An abstract from Esther Dyson’s most recent report that takes a look at ICANN and the Domain Name System as a means to assess future benefits and challenges of all other emerging “active registries” in the digital world.

As the world grows more connected and more complicated, we all need ways of defining, identifying and keeping track of things and cross-referencing them with their owners. The simplest way to do that is with registries—everything from the Domesday Book, a medieval registry of land, property and people; to current-day auto registries on the one hand and the worldwide Domain Name System on the other. In today’s computer world, registries are gaining visibility, even though in some form they have been around for a long time (cf. Novell’s Directory, Microsoft’s Active Directory, and even IBM’s old mainframe catalogue).

But now, companies and organizations have to keep track of ever more things and people, not just inside their walls but across extended organizational boundaries. Call this new wrinkle an “external registry”. Finally, they may want to interact with things and people, rather than just look them up, via an “active registry”. The DNS is the best-known example of an active, external registry. Its technical function, automatically translating the logical address of a URL into a (semi-) physical IP address, supports the Net’s rapid growth while the mnemonic character of URLs facilitates the e-mail and Web connections among businesses and people that have driven that stunning growth.

Several new online registries are positioning themselves to play analogous roles in the growth of specific new capabilities on the Net. The implementation of Electronic Numbering (ENUM), a new standard for turning phone numbers into domain names, could vastly expand the use of new ways to connect people such as instant messaging, voice over IP (VoIP) or video conferencing services, and promises to enable a more real-time Internet. Other registries such as the Handle System¹ and AutoID’s Object Name System (ONS)² will help enable new product and service offerings such as content purchases or product-tracking updates through indirect addressing that references data and functions rather than servers and locations. The kind of growth the DNS supported for the Internet as a whole may be fostered in other marketplaces through registries that enable interconnection among people and applications with functions specific to particular industries.

But these capabilities raise hoary questions about information ownership, access and modification control, version management…and the neutrality of the registries in the markets they serve. In short, registries can’t simply be inserted into the infrastructure and work (even though the software itself is simple enough). Registry systems need corresponding infrastructure: standards, federated identity management, business “protocols” and policies, and sometimes even political buy-in or active regulation. Lack of appropriate governance can foment pointless and costly disputes that benefit almost no one other than lawyers, along with meretricious business models based on artificial economics.

Yet most of the governance questions about active, external registries are still unresolved. The “experiment” of the Internet Corporation for Assigned Names and Numbers’ (ICANN) oversight of the DNS is still not conclusive, but ENUM and to some extent ONS rely on the DNS. Meanwhile, the Handle System and its “digital object architecture,” developed by Bob Kahn (who also has much of the Internet’s architecture to his credit), offers both better technology and “lessons-learned” governance…but unfortunately it lacks the visibility of the DNS, which is an order of magnitude larger in the number of things registered (approaching 200 million vs. less than one tenth of that).

—-
1. One of CNRI’s leading initiatives is the “digital object architecture” and its implementation through the system of unique, persistent identifiers it calls handles. The Handle System is somewhat like the DNS, in that it provides a registry to find things online. But it has important differences, says Kahn: “While the DNS registers machines [with domain names], the Handle System registers digital objects. The DNS served its purpose well, but now we have something better and different, with persistence and location independence. The digital object architecture is a reconceptualization of the Internet to deal with specific information objects, instead of just flows of packets between servers. It takes transport out of the picture [although you still need it underneath].”
Source: Release 1.0, Online Registries: The DNS and Beyond, page 22

2. One of the major new registry efforts, the Object Name System (ONS) for manufactured consumer goods, is moving forward ?- with implementation details that illustrate the technical and commercial realities of such initiatives. What started as a set of research projects is becoming a reality, even as the parent organization, the Auto- ID Center at MIT, undergoes its own transformation.
Source: Release 1.0, Online Registries: The DNS and Beyond, page 12