[jerf]

I know the protocol is a bit complicated, but didn't they say a while back they're going to open source a reference implementation of the server? Is this it?: http://code.google.com/p/wave-protocol/

I do not know if that's a complete Wave server, or if it's missing a piece. But if it is a complete Wave server, there's nothing to "save". It's already been saved. You just need to run it yourself, and possibly support from Google for exporting your data once. If it isn't, maybe some people will fill in the blanks. If nobody steps forward to fill in the blanks then the demand probably wasn't there in the first place.

[derefr]

What people want to "save" is Google's backing, proselytization, and continued investment into the project. The demand was never there—because Google never gave Wave a big marketing push, the way they did with, say, Chrome (no Wave TV commercial?) Wave is a new and unadopted protocol, and as such it won't grow organically; it needs a big company to force it down people's throats. People want to "save" the unfulfilled (implicit) promise that Google was going to be that company.

[jbooth]

That's terrible reason. If it was a startup, they'd either sink or swim. Maybe better marketing would have helped, maybe the client sucks, maybe the concept sucks. But there's no reason for Google to throw good money after bad, it's not like there's nothing else they could be working on (including the next, better Wave from the etherpad guys).

[derefr]

> But there's no reason for Google to throw good money after bad

That's the thing—they haven't thrown any money at the problem at all. The whole problem is that it was a half-hearted effort from the start; Google didn't put Wave anywhere visible, they didn't integrate it with Gmail or Google Talk, or do anything else to get traction. They just sort of put it out there and hoped people would subscribe. That's how products work, but Wave isn't a product, it's a technology—and you have to sell a technology, company by company, until it's in use in a sufficiently large user-base that it becomes self-sustaining.

Imagine if the concept of "electronic mail" was invented today. You couldn't pull that off as a startup; you'd have to be Google-sized to even get off the ground.

Now, what Google could have done, would be to go to Microsoft, Facebook, Yahoo, and whoever else that has any product or service that's vaguely message-/chat-oriented, and offer to help them rebuild that product/service on top of Wave. Wave Facebook walls, Wave MSN, Wave Flickr, etc. Just making one crappy AJAX client is exactly not the winning strategy.

[roc]

> "Imagine if the concept of "electronic mail" was invented today. "

I don't think it'd be all that different from launching the concept of microblogging or social networks. Startups did fine with that.

What Google should have done, having a technology on their hands and not a product, was make their other products either built on Wave or compatible with Wave (Docs, Chat, Mail - using Wave. Calendar, Pages integrated via robots, etc).

Then, you'd have a huge built-in user base that can ignore the complexity until they grok it and if they want it. And all their data will be waiting for them.

In the meantime, Google could develop and throw robots into their products as features. Being able to directly send messages to a robot for publishing on your blog platform of choice, or directly drag attachments to a robot that populates Dropbox/Flickr/whatever? Being able to add a plugin to schedule a party into an email chain that automatically updates Google Calendar? Having a service that detects tracking numbers and provides mouse-over summaries?

Even for users who would never want a Wave-like client, those features would make Google's existing products better and stickier. And none of it would involve burying every would-be user in complexity on day 1.

[jbooth]

Well, sure. But maybe if the one client they had was better or the protocol more natural it would've had more uptake. Or maybe there's just not enough of a need for those rival corporations to take up anything wave-like.

At this point, considering the Etherpad acquisition, you have to figure they're scrapping it and starting over -- not knowing anything about the situation I'd generally assume that was the right move.

[hs_100]

Well said!

If a company like Google is not going to invest in such technologies who will. This makes me think if Google is turning into a company which jumps on to a bandwagon once it starts rolling rather than be a creator of new ideas and technologies.

[elblanco]

I haven't run it, but by the description installing this software will get not you a whole heck of a lot closer to having your own Wave implementation in terms of server-protocol-web client that I think everybody is envisioning.