Thursday, March 06, 2003

If you haven't read "True Names" yet . . . People have been talking about Vernor Vinge’s novella “True Names” for quite a while. It’s particularly relevant to the digital identity and identity management discussion because Vinge, well before folks like William Gibson, not only anticipated cyberspace, but the potential benefits and dangers of digital identity in cyberspace. That Vinge wrote it in 1981 makes it all the more remarkable.   I bring this up only because I recently read True Names: And the Opening of the Cyberspace Frontier. This book couples the novella with essays written by a wide variety of interesting folks, many from technology circles, who comment on how Vinge’s story relates to the evolution of the Internet to date, and its future. It’s all the more interesting because many of the essays pre-date the bursting of the technology market bubble. (The book was published in 1999.) To see how things have changed and how some folks were still thinking clearly back then (and some folks weren’t) is quite interesting. In particular, the essays by Timothy Mays ("True Nyms and Crypto Anarchy") and Alex Wexelblat ("How is the NII Like a Prison?") will give anyone contemplating the implications of digital identity a lot to chew on. The many references to Bruce Sterling’s “four horseman of the modern apocalypse” (terrorists, pornographers, drug dealers, and the Mafia) are eerily resonant today.   I’ll have to admit that I don’t like the way the book was organized; it puts Vinge’s novella at the end. After first attempting to read it in the order the editor intended, I cheated, skipping to the back and reading “True Names” and then reading the commentary. Made the essays—and the references to the novella they contain—much more meaningful.   Still, it’s highly recommended reading. 12:08:25 PM

US Federal Government Supports Liberty The Liberty Alliance announced that the General Services Administration (GSA) and the US Department of Defense (DoD) have joined the organization. When you combine these obviously important government agencies with the large number of corporations that have joined the Alliance, it’s clear that Liberty has critical mass.   The GSA’s and DoD’s decision to join the Alliance has a couple of important implications. First, the government’s involvement can help speed the development of the legal and technical infrastructure necessary to enable federation, increasing the legitimacy of whole effort. The US federal government has been working on many identity fronts, whether it’s the Federal PKI project, a variety of smart card deployment projects, as well as the more recent eAuthentication initiative. (Phil Becker makes these same points on the Digital ID World blog.) Things will move forward if the government, building on the Liberty spec, establishes a means by which companies doing business with the government can federate identity with the government. As these government agencies establish legal and technical frameworks for federation on which they rely, many businesses will feel more comfortable in moving ahead, relying on some of those same legal and technical frameworks.   Second, the federal government’s involvement in Liberty could, at least in theory, lead to a stronger push toward international federation agreements. As Craig Mundie, Microsoft’s CTO, pointed out in his speech at Digital ID World last year, we haven’t even really begun to define how governments, which typically feel they have sovereign rights to manage identity within their borders, will federate identity internationally. Mundie has been working with folks like Lawrence Lessig and Esther Dyson under the auspices of the Center for Strategic and Internal Studies to generate recommendations to the world’s governments on how to federate digital identity in cyberspace. But governments have to get their toe in the federation waters before things move, so maybe this will start to catalyze those efforts.   Liberty’s building momentum is also good news for the Security Assertion Markup Language (SAML), which Liberty based its spec on. SAML continues to show strong momentum, providing a starting point for federated identity. Microsoft and IBM are promising a WS-Federation specification, and Microsoft continues to push XrML as a more robust assertion model. But SAML, precisely because it was narrowly scoped, was completed quickly and vendors are implementing it. If the government moves to implement the Liberty spec in its own identity efforts, it will further strengthen the case for SAML.   We’ll be releasing a report on Liberty and where the organization is headed next week. 10:15:03 AM