Slashdot Mirror
Google Open Sources Wave Protocol Implementation
eldavojohn writes "Certainly one of the most important steps in adopting a protocol is a working open source example of it. Well, google has open sourced an implementation of the wave protocol for those of you curious about Google's new collaboration and conversation platform. It's been reviewed, skewered and called 'Anti-Web' but now's your chance to see a Java implementation of it. The article lists it as still rapidly evolving so it might not be prudent to buy into it yet. Any thumbs up or thumbs down from actual users of the new protocol?"
It clearly can't be anti-web.
Too confusing. Requires a browser. Won't run on my iPod. Lame.
The CB App. What's your 20?
here from 4 days ago
http://en.wikipedia.org/wiki/Google_wave#Product
That wasn't so hard, now was it?
this
I've read reviews of it as real time collaboration. Think of it as private e-mail, IM, and document collaboration all in one system.
-- "Freedom is the right of all sentient beings" -Optimus Prime
I for one feel bad for the helpless telecoms. They have done everything in their power provide nothing but exceptional service to customers over the past 30 years, including protecting our privacy and investing in infrastructure ~/sacrasm. Regardless of 'infringing' business models, we should be rejoicing the opportunity to compliment the current, and broken, communication model. By providing an alternative protocol with specific functionality it's not replacing current technology, simply enhancing it. Let's just hope it's not a product of the PR machine.
What is The Wave's motto?
Strength through discipline, strength through community, strength through action!
Google chose a very fitting name.
If you have something that you dont want anyone to know, maybe you shouldnt be doing it in the first place -Eric Schmidt
http://en.wikipedia.org/wiki/Google_Wave_Federation_Protocol ?
Beware: In C++, your friends can see your privates!
It seems to be a different approach to the same problem, with Croquet using distributed synchronization of computation rather than synchronized distribution of updates.
Many people have responded to your post with links, but I know people are really, really, really lazy. So Google Wave is kind of a nifty new communications paradigm designed to replace e-mail, IM, IRC, and other collaboration tools. The basic idea is to create communications centered around a conversation with as many participants as needed, rather than trying to take a two way communication like a letter and expand it to sort of work for more people.
If you're the only person in the conversation (or wave) online, it works like e-mail. As soon as a second person is online at the same time, it works like IM. It is sort of timestamp version controlled so you can rewind conversations and see how the conversation branched and you can embed the conversations in generic Web pages. It's extensible so you can add additional communications to it, and they've added a way to post images and host them as photo galleries.
In short it's new, but similar in ways to IM and e-mail and it's fairly cool, but watching a video makes more sense than reading a lengthy explanation.
It defines a protocol that allows servers to publish documents with threaded conversations, and allow users on different servers to edit those documents and append to the threaded conversations in real-time. It also defines an API that lets developers extend the kind of media that can be placed in the documents, and make documents interactive with the user or other services. It also uses a messaging semantic based on operational transformation, that allows users to browse the complete editing history of any document or thread, and allows agents observing a document to resolve their local state by reading a document as a stream of deltas (it's more complicated than this, but I have yet to wrap my head around OTs completely).
People say it's like email because it lets you do messaging in non-real-time, and has threaded conversations, and documents and media attachments, and it's an open standard. People say it's like IM because conversations are posted to threads in real-time, keystroke-by-keystroke. People say it's like Google Docs (or other such things) because it allows collaborative editing of documents, except this lets you edit the document contemporaneous with other people, since the server protocol merges all updates to the document keystroke-by-keystroke.
Don't blame me, I voted for Baltar.
Whatever you do, don't read this
http://sites.google.com/a/waveprotocol.org/wave-protocol/draft-protocol-spec
I tried reading it and its like the South Park episode with the Marklar, only replace Marklar with Wave as the only Noun/Verb in the language.
Its an adressible service like email or newsgroups, users have usernames @ domains and can subscribe to or send content to lists/groups.
It has a collaborative aspect, parts RSS feed/Twitter/Wiki and I think it will be easier to understand when there is more content availble. I do wonder how the providers are supposed to keep these documents like newsgroup retention, with conversations dissappearing after a retentionwindow ends, or like hosting wikis. I don't see where I would download any waves that I was participating in or monitoring, but I'd have to see an implemented client I guess.
