There’s a lot of buzz going around about Google Wave, a web-based “communication and collaboration” tool going live later this year, but introduced to developers at Google’s I/O conference yesterday.
Google frames this new… thing… by asking the question, “What would email look like if it were invented today?” Wave is based on the idea that it would look very different – that email’s reliance on the same metaphor as “snail mail” is due to the state of the technology when it was invented, not any inherent advantage to communicating that way. So they’ve created a system where email, chat, wikis, bulletin boards, etc., etc., are all the same thing, and it all happens in a browser, and all in real-time.
You can reply to specific parts of a message (a “wave”) inline without having to re-edit the text yourself; you can add people to the conversation arbitrarily as it goes on (and they can “replay” the conversation up to the point they joined to catch up); you can send private messages to certain people, and it all goes from your browser to their browser in real time, no refreshing, etc. Pretty slick.
The demo is impressive technically, and shows how much can be done with a HTML 5 web app. The crowd ooh-ed and aah-ed and cheered all over the place as they showed off drag-and-drop from the desktop, blog integration, and the whiz-bang spellchecker.
Tim O’Reilly is very excited about it, and I’m starting to come around a little. The keys to success will be that Google is making Wave open-source, with a rich set of APIs, and most importantly that Wave is also a protocol, so Wave services built by different companies can interact with each other and with the main Google Wave product. Google is trying to set a new standard for communicating and collaborating online, a hugely ambitious goal even for them. Wave does have potential, though, and Google has clearly put a lot behind it and thought it through in ways that give it a chance to succeed.
The keynote and demo for those with 1:20:11 to spare..
For tons more detail and examples, check out TechCrunch and O’Reilly.
While I don’t look forward to Google getting yet more information about people, I predict this is what will replace Facebook.
Well, what should be different if email were reinvented today is:
1. Based on top of certificate technology to ensure that there is always a verifiable chain of attribution. This would shut down pfishing and email viruses.
2. There would be a small computational cost assocated with sending an email, in the form of hash-cash or a similiar system. This would shut down spam.
But given the current Web 2.0 / social networking / splashy colors obsession, what Google seems to think should be different is that it involves lots of chat and video features, and of course that it all flows through Google servers.
I want this now.
NOW, GOOGLE.
Obey Baconcat. Obeyconcat.
I’m not sure hashcash would eliminate spam, even it could be implemented everywhere. Zombie networks have lots of computing power.
It does look very interesting and I think it has great potential.
@2 – Yes, of course email should be more secure and free of spam and other nuisances. There’s no reason to think that Wave would be subject to these things, so if (and that’s a very, very big if) Wave were to supplant email in any significant way, it would have accomplished these things. Of course, any system can be, and will be, abused.
And yes, what’s with people’s current obsession with technology having good UI, working well, and fitting into their lives nicely? It’s clearly just a fad.
Wave being open-source and open-protocol means it won’t necessarily have to flow through Google’s servers, though of course most of it will.
For anybody who’s ever used “track changes” in word, this sounds like a headache and a half.
When do we get robot servants/killers?
The spam industry is studying the Wave API diligently. Lots of external hooks = lots of annoyance potential.
Blah meh @ 4: Bot-nets are often brought up as an argument against hashcash, and they certainly do reduce its value as a spam-fighting tool. But they hardly eliminate its usefulness.
A good implementation would allow me to set the cost to email me, and to set different costs for different senders (remember, we are coupling this with a certificate-based identity verification system). So I could let good friends email me for free, let businesses with whom I work email me for 1 second of CPU time, but set the cost of email from others to 10 minutes of CPU time. An old friend trying to get hold of me would pay that. But you can be sure that a bot-net operator, used to sending out 10 emails/second from a bot, is not going to spend 10 minutes sending one email. Even bot net operators don’t have 10,000x overcapacity. (And of course, to the extent that the operators do try this, the bot’s owener is more likely to notice that his PC is not just slow, but entirely frozen.)
i’m not sure i want my IMs treated like emails. i know that’s not the whole idea….
but the interface seemed a little to free form… i dunno…. i think i’d prefer a visual map of what i will call the “wave thread” (google wave seems to have the messages forum/facebook style). with as many messages per day, with the various permutations, the example interface would get cluttered way too quickly. also, the main display has too much going on as well… what’s with the up to three pics? that sort of stuff just gets in the way, and makes it more difficult to mentally categorize information.
despite my opening statement, i do like the idea of a tracked office communication system. but what about all the snide remarks i IM on a regular basis?
Anthony @ 6: “There is no reason to think that Wave would be subject to” these problems?! Are you Google-ly eyed?! Any system that allows someone you don’t know to send a message to you without your permission and at no cost to himself, is subject to these problems. There is, on the contrary, no evidence that Wave has even attempted to deal with these problems.
groove, gist, xobni, live services…wave. the way people work and think has to catch up before these catch on.
@12 – True, I suppose, but it appears from the demo that a Wave is basically invite only. We’ll see what it all looks like when it’s released, but from what I’ve seen, it’s not accepting random messages from anyone. The main thing, though, is that any system can be abused and used to annoy people with unsolicited messages. Google is extremely good at filtering.
David @ 10: 10 minutes processing time per email? My computer would be frozen just from the emails *I* send (I guess maybe I use email more prolifically than some). That’d be much more of an inconvenience to me than the increased network costs associated with spam, and relatively minor inconvenience of personal spam filtering. I do agree that it’s possible that hash cash could reduce spam, but I don’t think it’s going to eliminate it. I’m only willing to spend so much to do that, though.
Back about Wave, one thing I *love* about the idea (though I’m wary of entrusting my data to Google), is that it’d finally let me archive all my messages in a common place. Being something of an OCD backup-fiend, I currently have to use separate systems to archive email, SMSs, and IMs.
Bleh Meh @ 15: 10 minutes per email (actually per recipient) if that’s what the recipient wants to charge. The whole point is to put the recipient in control, not the sender.