly chose nothing.
I remember watching Google I/O last year and being pumped… This quickly turned to disappointment almost as soon as I got it into my hands and saw that there was really nothing being done with it.
Well, I guess they can't all be winners.
I was still pretty psyched about it when I first got my hands on it, because I figured "well, it hasn't got a very clear purpose or defined use case right now, but it's called a 'preview version' for a reason I suppose. They're Google; they'll iterate, and iterate big!"
… fast forward to present, I'm very disappointed 