As indicated in the comment at the top of that file, that code was generated by the Protocol Buffers compiler, protoc. You aren't supposed to edit that -- edit the .proto file instead and regenerate. I'm not really sure why they checked the generated code into VCS -- normally only the .proto would be checked in and protoc would be invoked at build time.
I think that every web developer that misses this out, will pay it hard.
Experts say that true innovations are hard to detect. I would say, keep an eye on this, or you will regret it.
Waves are exactly as local as email is. WFP sits atop XMPP (the protocol that runs Jabber). Waves do not reside "out there"; they reside on your XMPP server. I would expect any organization using Waves to maintain its own XMPP server or 3 (but I have seen stranger things).
WFP isn't perfect, but if you're going to complain about it not residing on your local machine, you'd better be prepared to make exactly the same complaints about email. Personally I think email has proven itself to be a plausible communication medium, even occasionally for mission critical work. WFP has the same potential in a few years time when it's stabilized.
Err, try again. The whole point of wave is that google are open sourcing the spec, and plan to release an open source *server* reference implementation.
The concept of wave servers appears to be similar to that of smtp email. Companies can run their own internal servers, and configure links to the outside world as needed.
I assume you've tried signing up? You should be able to develop something however if you want to get a peek.
No, it is not necessary to run on a Google server, you can run your own server. Check out http://wave.google.com/ for info. Wave looks awesome - seriously awesome
If you can read this... 01110101 01110010 00100000 01100001 00100000 01100111 01100101 01100101 01101011
just like "clouds", "waves" do not reside on your computer, but rather *out there* somewhere, that you can *probably* get access to if:
-the service is up and functioning properly
-you have the required hardware and software
-there are no connection issues between you and the server
You can set up your own wave server, just as you can with e-mail.
if your internet goes down, suddenly you've lost access to even internal communication at your office, as well as all archives and logs of past communication. Without local storage, you cannot do efficient search and retrieval of your own information.
Companies can set up their own wave servers and communications between members of the same server will never leave the network.
there are serious privacy issues as well, no doubt google will be surfin the "waves" looking for terms to market to you, but perhaps it is more shady than that even. google has agreed to censorship in foreign markets over the years, does it really make sense to let them hold onto your data in this way?
Yeah, they can - on their own server which will probably become the most popular one but you can use alternate servers to those of Google.
then again.. it's cool technology, and now that it's being open sourced, it means feasibly you can run your own "waveserver" and mitigate the issues above somewhat.
Not somewhat but pretty much equally to e-mail.
I think the whole idea of wave is AWESOME. My one question is ... how is Google going to make money off of it???
If you can read this... 01110101 01110010 00100000 01100001 00100000 01100111 01100101 01100101 01101011
Anyone know how on earth Google expects to make money off of this thing? It looks amazing, but how do you make money off of this if it's open sourced, free, and took a ton of development time to build (and presumably support in the future)?
Obviously they plan to make money the same way they do with GMail. They'll offer a free in the cloud service to normal users and either provide ads alongside the client and/or robotically harvest the conversations to better target online ads to their users. They might even sell corporate Wave hosting services to corporations or sell servers with it pre-installed and ready to go to corporations.
By yet again re-enforcing their brand image as being synonymous with the web.
but what good would that do you, if it's an ever changing document, like a conversation between multiple people?
I think the whole idea of wave is AWESOME. My one question is ... how is Google going to make money off of it???
I doubt if Google will make much money off of the wave protocol or message format... Much like SMTP, it'll just kind of be out there for other people to implement.
I'm sure they'll offer a free (ad supported) Wave service however, much like they've got Gmail now... And they'll probably offer a paid subscription to business users, like with Google Apps right now...
Of course, they're spearheading the whole thing... So they could probably get an actual wave server (hardware/software/whatever) to market long before anyone else can. They could sell that, but it doesn't really sound like Google's way of doing things...
"Work is the curse of the drinking classes." -Oscar Wilde
Google is winning because they are as smart on technical stuff as they are on getting money from advertising without pissing of users.
Make no mistake, whoever is in charge of ad marketing in Google is a pure genius.
Love many, trust a few, do harm to none.
