Is OpenSocial open enough? The problem with OpenSocial, Google’s new platform for social-networking apps, notes Tim O’Reilly, is that it doesn’t go quite far enough. It lets applications out from the confines of any one Website, but it does not let the data out. Apparently, you cannot mix and match data from more than one app to create a new social app. This is wrongheaded, argues O’Reilly:
It’s the data, stupid.
We don’t want to have the same application on multiple social networks. We want applications that can use data from multiple social networks.
That is an important point. You cannot have a proper mashup without complete data portability. Programmers would prefer to have access to the underlying data powering these social apps so that they can create their own new social mashups. Such new social apps would also benefit consumers who would have more social apps to choose from, and broaden the market overall.
Unfortunately, the business models have not been worked out yet to accommodate such mixing of data. If a social mashup starts making money from ads, how would that be split up between the host site, the app developer, and all the other applications or social networks from which that mashup pulls data? O’Reilly doesn’t really have an answer for that one.
Under the OpenSocial rules, it is the host sites who make the rules. The more players involved in any given app, the more complicated the economics—unless they take the Facebook approach that data can move everywhere, but you only make money on the pages you control.
“We want applications that can use data from multiple social networks.”: YES, YES, YES. This is what i am fantasizing.
HAT’s of to O’Reilly: The only guy who understands depth of WEB.
Freebase can solve this problem, i think.
Why shouldn’t facebook and myspace open port 3306 and give us all anonymous access????
That’s what your essentially asking for Erick.
NOW, YES NOW.
Why doesn’t TECHCRUNCH open up their wordpress DB on 3306 and provide US with anonymous access???
Huh? well….
…. I’m waiting….
hmmm…..
Perhaps that’s why Google searches often state that there are millions of results but you are only privy to 10 pages of them? Didn’t ever notice that did you?
So, I have fresh copy of phpmyadmin ready. When can we connect to the Techcrunch DB and start freeing the data away ????
or doesn’t that work when it applies to you guys?
It’s all about the marginal cost — the spread of the income you bring in versus how much you need to pay to get your product out there. We are going to reach a point where so many people who have their hands in the pot, that income generated from “mass media” advertising will not make the time expended creating widgets worthwhile. This is the inherent flaw in Open Social that I see at this point. Furthermore, I don’t understand why everyone wants the data to be free. Is this because it gives everyone an equal opportunity to establish a competitive advantage? IMHO competitive advantage is exceptionally difficult and a greedy way to generate income, creating contempt from your competitors who are out to see your eventual demise.
Yes, “Open Social” is not an appropriate name for this initiative. Users will one day, through a new model application (like ours
) be able to let their profile data loose in a ‘open’ and knowing manner, finally fixing this absurdity of writing the same profile stuff over and over. Because the big players are now too fearful to let the data go ‘free’ and travel around, least they can’t monetize it, they are forgoing their chance of being the source of the portable profile. Necessity will win this game. They should try to monetize their destinations as experiences. That would be fair, and ultimately wiser.
You guys make this so complex. If the issue is not re-entering your profile data over and over again on different sites, why not create a form aggregator? You enter the data in once, and the forms will populate on any site that you register for. The host wins because they get your data to monetize on, and you win because you save time. Does it REALLY need to be this difficult?
Facebook does this, but once you enter your info on their site, they use it to power third party applications. Is it a “walled garden”? Perhaps, but if it works, its users are happy, and fresh widgets and applications are being made, why try and reinvent the wheel?
I swear, the constant focus on the newest technology will have you all running around in circles while those that can decide on a platform are moving forward.
Profile data can never be let loose…there will be privacy issues…
http://www.meetingflex.com
If you guys saw what I’m working on right now you would go absolutely bananas.
Sorry Erick, I was just playing devil’s advocate.
I’m a nametag wearning offender.
Open data would probably be the most exciting phase as far as power of social networking is concerned. That will allow true killer-apps to come in but then it does lead to a lot of issues…
1. How do social networks lock users within their sites? I currently stay on orkut since all my buddies are there – even though i know facebook is wayy ahead. Something like free moving data may allow those apps to be built that would allow me to be on facebook and still connect to orkut folks. While, for user, such capabilities would be amazing – social networking sites are scared of loosing users to one another.
2. As meetingflex already pointed out, privacy issues are always going to come up.
3. As article points out – business model for this stuff needs to be worked out.
Anyways, sooner or later, this has got to come.
Google for now – I agree didnt get it right.
O’Reilly is a communist. Data = money. There is no way companies are going to give this away. If you can get the data you can move it off site, and if you can move it off site, you are stealing page views from the social networks. Regardless, Google wins either way.
I’m sitting here LMAO and I can’t tell you why. It’s driving me insane. arghhhh…. Stupid capitalism….. I’ll reveal all at TechCrunch in Boston, when it’s far too late.
Another OpenSocial post?
I wish TechCrunch would pay more attention to sites that are useful and profitable, like the the ones in the online dating industry.
