Mashup Wrapup

I'm writing this from the cavernous downstairs at the MIT Media Lab in Cambridge, Massachusetts, where we are nearing the end of the third and last day of the Identity Mashup conference put on by Harvard's Berkman Center for Internet and Society. The first two days' sessions were held at Harvard Law School (out of which Berkman grew). The third was held here.

Forward movement has been tectonic. Just like it was at the last Berkman identity event, earlier this year. And at the Internet Identity Workshops (IIWs) held in the Bay Area last October and this May.

Things have not always moved so fast.

That's what I expected back in 2002, when Andre Durand, one of the fathers of the Jabber/XMPP movement, told me that identity was personal -- and that personal control was a huge piece missing from all the identities (mostly in the form of cards) that we receive from credit card companies and other outside parties. Andre painted a vision of a networked marketplace in which individuals held the most important cards -- namely their own -- which they would use to interact more efficiently with businesses, and to establish a new balance of market power that favored customers for the first time since the dawn of the Industrial Age.

I got on board right away; but the bandwagon turned out to be more like a wheelbarrow. For the next two years I spoke often and earnestly on the inevitability of a future business world in which "user-centric identity" would change everything. While almost nothing actually happened.

Then, at the Fall 2004 Digital ID World, I pushed a number of like-minded folks in a converging direction. Could be there was a gravitational attraction in any case, but I like to think that I helped move things along a bit.

Especially when Steve Gillmor and I convened The Identity Gang -- a group of user-centric identity figures -- for the Gillmor Gang podcast on December 31, 2004. A conversational snowball started rolling, and it has since increased enormously in size and importance.

But not in formality. The Identity Gang is open to anybody with anything to contribute to the conversation - or to the assortment of code bases that comprise a converging array of solutions to different, related and overlapping identity problems.

All those code bases share a single central concern: the individual. Known variously as "individual-centered identity", "identity 2.o" and "independent identity" - among many other labels - Identity Gang members' common perspective is outward toward markets, and toward relationships of all kinds, from the perspective of the independent individual.

This is new. In the past, most -- if not all -- identity solutions came from business, government, standards bodies and other institutional sources, and shared those sources' top-down perspective toward individuals. We see this perspective behind every plastic card in our wallets that carries our name. In every case, our identity is organizationally peripheral (if not also subordinate) to the organization that identifies us -- whether that organization is the DMV, the credit card company, the city library, or the local Y. Even some identity standards development efforts, such as the Liberty Alliance and the WS-Federation, are projects launched by and for large companies. These efforts were highly respectful of individuals from the beginning, and necessarily cover large areas of the growing identity solution map; but they do not come from individuals.

All the efforts developed by Identity Gang members -- ranging in scale from Brad Fitzpatrick's OpenID (created for Live Journal blogs) to Microsoft's InfoCard (created for Vista, but due to be released for XP) were zero-based around individuals and their sovereign identity needs in the networked marketplace. (Kim and Microsoft's remarkably open-oriented efforts were visited at length in Independent Identity, in last September's issue of Linux Journal.)

From the beginning, open and encompassing concepts for identity solution interoperability and convergence have also been no less important than the code development those concepts have guided. These include Kim Cameron's Laws of Identity and Identity Metasystem construct, Dick Hardt's Identity 2.o, Drummond Reed's dataweb, my own Intention Economy -- and others I insult by forgetting right now.

From its beginnings, the Identity Gang and its friends have been doing their best to keep their separate development projects as open and mash-able as possible. These efforts have been encouraging, to say the least. In the first Internet Identity Workshop (IIW), held this past October in Berkeley, OpenID, LID and XRI/XDI (i-names) mashed up Yadis (a simple URL-based "Identity and accountability foundation for Web 2.o.") from parts of their own separate development efforts. Still, as Johannes Ernst of LID, "we could see we were stepping on each other's toes". And a concerted effort was made to stop doing that.

The second IIW, held last month in Mountain View, saw the first open source implementation of Microsoft's InfoCard, by the University of North Carolina at Charlotte. It also saw Identity Gang members from the IETF, Red Hat, Verisign, Novell, SXIP, Tucows, LID, Linux Journal and other organizations meet with Kim Cameron and Mike Jones of Microsoft to talk about working compatibly with InfoCard, and launching both commercial and open source InfoCard-compatble development projects. Now there are weekly OSIS (Open Source Identity Selector) conference calls, each intended to keep independent developments both open and mash-able.

And that was just one among many mash-worthy efforts that snowballed at the Workshop.

The Berkman Center became involved with the Identity Gang at Esther Dyson's PC Forum in March 2005, when Senior Fellow John Clippinger joined the gang's meetings there, and volunteered the center as a virtual "clubhouse". Not long after that, Paul Trevithick - John's colleague at Social Physics - launched the Identity Gang wiki.

The Identity Mashup is the second identity event hosted by the Berkman Center. The first was a workshop on user-centric identity and commerce held in February. As with every other Identity Gang gathering, there was tremendous progress there. For example, this was where Paul Trevithick introduced the Higgins framework for integrating identity-related information across heterogeneous systems.

I don't know where to begin listing all the tectonic movements that happened here in the last three days. And I don't have time. My plane leaves shortly, and I need to get this out to the newsmill for mailing.

So I'll point to a posting by Johnannes Ernst, about OSIS: http://netmesh.info/jernst/NetMesh/announcing-osis-zdnet.html.

The next IIW will be in November. I expect to see more of ya'll there.

Meanwhile, I'll have a lot more to report about the whole workshop when I get back home on Monday. Watch the Linux Journal website for that.