SmoothSpan Blog

For Executives, Entrepreneurs, and other Digerati who need to know about SaaS and Web 2.0.

Archive for November 2nd, 2007

Google is in the Cat Bird Seat for Identity Matching in OpenSocial

Posted by Bob Warfield on November 2, 2007

David Berlind ponders how identity is handled in OpenSocial.  Specifically, how do we map identity from one Social Network “Container” to the next?  If I query “Open Social” for friends, it sends me back a text based name like “Cobra427”.  That name might be “Bob Warfield” over on another site.  How can I tell they are the same?

I had speculated previously that the application knows the names and that it can do the mapping between networks to the extent that all of your friends use the same OpenSocial application.  I called this “leaking” information across Social Networks and speculated one could even build an app to help move graphs between networks.

It turns out there will be an Uber Leaker in the whole system.  It should come as no great surprise that this pivotal role is played by Google.  In looking over the OpenSocial API’s, I was struck by the authentication scheme.  Before you can access any private information, and the Friends list is definitely private, you have to be authenticated to the API.  There are two ways to do this.  One is you can logon with your email and password each and every time you use the application.  That stinks.  And details are unclear of how such “accounts” are established.  Google recommends that approach for desktop applications.  Perhaps that’s intended for disconnected operation.

The preferred approach is to use your Google account, a mechanism they call AuthSub.  Can you see where this is going?  If everyone who uses an app has a Google account, and is logged in while they use the account, Google has the treasure map.  They know how all the names link up across all the Social Networks.  Pretty cool, eh?

The identity fabric people are searching for boils down to Google accounts.  The API’s map all your Social Network info back to a specific Google account.  Now Google is in a position to build out an extremely sophisticated and varied collection of information on anyone who has a Google account and participates with an application on Open Social.

I’m sure the conspiracy theorists will have a ball with that one, but let’s be real, if you want to have a meta-Social Network, you have to do something like that to link things together.  You’ve already got individual Social Networks madly gathering dossiers on everyone that they type in themselves.

What am I going to do about this?  Note to privacy buffs:  you’re going to need a lot of Google Accounts.  In fact, maybe you should never use the same one twice.  Note to self:  call broker and buy a few shares of GOOG before everyone figures out where this is going…

Breaking News Update 

Despite what the current preliminary documentation says, it is possible to invoke OpenSocial without using AuthSub or having a Google account.  How do I know this?

2 reasons: 

First, Al over at Folknology asked the question over on the forums and left me a link to the answer.  Thanks Al! 

Second, after reading a new post by Marc Andreesen, I created my own Ning social network and checked it out.  I installed the iLike application as an available app on my Ning network, then I installed the app on my personal profile on the network.  At no time did it have access to my Google login.  Now in fairness, it also didn’t touch any of the profile or social graph info, so we still don’t know what really goes on there.  I think we’re going to have to wait until the api is more fully documented to tell.

BTW, Ning is pretty slick.  Give it a try.  It’s easier to create a new social network of your very own than it is to build the social graph.

Related Articles

What’s Next Google?  (You Won the Battle, Now What About the War?)

Ubiquitous Social Networks for Business

Posted in business, strategy, Web 2.0 | 5 Comments »

What’s Next Google? (You Won the Battle, What About the War?)

Posted by Bob Warfield on November 2, 2007

Google’s OpenSocial API for Social Networking applets looks like a fabulous coup, particularly after the surprise second shoe dropped and MySpace jumped on the OpenSocial bandwagon.  Suddenly it seemed that Facebook was surrounded with nowhere to turn.  But it’s important to remember that Google has only won one battle and not the war.

It seems that the applet platform aspect of Social Networks, at least insofar as this plug-in style of platform behaviour is concerned, will go to the Google standard.  But how important is that to the overall Social Networking experience?  Compete did a fascinating breakdown of what people do on Facebook.  Use of applications was in fourth place, behind looking at profiles, looking at pictures, and looking at the start page.  This is based on visits.  People did spend a lot of time in applications, and a lot of people tried them, they just didn’t visit them all that often.

It’s clear that the applications aren’t the main event, which means that Google has a solid victory in a battle, and a lot more battles to fight to win the war.  It’s only check, not checkmate, Mr. Arrington.  The potential of a truly open Social Network, or a meta-Social Network is huge.  I’ve written about some of the business possibilities it would open up, but it’s going to take a few more battles before that happens.  One of the biggest things will be portability of one’s social graph.  Can I carry mine from one participating Social Network to the next?  That’s a touchy business.

Another question is whether the apps can “leak” information between Social Networks.  Presumably they can take everything they learn through access to data on one network, ship it to their own mother ship, and then if you install the same app elsewhere, the info will be available.  Who will be first to write an app whose sole purpose is to carry your identity and Social Graph from one network to the next?  Who will be the first to put that data leakage concept to some nefarious business use?