Well, O’Reilly wrote his post after I wrote mine, but I still disagree: it’s not about data, it’s about distribution.
http://20bits.c...ibution-stupid/
Most users don’t care about data portability. Indeed, outside the technology world I doubt it’d even be on people’s “top 10 lists of things wrong with Facebook.” The ability to activate a large userbase using some sort of distribution system (in Facebook’s case the newsfeed) is what will win the game.
In that respect OpenSocial is, at best, unproven.
I wonder how this will effect thefacebook… considering social networks have a lot of money in them and the data they produce is worth millions… how will this affect the users and or developers who will incorporate this system…
Every one is looking for ad. revenue and the people in the network are going to loose privacy. Every information in a social network is going to be sold and there will be good and BAD things as well because of that.
It will be a fun shortwhile!
@Chris …. you didnt get what O’Relly was saying. Did you? The data in social networks is data about you, your profile, your friends. What kind of data does techcrunch have about you except your not so funny comments, that they would open it up to you.
The only company so far that successfully pried websites from their data and fended off lawsuits is Google. They’re about to have unwelcome company real soon. Peace to everybody!
Let’s see some “open data” proposals that include a business model to support it!
@17 I wrote in a later comment that I was simply playing devil’s advocate. People think too much about concepts. Where’s the “API” for this, ect….
Remember when everybody coded in assembly language for DOS? There was no api, just a few calls to reserved memory locations in the RAM with function/returns on the stack. Nobody made it nice and neat to do anything.
These web 2.0 programmers are the equivalent of people paralyzed in wheelchairs. They can’t so much as print something to the standard out without there being some highly documented object oriented framework to do it. It’s a laughing matter.
Then they wonder why other companies fly by them at 100 MPH and they’re standing still.
“It’s the data, stupid.”
No, it’s the code stupid.
Social networks could easily monetize this by charging mash-up sites a per-transaction fee to use their API (for example: $0.0001 per hit), or charge for bundles (1 million hits for $100).
http://www.goog...G=Google+Search
“Social networks could easily monetize this by charging mash-up sites a per-transaction fee to use their API (for example: $0.0001 per hit), or charge for bundles (1 million hits for $100).”
Google seems to be using 1M+ pages from any given SN site to drum up adwords, without paying them a dime?
You guys are absolutely STUCK on social, networking. It’s not about social networking. SN is just a concept. In terms of software it’s meaningless. Think back to the days of PCDOS and how people were able to do so much with so little. That’s what makes Google so great.
I’m actually working on an application that’s leveraging existing ideas and standards to help you pull your entire social graph into one place so you can reuse it wherever and however you like. OpenSocial may not go the entire way, but a single standard API with good uptake would be a huge time saver for us for the simple fact that we don’t have to write a custom scraper for every single site we want to support.
We’re still in early alpha, “getting dressed in public,” if you will, but we’re all about being a part of the conversation and meeting real user needs. So feel free to check us out and offer some feedback. http://mywhurls.com/
What if a search engine, like Google, was so smart, so intelligent, that it could read web pages like you can? What if it could logically organize data like you can?
What if you could ask it questions?
Wouldn’t that essentially free the data not only from SN sites, but from everywhere?
That must be so complex. It has to be impossible to code something like that. Only a really truly awesome code GOD could do that. Like awesomer than Larry Page.
I have been saying this since the beginning. Just read my blog, “OpenSocial” is not open, it is just a glorified Google widget.
Chris? What do you mean 10 pages???
It is amazing to me that Google got all these top sites to give their user’s social graphs away. How long before google’s uses that data to their advantage?
http://www.opensocializr.com
OpenSocial Reviews, Code & Community
“Chris? What do you mean 10 pages???”
“Sorry, Google does not serve more than 1000 results for any query.”
I used 10 as an approximate generic example. They do not let the users pull any more than a hard limit of 1000 to be precise. There are a lot of reasons for this which I will not go into, but would like to if time allowed.
Google can’t be pushed because no one is willing to push them. No one is willing to give more than Google gives. Google has all this data neatly stacked that Erick is looking for. They have A LOT of data neatly stacked from the entire web. They will never let it be exported as raw.
Companies will only do as much as they NEED to, to continue making money. Only when a new guy comes along and pushes the bar up do they change. Facebook did it for silly Social networking API. Otherwise the other sites would have NEVER EVER, but NEVER EVER have joined an initiative like open social.
Somebody’s about to push Google in the ass in the same way, and I can’t wait.
Let My Data GO,
Privacy is exactly the point. Context is also the point. It can’t be assumed that data made available to one network by a user would have willingly been made available to another. It would make a mockery of privacy controls, for one. Is there any personal data on Facebook that *is* entirely public? I’m a believer in openness when it’s appropriate, but it’s flat out evil when it’s not.
