Identity Management Will Destroy Social Software

There's a meme growing on the web about social networks and identity... and I thought I'd inject my opinion before it grows into some kind of wretched mutant...

Firstly there was the long-standing opinion of Read/Write Web about the attention economy: concerning that the pages a person visits on site A is incredibly valuable to site B... Similar to how info about previous Amazon purchases would really help NetFlix, and vice versa. I liked their analysis, but I think they missed the point. Then there was the article in Wired about opening up Facebook: shouldn't Facebook be required to open up their APIs so people can copy them more easily? Um... yeah, that'll happen...

That was followed by the decent analysis by Nishant Kaushik from Oracle about how social software sites are popular only because there is no internet-wide identity management scheme. I don't believe Nishant went far enough... a distributed, user-centric identity scheme would destroy almost every "walled garden" social software application on the web... Which is exactly the reason why very few people would choose to implement it.

The true next generation of social software will embrace and share identity -- like does -- and thus enable users to replace old-school identity silos in a flash.

Think about it: if we had an internet-wide user-centric identity scheme (like SXIP/OpenID), then I could identify -- in one single place -- what my likes, dislikes, bookmarks, and contacts are. I wouldn't have to copy all my LinkedIn contacts into Facebook. I wouldn't have to tell every news site that I dig technology. I wouldn't need a hundred passwords... They would all just work together, and know me!

Or, more specifically, I would log in with my internet-wide username and password... then the apps would ask me if they could sync with my identity management server for contacts and interests. Then I would explicitly choose yes or no depending on what site I'm connecting to.

Personally, if offered some kind of SXIP/OpenID implementation, I'd use it in a flash... I already have all my bookmarks there, along with my tags. I've always said, my tag cloud says more about me than any customer survey I've ever taken... is truly sitting on a goldmine that could destroy StumbleUpon, if they only knew how.

The Wired guys noted that it takes very little effort to duplicate 90% of what Facebook does... Nishant correctly states that the remaining 10% is totally solved with internet-wide identity management. I'm not sure if he's backing the right horse -- IGF style governance documents??? WTF??? -- but he has the right idea.

Frankly, Mark Zuckerberg, the wannabe-Steve-Jobs-CEO-of-Facebook, is really close to screwing the pooch on this one. He's sitting on a house of cards, not a goldmine. If Yahoo offers $1 billion for your company, frigging take it! Hell, if they offer $50 million its a bargain... unless you're certain you can con somebody else into offering more. Then jerk around the jerks for as much as you can get.

But don't take too long: once Ning dumps their NingID in favor of OpenID, Facebook's days are numbered. And frankly, if Ning was motivated, that could happen in under a month.

Sorry to bring this to your attention, capitan fleece, but somebody had to.

Further reading:



I agree, Ning is a very user friendly app and easy on the eyes.

forgot to mention...

it might be worthwhile for Yahoo to purchase Facebook, blend it with, and add an OpenID interface to it. That would be the ultimate way to charge forward into the market, and simultaneously take out a lot of the competition (like MySpace and LinkedIn).

Of course, it only makes sense if Facebook sells for the value of their "contacts" database, and little more...

SXIP and OpenID?

How can you mention Sxip and OpenID in the same breath? Part of the challenge with an "internet-wide user-centric identity scheme" is aligning on one...

Mixing Sxip and OpenID only adds to the confusion...

I hope not...

I usually say SXIP/OpenID to make it clear that SXIP is a part of the OpenID 2.0 spec for SXIP fans. Sorry if I confused the issue. More info:

Bex, you nailed it.

A while back, I described 3 potential resolutions to this problem. Like yourself, I prefer the open/decentralized version (#1)

I find the Twittergram work very interesting as an end-around to OpenID adoption - authenticate with credentials you already have (could be OpenID, TwitterID, or GoogleID). A use-what-you-have approach vs. creating a new silo.

Overall, I think we'll get some progress on identity management when pageviews stop selling.

that leaves a question...

In your article here:

Should you have a twittergram for every app -- Flickr, LinkedIn, BaseCamp -- or should every app implement Twitter?

I think technologies like Jabber can be easily tweaked to give Twitter support to any system... not to mention other a plethora of other chat-like interfaces... as others have noted:

Twitter, other than as a brand, is also a house of cards that could collapse with decent identity management...

RE: that leaves a question

I'm saying any app with an API including name/pass could be used for authentication for another app. Sure, Jabber could be used, same with OpenID, or Twitter, any other app, or all of the above. Authentication could be customer-centric rather than app-centric.


Very true... Although I'm now wrestling with the question, "who owns a relationship?"

Not finding any easy answers...

RE: gotcha

dare we say the customer?


Regarding unified ID and social networks, perhaps a combination of FOAF - cf. the FOAF Online Account Description Generator -, OpenID, the foaf:openid property and XMPP authentication for OpenID could help - though I haven't figured the whole picture.

facebook $50 million/ now $100 billion IPO

I remembered your comment about facebook long time ago and I couldn't help coming back and comment about it :-) .
Who would have guessed it would be $100 billion company.

Recent comments