What about the question of whether the platform will turn everything inside out?  Facebook and most of the others in this space have gone from compelling applications to platforms, a tried and true formula.  Things built on their platforms get referred to in the diminutive:  they’re applets or widgets.  But what if the Social Networks themselves go back to being regarded as applications that can be arranged and composed to personalize your networking experience.  Wouldn’t that be fascinating to consider? 

Let’s say you want to lash a bunch of these things together to make your own view of the web, your own portal to use that apparently antiquated term (though Yahoo are quite proud of it).  For example, I pretty freely interoperate between Google Blog Reader, Google Blog Search, LinkedIn, Google Search, Facebook, and my E-Mail.  I use LinkedIn when I encounter a name and want to know who they are or what they’ve been up to lately professionally.  Facebook sometimes gives me a more up-to-the-minute (and often inane) idea of what someone is up to right this minute and a little more profile color that may be missing.  The blogosphere feeds ideas and I tend to search there first.  If I don’t find it in the blogosphere (and somethings I know I won’t, so I don’t try) I go to Google Search.  I personally prefer E-Mail to make direct personal contacts.  You see how all this is fitting together?  Now some people try to do a bunch of that sort of thing inside Facebook, but I don’t see how it will ever be as rich as the whole Internet.  But yet, having the Social Aspects tying everything together would be really cool.    Can the Social Network piece be a thin mortar that cements all these app bricks together in powerful ways?  I think probably so.

Anything Google and the rest of the crowd that want a meta-Social Network can do to foster such visions will further their cause.  It’s really the opposite of trying to rebuild your ideal web experience inside Facebook.  Rather, the Social Net is just one aspect of the web experiences you already have, and perhaps it is a key aspect that ties things together via people.  That starts to sound just a little bit like pieces of the Semantic Web dream, doesn’t it?  Interesting.

Google has a long road to make that vision real, but it would be a big step up for the web.  Lots of ways for it to fail along the way.  Some of it is in Google’s hands, and some is in the hands of the participants.  There are third party issues too.  What will Microsoft and Yahoo do?  What will properties like eBay and Amazon do?  What should Facebook itself do?

I can’t imagine why Yahoo isn’t stepping up to embrace this thing.  It just makes them look clueless and unable to respond.  They sure can’t build an alternative, so why not jump in?  Seems like MySpace looks great for jumping in, and nice timing to follow the others too!  Yahoo looks like a deer in the headlights until they say something.

Microsoft faces a messy business and I would bet they shun the platform.  First, they’ve just made the big Facebook investment.  They must be smarting from all this.  I can already hear the distant spinning that all this OpenSocial stuff is a bunch of vaporware that doesn’t really mean anything.  Second, Microsoft almost never gets on board with an open standard, let alone an open platform.  Microsoft’s business is to own platforms.  Yet, how can the not saddle up all their technologies and make them work with this stuff?  It is truly nose cutting off to spite one’s face kind of drama.

I believe other web properties like Amazon, eBay, and even Netflix ought to be joining up.  They all have communities going, and could benefit.  I want to see eBay auction stats while I’m off Social Networking so I don’t miss out.  I’d love to be able to track down buyers and sellers on other venues too.  In some ways, I would treat being able to access them as adding to their credibility as much as some of the eBay feedback, for example.

As for Facebook, they got Uber-Face-Slammed since they claim they were never briefed on the initiative and Google doesn’t seem to dispute that.  That’s a bit of evil on Google’s part, but I suppose it was justifiable that they didn’t want Facebook to be able to pre-empt in any way.  Facebook ought to embrace the standard despite feeling sucker punched, but they should wait until they’re ready to release support to talk about it.  Here is why:  I can’t imagine an app developer not supporting OpenSocial or Facebook.  They’ll all do both.  So what is the competitive advantage for continuing to be proprietary?  Seems like they gain face (pardon the pub) if they adopt the standard.  The trick is to start thinking chess moves ahead now that they see how this game will be played.  What open API could Facebook kick off?  They certainly have currency to persuade others with.  Everyone wants to be a part of Facebook.  How can they parlay that into a successful counterstroke and win the next battle?  If they come up with something smart, maybe they should wait and announce that alongside support for OpenSocial. 

There’s definitely time for Facebook to cook up something good.  It will take Google quite a while to really get OpenSocial underway, it’s just an announcement (albeit a powerful one!) at the moment.  From a PR standpoint, I think it’s good if they let this all cool off a bit and be careful not to over react.  Michael Arrington asks whether Facebook will be forced to adopt the standard, and waiting a little while can help to dispell that impression.  But it seems to me they need to make their announcement before end of year.

Readers, what do you think the right master counterstroke for Facebook might be?

Related Articles

Google is in the Cat Bird Seat for Identity Matching

Posted in strategy, Web 2.0 | 7 Comments »