Google Wave is dead, long live Google Wave? Shortly after the company announced that it was mothballing the Wave platform, the company is now pushing "Wave in a Box," as an open source second try for the dead platform.
Typically, I don't think much of the "open source as last resort" strategy. You know the one, when companies chuck a project over the wall as open source when it didn't figure out how to make it work as a proprietary project. Wave is slightly different in this regard. First of all, Wave never should have been a proprietary project in the first place. Google was trying to build, in many ways, the replacement or second generation of email. That's not something that flourishes in a walled garden, no matter how big the garden is (and Google's garden is pretty big).
Aside from the fact that Google could have really used outside suggestions on the design of Wave's UI, Wave sank in large part because it was a collaboration tool that didn't have enough collaborators. When I finally got a Wave invite a week or two after the launch, I could only collaborate on documents with a select few people. At the time some members of the openSUSE Marketing Team really wanted to try Wave — but we couldn't get invites for everyone who'd want to participate, and it seemed ridiculous to use a closed tool to help collaborate on marketing an open source project.
Now Google is doing the right thing, belatedly. But this is one of the few instances of too late, but not too little. The "Wave in a Box," is likely to live on and be refined. A few companies, like Novell, are still soldiering on with products based on Wave. And the buzz suggests that the community is willing to pick up Wave and carry it forward.
The lesson here is that Wave might have succeeded from the start if Google hadn't bungled its approach. If Google had launched Wave as an entirely, or mostly, open project from the start it might have had enough participation to make it useful.