Wave is surely an interesting concept and application, but if there's any web app that just makes you want to scream for a native implementation, it's Wave.
I think focusing on making one cross platform Web application that can be embedded into Web pages is probably the most effective use of their resources. No one is going to bother downloading a client unless there is some significant use of Wave first or it is being deployed in a corporate/large organization setting. Google needs to get it out there and a Web app makes a lot of sense as a first attempt.
Why is Google spoiling good concepts by tying them to the browser exclusively? They just need to develop for the three major platforms, Windows, Linux and OS X.
Again, I disagree. For geeks, maybe this would make sense if Google had the resources to accomplish it at the same time as creating the Web application, but for normal users this isn't going to happen. Most users just don't install things like this or they'd have a Jabber client by now. How many people with Jabber clients right now do you think have ichat compared to all the other clients out there. What Google needs to do is push this as a Web app and then partner with other companies to get them to develop native clients to be pre-installed on their respective platforms and Web services. By open sourcing the Web client Google potentially gets AOL, Yahoo, and MS to expand their chat and e-mail clients both on the Web and the desktop. By talking to Apple they might get this on OS X and/or the iPhone by default. Someone will write a native Linux client no matter what Google does. There will eventually be clients for Windows and OS X, but very few people will use them if they aren't pre-installed with their computer or unless Wave really takes off on the Web first.
Malkovich, Malkovich Malkovich Malkovich....MALKOVICH!!!
I definitely agree with you. A rich client, maybe implemented with C++ and Qt4, would be very useful. The demo video actually shows a native command-line client for Wave. If that's possible, you should be able to develop any kind of interface. If Google doesn't release a thick client, maybe that's a business opportunity right at your doorstep.
It's not currently a service, so you're point still isn't valid. Yes, they have ONE server up. It's a BETA. An INVITE ONLY beta. Where did you get the idea that someone is putting mission critical things there? Now that this reference implementation is out you can even have your own server running NOW, before Google's even hinted at running their own service.
Troll harder please.
I tend to agree--I don't particularly care to have my email/IM/collaboration software all tied up into my browser (particularly in FF or IE). I want something that does one thing and does it well. But if the protocol is open, as far as I can tell (note: IANAD(eveloper)) there's nothing stopping anybody from building a nice lean, writing-focused Wave client.
But isn't that what Chrome OS is for? Ultimately, Google wants the browser and the OS to be identical.
how is Google going to make money off of it???
Keep an eye on the various "Robots" and "Extensions" they'll be offering as services.
Also, destroying the competitive advantage of Exchange and Lotus Notes will have certain long-term strategic benefits.
Don't blame me, I voted for Baltar.
Why develop for three platforms and let geeks port that to many when you can develop as open source for one platform (DHTML) and actively encourage the geeks to port that to 3, and then to many?
Hell, if I worked at Google the last thing I would want would be to get involved in GNOME/KDE turf wars, piss off apple fanbois if it doesn't look precisely like a macintosh app or really develop anything for the Windows desktop. Like, ever. So instead, Google puts it on the Web where everyone can get at it from any modern platform (even/especially smartphones, and if you want a native copy then you or some hobbiests are free to write one (and thereafter support it :P)
Seriously, do you recall their last attempt at a Jabber desktop application, Gtalk? It even worked well, but then they dropped it like a bad habit in favor of the web-based version.
Additionally, one of their main design goals was to make Wave conversations embeddable into web pages. They would like this to be used for CMS, to replace forums, to replace blog comments, essentially they wish people to mash up Wave content with their own web pages. If they focused their deployment first to the desktop, they would miss out on that opportunity.
People willing to trade their freedom of expression for temporary entertainment deserve neither and will lose both.
What question?
Its a published protocol built on top of XMPP, with a defined data model; nothing is stopping people from building native apps that produce and/or consume wave updates. Certainly, Google is doing nothing to prevent this.
Or, you know, they could just publish the specs and leave implementation of native applications, which are far less relevant to Google's business than web applications using the protocol, to people who have an interest in and use for native applications.
Nah, Microsoft would have several separately licensed and managed components. They'd call them something like Exchange, Live Communications Server and Sharepoint.
Calling someone a "hater" only means you can not rationally rebut their argument.