“It would make a mockery of privacy controls, for one. Is there any personal data on Facebook that *is* entirely public? I’m a believer in openness when it’s appropriate, but it’s flat out evil when it’s not.”
Does information showing up in the Google cache mock privacy?
I don’t think it does. Most of the profiles on the major SN sites are there.
Are we talking about FB or the sites on the OpenSocial network?
Sites flagged as private can’t be exported or you have control of what can be exported.
The point of the article was the limited interoperability between the partner sites on opensocial and the internet in general. They’ve created their own walled garden as Mr. Robert Scoble likes to say.
The walls of everything are about to be blown apart by a cannon.
As a linkedIn user, I don’t want my profile data shared with facebook, myspace and plaxo users. I would be the first to terminate my account in LinkedIn if OpenSocial were proposing or forcing this kind of interoperability.
However, as a Flickr user I do want to be able to move my photos to whatever other service I want, and I am glad that I am already able to do that using Flickr APIs, the same is true for Ning and Freebase that have have their public APIs, and already allows cross-network use of the data, the two remaining issues left to solve are: 1- standardization of the different data access APIs, and 2- A way for you as the user to selective authorize which data from which networks you trust to be shared with which other networks.
The first one OpenSocial can address, the second one maybe OAuth. The fact that one sn company in particular locks your contributed data (for instance, an app that generates rss for your facebook photos is AFAIK against their TOS) doesn’t mean that the web as a whole is not open.
To “let the data out”, to “Set the data free!” and to “Allow social data mashups” as this article purposes, is a decision that should be made by each network owner/company individually, not by a standardization initiative (opensocial).
As a user of course the best you can do is to vote with your foot/wallet, and give preference to services that already let you do whatever you want with your data.
Madness? This is DATA!!!!!!!!
Leave!! My data!! Alooooone!!!
I am SO sick of Chris (from BeerCoSoftware) and his using this blog as a freaking running monologue. Dude! Launch your lame product, see if we care, but enough of the blah-dee-blahblah-dee-frickinblahblah.
Lame.
My point is that the data is already available. People are stupid and lazy and they can’t go get it. Google can and did.
O’reilly should shut up and stick to publishing. I am sick of writers writing about these concepts as if they are hard and fast and absolute.
Wahhhhh, so and so didn’t export this. And Facebook is oppressing Ron Paul. This data he wants is readily available via any TCP connection. So what if Google won’t let you get it from their servers.
It’s like complaining that somebody didn’t make my bed this morning. Who was *supposed* to make it? These web 2.0 programmers need a serious kick in the *ss.
We need OpenID + OpenSocial + Open Source Containers.
Maybe we can tag data with privacy metadata.
OMG, my username changed to Asshat. There is clearly a cookie error here with Wordpress.
I am not #35. This is really messed up.
Tim is right. Hard not to blow edgeio’s trumpet here. We have 20 million items from about 7500 listings publishers that are able to be sliced and diced by keyword, tag, vertical, geography, provider. These are freely available to creat classified ad sections or drive classified search on any web site.
API here:
http://www.edge....com/search-api
Bind this to “open-social” and you can get a feed of people and their stuff and slice and dice it in many ways.
Keith Teare
ceo/founder/edgeio
I’d just like to see Mr O’Reilly propose what a social data mash-up might comprise and how it would actually benefit users and not just fulfill a wet dream for the data miners and monetisers.
As Fabricio put it:
‘To “let the data out”, to “Set the data free!” and to “Allow social data mashups” as this article purposes, is a decision that should be made by each network owner/company individually, not by a standardization initiative (opensocial).’
wp-comments-post.php, the server sent me bad cookies. That’s not cool.
People care about their privacy.
Asshat from #35, did you send the bad cookies to my browser through a hack?
I am looking through the WP code to see what could have done that.
The setcookies() function is in wp-comments-post.php
I don’t see how that could have happened. What did you do when you posted?
The endgame of social networking is a p2p structure that is totally centered around each individual. There would be no need for any centralized data storage, except for a common directory where you can look up people you don’t know. People would have total control of their own info. They would decide for themselves what info would be accessible to other people, as well as search engines.
yeah, I’m not going to your event in Boston anymore. I don’t deal with people who send malicious code to their users. Have a nice time. I am filing a report.
Quoting from Tim’s article
The audience member asked something about building applications that can remix data from the participating social networking platforms. Patrick’s answer was along the lines of: “No, you only have access to the data of the individual platform or application.”
I am confused. If I can get data from individual platform, why couldn’t I remix it on my app?
It’s funny that people are just now picking up on this, when it seemed clear from the initial press releases. Way too many people jumped on this bandwagon without looking at it first. It seems clear that OpenSocial could grow into a common data sharing technology, but it’s certainly not there now.
Hey, one good thing came from me wasting my time reading that blog from Tim: now I know to NEVER listed to anything that dolt says.
…er: liste-N
Unfortunately, the current business models have not been worked out yet to accommodate such mixing of data. I don’t think it will work out.